This action might not be possible to undo. Are you sure you want to continue?
A COMPILATION OF RECENT RESEARCH RELATED TO THE APOLLO MISSION By Langley Research Center Staff Langley Research Center Langley Station, Hampton, Va.
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
CONTENTS LUNAR ORBITAL ENTRY AND DESCENT 1. LUNAR-ORBIT LANDING SITES AND STAY TIMES . . . . . . . . . . . . . . . . By Robert H. Tolson 1
2. RETURN-TO- EARTH CONSIDERATIONS AND STAY-TIME CONSTRAINTS FOR THE
LUNAR MISSION . . . . . . . . . . . . . . . . . . . . . . . . . . . By John P. Gapcynski By Richard Reid LUNAR LANDING
3. A SIMPLE GUIDANCE TECHNIQUE FOR LUNAR LETDOWN . . . . . . . . . . . . .
4. VISUAL SIMULATION STUDY OF LUNAR HOVERING, TRANSLATION, AND
TOUCHDOWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . By Maxwell W. Goode By Ulysse J. Blanchard
27 37 47
5. LANDING CHARACTERISTICS OF A MODEL OF THE LUNAR EXCURSION MODULE . . . . 6. LUNAR TOUCHDOWN OPERATIONS AS AFFECTED BY ABORT CONSIDERATIONS .
By Gary P. Beasley
7. DEVELOPMENT OF A SIMULATOR FOR STUDIES OF SELF-LOCOMOTION OF MAN
UNDER REDUCED GRAVITY . . . . . . . . . . . . . . . . . . . . . . . By Donald E. Hewes and Amos A. Spady, Jr. TAKE-OFF, RENDEZVOUS, AND DOCKING
8. SOME ASPECTS OF MAN'S VISUAL CAPABILITIES IN SPACE . . . . . . . . . . .
By Jack E. Pennington
9. FIXED-BASE GEMINI-AGENA DOCKING SIMULATION . . . . . . . . . . . . . . .
By Byron M. Jaquet and Donald R. Riley 10. THREE-DEGREE-OF-FREEDOM FIXED-BASE SIMULATION OF PILOT-CONTROLLED LUNAR TRAJECTORIES FROM LIFT-OFF TO RENDEZVOUS . . . . . . . . . . . By Charles P. Llewellyn 11. MANUAL CONTROL OF A LUNAR LAUNCH . . . . . . . . . . . . . . . . . . . . By Lindsay J. Lina
79 85 93
12. AN OPTICAL DEVICE FOR OBTAINING ATTITUDE AND ALTITUDE . . . . . . . . .
By Alfred J. Meintel, Jr. i
13. ANGULAR SEPARATION OF APOLLO AND LUNAR EXCURSION MODULE AT LANDER TOUCHDOWN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . By James L. Williams and L. Keith Barker 14. SIMPLE ABORT SCHEME FOR SYNCHRONOUS ORBITS . . . . . . . . . . . . . . By G. Kimball Miller, Jr., and L. Keith Barker 15. ABORT CONSIDERATIONS FOR THE LUNAR EXCURSION MODULE . . . . . . . . . By David B. Middleton .
101 107 115 123 131
16. PENETROMETER TECHNIQUES FOR LUNAR SURFACE EVALUATION . . . . . . . . . By John Locke McCarty, Alfred G. Beswick, and George W. Brooks 17. PROPOSED SURVEYOR LANDING EXPERIMENT By Sidney A. Batterson . . . . . . . . . . . . . . . . .
18. RADIO-FREQUENCY SIGNAL ATTENUATION BY PLASMAS OF ROCKET EXHAUST GASES ............................... By Duncan E. McIver, Jr. 19. DYNAMIC PENETRATION AND EROSION OF DUST-LIKE MATERIALS IN A VACUUM ENVIRONMENT . . . . . . . . . . . . . . . . . . . . . . . . By Leonard V. Clark and Norman S. Land 20. VISIBILITY AND DUST EROSION DURING THE LUNAR LANDING By Leonard Roberts LUNAR RESEARCH FACILITIES . . . . . . . . .
21. LOLA, THE LUNAR ORBIT AND LANDING-APPROACH SIMULATOR . . . . . . . . . 171 By William T. Suit and Ralph W. Stone, Jr. 22. DESCRIPTION OF A LUNAR-LANDING RESEARCH FACILITY . . . . . . . . . . . By Thomas C. O'Bryan 23. RENDEZVOUS DOCKING SIMULATOR . . . . . . . . . . . . . . . . . . . . . By Howard G. Hatch, Jr. 24. NEW DYNAMIC RESEARCH FACILITIES SUITABLE FOR SUPPORT OF APOLLO-®LUNAR-EXCURSION-MODULE MISSION . . . . . . . . . . . . . . . . . . By D. William Conner 179 187
25. LUNAR-GRAVITY SIMULATOR FOR FULL-SCALE IMPACT TESTING . . . . . . . . . 199 By Robert W. Herr
LUNAR ORBITAL ENTRY AND DESCENT
1. LUNAR -ORBIT LANDING SITES AND STAY TIMES
By Robert H. Tolson
The determination of possible lunar-landing sites and the corresponding exploration times are discussed for missions in which the lunar-orbit-rendezvous technique is utilized. First, the geometrical properties of lunar orbits, which can be established from typical earth-moon transfer trajectories, are reviewed from the standpoint of the lunar-landing-site problem. Use of a landing procedure which allows mission abort during the entire exploration period indicates that a large region of the moon can be explored for a few days or less. Finally, a brief discussion is given of the possibility of increasing the landing-site capability by varying the parameters in the analysis.
In the initial Apollo missions the location of the landing site on the lunar surface will probably not be one of the primary parameters determining the overall mission profile. However, in later missions, landings in particular regions of the moon's surface may be desirable for scientific or other reasons; therefore, it is of interest to determine the lunar sites which are accessible for the Apollo mission and, in particular, to determine the possible lunar-landing sites when the lunar-orbit-rendezvous technique is utilized. The establishment of a lunar orbit is, of course, an integral part of the lunar-orbit-rendezvous technique and, as illustrated in figure 1, the types of lunar orbits that can be established play an important role in determining the possible lunar-landing sites. It is assumed that the mission is initiated with a launch from Cape Canaveral and that after an appropriate coasting phase the injection into the transfer trajectory occurs at an altitude at 480 km with an injection angle of O o . Typical values of these two parameters are chosen, because it simplifies the analysis and because the results do not sensibly depend on their particular values. At the proper point in the vicinity of the moon an impulse is applied to establish the lunar orbit. The orientation of the resulting orbital plane is specified by the inclination of the orbital plane to the earth-moon plane i and by the longitude of the ascending node of the orbit measured from the earth-moon line S2. Although the lunar-landing procedure for the lunar excursion module has not been specified in detail, it is expected that the lunar excursion module will always land nearly in the orbital plane of the command module. Hence, for a given lunar orbit, the possible landing sites constitute a narrow band around the moon directly below the orbit of the command module. The width of this band will depend on the propulsion capability of the lunar excursion module to make orbital plane changes during its descent to the lunar surface. Thus, the location of
possible lunar-landing sites is directly related to the geometrical characteristics of lunar orbits which can be established efficiently and which at the same time are consistent with overall mission requirements.
GEOMETRICAL PROPERTIES OF LUNAR ORBITS
In order to investigate the possible landing sites it is advantageous first to determine the allowable range of i and 2 consistent with typical constraints on the transfer trajectory. For this purpose, it would be convenient to have explicit relationships between the lunar orbital characteristic and the injection conditions. Unfortunately, no exact expressions are known and some approximations are required in order to obtain general analytical information. For this study the earth and moon are assumed to be spherical masses and to move in circular orbits at their mean distance; in addition, a "patched" conic technique (ref. 1) is utilized. This technique is based on the assumption that there is an imaginary sphere centered at the moon with a radius of about 58,000 km. When the vehicle is inside the "sphere of influence," the earth's gravitational effects on the vehicle are neglected; and when the vehicle is outside of the sphere, the moon's gravitational forces are neglected. With these assumptions it is possible to obtain explicit relationships between the lunar orbital characteristics of interest here and the transfer trajectory injection conditions. Details of this analysis have been reported in reference 2, but a few of the results are mentioned herein because they have a direct bearing on the lunar-landing-site problem. In figure 2 the sphere of influence is depicted with the moon at its center. The latitude q measured from the earth-moon plane and the longitude measured from the earth-moon line are used to specify the location of any point on the sphere of influence. The dashed lines represent the projection onto the sphere of influence of two typical lunar orbits and as before the orientation of these planes will be specified by the inclination and nodal position. A nominal transfer trajectory is defined as one with a specified injection energy and a specified inclination to the earth-moon plane I. A number of free parameters at the injection point are still to be considered and if these free parameters are properly varied, it is in general possible to obtain a wide variety of lunar orbits, namely, lunar orbits with a wide range of values of i and Q. However, the analysis of reference 2 showed that the orbital planes of all lunar orbits established from a given nominal trajectory pass through a common point on the sphere of influence, that is, all the orbital planes have a common line of intersection. This result provides a relationship between the inclination, nodal position, and the latitude and longitude of the common point of intersection of the orbital planes, as follows:
tan i = tan sin(Q - )
Thus, if the energy and inclination of the transfer trajectory are specified, the values of ^ and q are fixed; consequently, the inclination and nodal position are not independent parameters and once one is chosen the other is determined by
this relation. (The importance of this result on the location of possible lunarlanding sites is discussed subsequently.) This common point of intersection is often called the entry point because this is the point on the sphere of influence through which the vehicle must pass if it is to impact normal to the moon's surface; that is, the extended trajectory would pass through the center of the moon. The utility of the patched conic technique is that the locations of these so-called "entry points" are defined by a set of algebraic equations that can be solved to give the latitude and longitude as a function of the transfer trajectory characteristics. A typical solution of these equations is shown in figure 3. The lower part of the figure gives the locus of entry points for four injection velocity ratios (1.0 corresponding to parabolic velocity) and for a range of transfer trajectory inclinations to the earth-moon plane. At the lower velocities the flight time from the earth to the moon is excessively long and to inject at the higher velocities generally results in a payload penalty, so that it is actually only the middle third of this region which is acceptable for manned missions. In other words, all of the orbital planes must pass through a rather small region on the sphere of influence. Briefly, what has been shown about the establishment of lunar orbits is that if a nominal earth-moon transfer trajectory is chosen by specifying the injection energy and inclination, then the orbital planes of all lunar orbits established from this nominal trajectory pass through a common point or entry point on the sphere of influence. This result gives a relation between the geometrical properties of the resulting lunar orbits in terms of the latitude and longitude of the entry points. Finally, for manned missions the allowable variation in ^ and n as they appear in the relation cannot be changed appreciably by varying the injection conditions of the nominal trajectory.
LUNAR-LANDING SITES AND EXPLORATION TIMES
With the foregoing relationships established between the lunar orbital characteristics and the earth injection parameters, the allowable stay times on the lunar surface as defined by lunar-orbit-rendezvous requirements may be considered. Some of the parameters in such an analysis are illustrated in figure 4. Suppose a landing at some selected latitude on the lunar surface is desired. One approach to the problem is to establish a lunar orbit with an inclination i greater than the required latitude by the offset, 8. When the excursion module starts its descent to the lunar surface, a small out-of-plane impulse is applied so that the path is along the dashed line and the landing site is 8 degrees out of the orbital plane of the command module. Because of the moon's rotation on its axis and its orbital motion about the earth, there will be a relative motion between the landing site and the orbital plane. In figure 4 the landing site will appear to move to the right along a parallel of latitude at about 13.2 0 per day. (Note that no perturbations of the command module from pure Keplerian motion are considered here.) After a certain angular travel 08, the landing site will again have an offset of 8 degrees and the offset is increasing so that the "return to orbit" phase of the mission would be initiated. The exploration period (stay time) on the lunar surface r is given by
P = GO
where D8= cos i sin ^ - sin 8 8 cos — cos 'T sin i 2
At any time during the exploration period the lunar excursion module can return to the command module with not more than an offset of S degrees required. Now, the significance of the previously derived relationship between orbital inclination and nodal position is realized. For if 8 is specified by the propulsion capabilities of the excursion module, then the latitude of the landing site determines the required orbital inclination by the relation that Inclination = Latitude + Offset. However, as was shown previously for a specified transfer trajectory, the nodal position is determined once the orbital inclination is chosen. Hence, if this type of landing maneuver is utilized, the landing-site location on any parallel of latitude is uniquely determined by the transfer trajectory characteristics and the offset. Thus, if a maximum stay time is to be obtained for fixed offset, a rather strong restriction on the possible landing sites must be accepted. The landing-site capability can be increased at the expense of decreasing the allowable stay time on the lunar surface by a slight modification of the landing procedure illustrated in figure 4. Instead of landing at the most extreme point to the left, the landing can be made at some point between the two extremes along the desired parallel of latitude. Again, the exploration time is limited to the time it takes for the landing site to move to the right-hand limit point. With this variation of the original landing procedure there is a range of possible landing sites along each parallel of latitude, and each landing site has a certain stay time associated with it. The location of the possible landing sites on the lunar surface and the stay time at each site for any nominal earth-moon transfer trajectory can be calculated using equations (1), (2), and (3), together with the results given in figure 3. Figure 5 gives the landing-site stay-time restriction for a median-energy, lowinclination transfer trajectory. The figure shows the face of the moon as seen from the earth. The boundaries are the locus of points on the lunar surface at which landings can be made and at which the lunar excursion module can stay for the specified time without requiring landing and take-off offsets of more than 50. The hatched line is the line of maximum stay time corresponding to the landing procedure illustrated in figure 4. In the equatorial and polar regions the excursion module can stay on the surface indefinitely and have the capability of returning to the command module within the limiting amount of offset; for example, if it is possible to establish an equatorial lunar orbit, then the excursion module can land within 5 0 of the lunar equator, and the rotation of the moon on its axis will not affect the relative position of the excursion module with respect to the orbital plane of the command module.
It is seen that a considerable portion of the lunar surface is subject to manned exploration for periods of a few days or less. However, using this landing procedure with the specified offset and the specified nominal transfer trajectory will not allow landings in the midlatitudes on the western limb of the moon. Before the landing capability can be extended into this region, either the offset, the transfer trajectory characteristic, or the landing procedure must be changed. Increasing the offset will not be considered since this parameter is directly related to the propulsion capabilities of the lunar excursion module, and a sizable increase above the value used here is not expected. In addition, changing the transfer trajectory characteristics will not yield a large increase in the possible landing area because the location of the landing sites on the lunar surface is essentially determined by equation (1), and the results given in figure 3 indicate that for manned missions, the location of the entry points is limited to a small region on the sphere of influence. Thus, the values of ^ and T1 that appear in equation (1) can only take on variations of about ±80 from the mean values used to obtain the results of figure 5, and consequently changing the transfer trajectory characteristics would result in a displacement of the boundaries shown in figure 5 by only t80 in both latitude and longitude. Finally, there are a number of ways in which the landing procedure can be altered so that the excursion module can land at the midlatitudes on the western limb of the moon. First, the landing procedure considered heretofore was designed so that the excursion module could return to the command module at any time during the exploration period without requiring an offset of more than 8. As mission experience is gained, such a requirement may be relaxed in order to increase the landing-site capabilities inasmuch as without this requirement, it is possible to land at any point on the lunar surface. The motion of the moon may take the excursion module a large distance out of the orbital plane of the command module; however, twice during the lunar month the orbital plane will pass over the landing site and the return flight can be initiated. The exploration time at each point on the surface is fixed by simple geometrical constraints and again by the results illustrated in figure 3 and equation (1). Consider again landing procedures which allow return during the entire exploration period. Note that for a specified landing site on the western limb it is possible to establish a lunar orbit such that after landing, the excursion module will have a displacement of 8 to the west of the orbital plane of the command module; however, the inclination will be greater than the sum of the offset and the latitude. The moon's rotation will again cause the landing site to move eastward relative to the orbital plane; and after a short angular travel, the landing site will be east of the orbital plane with an offset S and the offset will be increasing. If the excursion module returns to the command module at this time, the exploration period in days is given by;
sin- sin ^ cos i + sin 8 _sin-1 sin A cos i - sin 8 cos T sin i cos A sin i
This equation gives an exploration time of at least 3.6 hours per degree of offset. Therefore, with this landing procedure and a 50 offset, every point on the western limb can be explored for at least 18 hours. Although it has not been stated explicitly, it has been assumed that the lunar excursion module initiates its landing maneuver soon after the lunar orbit is established. Suppose, however, that the excursion module does not land immediately. Because of the moon's motions, the nodal position of the orbital plane of the lunar excursion module will appear to move westward along the lunar equator. If after some specified waiting time in orbit the landing procedure illustrated in figure 4 is initiated, the nodal position will have precessed westward relative to the lunar surface from its original position through an angle equal to the product of the moon's rotational rate and the waiting time in orbit. Therefore, for a specified waiting period, the landing sites and the corresponding exploration times could be obtained from figure 5 by simply rotating the areas in the figure westward through the aforementioned angle. To increase the area of possible landing sites to the entire western limb would require waiting times in orbit of about 6 days.
For lunar missions utilizing the lunar-orbit-rendezvous technique, the determination of the possible lunar-landing sites and the corresponding exploration times is seen to depend first on the geometrical properties of lunar orbits which can be established from typical earth-moon transfer trajectories and secondly on the particular lunar-landing procedure utilized. For a landing procedure which affords mission abort during the entire exploration period, a large region of the moon can be explored for periods of a few days or less. A sizable increase in the possible landing-site area on the lunar surface cannot be obtained by changing the characteristics of the transfer trajectory; however, there are a number of variations in the basic landing procedure which allow landings to be made at any point on the lunar surface.
REFERENCES 1. Plummer, H. C.: An Introductory Treatise on Dynamical Astronomy. Dover Publ., Inc., 1960. 2. Tolson, Robert H.: Geometrical Characteristics of Lunar Orbits Established From Earth-Moon Trajectories. NASA TN D -1780, 1963.
o " " » ~ ~ ~^
w ° ° ~ °° ^ ^ * ^ ~^^ ~
^ ^ ^^*
r °° ° ^~ ° ~ ^ = ° ° ^
^ ^ °^ ^ ° ° ^ ^ ^ «^^ v^
ESTABLISHMENT OF LUNAR ORBIT MOON ' LUNAR ^~~~~ MOTION`~~_~~ ^LANDING SITES TRANSFER TRAJECTORY
t' - l
i=LUNAR-ORBIT INCLINATION TO THE EARTH-MOON PLANE D=4SCENDING NODAL POSITION OF LUNAR ORBIT I , TRANSFER-TR4JECTORY INCLINATION TO THE EARTH-MOON PLANE Figure I
GEOMETRICAL CONSTRAINTS ON LUNAR ORBITS N INJECTION ENERGY ' —>c^^'^\ TRAJECTORY INCLI NATI ON) ` ' ~~~— OF `X -`^~JNFLUENCE
^ ^ .
LOCATION OF ENTRY POINTS ON LUNAR SPHERE OF INFLUENCE
LUNAR SPHERE OF INFLUENCE
(RADIUS = 58,000 K M)
MOTIONS TO EARTH LATITUDE, TRANSFER-TRAJECTORY 77, DEG INCLINATION, DEG 10 90 0 -10 60 .992 30 0 ,994
INJECT 10N-VELOCITY RATIO
_ 90 0 LONGITUDE, ^, DEG O
EXPLORATION TIME FROM RENDEZVOUS CONSIDERATIONS EXPLORATION TIME=
TAN i = SIN M-C)
PARALLEL OF LATITUDE
LANDING SITE-STAY TIME RESTRICTIONS FROM RENDEZVOUS CONSIDERATIONS, 8=50 N
Page intentionally left blank
2. RETURN-TO-EARTH CONSIDERATIONS AND STAY-TIME CONSTRAINTS FOR THE LUNAR MISSION By John P. Gapcynski
This paper presents a discussion of the restrictions which are imposed on the orientation of a lunar orbit so as to achieve a satisfactory return flight to a desired touchdown point on the earth's surface. In addition, the relationship between the orientation of lunar orbits established from earth-moon transfer trajectories and the orientation required for the return flight is discussed.
The return flight from the moon is one of the more important phases of the overall manned lunar mission, inasmuch as the restrictions imposed by the return requirements may well dictate the entire mission profile and timetable. The problem involved in defining these return requirements is depicted schematically in figure 1. It is assumed that a vehicle is in some arbitrary orbit about the moon as shown. The problem is to determine the injection conditions which will place this vehicle in a trajectory for returning to the earth, reentering the atmosphere at some point R, and landing at some touchdown point T after a specified angular travel through the atmosphere (defined here as the reentry range). If it is assumed that the vehicle is in a circular orbit about the moon at some arbitrary altitude, then the lunar orbit may be defined by the two parameters shown in figure 1 - that is, the inclination of this orbit to the earth-moon plane i and the ascending nodal position of this orbit P, measured with respect to the earth-moon line. For the return problem, it is necessary to define the permissible range of these two orbital parameters - the inclination i, and the ascending nodal position S2 - which results in a satisfactory earth-return trajectory for reasonable values of the vehicle injection velocity and total mission flight time.
EARTH REENTRY CONSTRAINTS
In solving the return problem, it is necessary to work back from the earth to the moon by first establishing the requirements for earth reentry and then determining the lunar orbital characteristics which meet these requirements. Two reentry conditions must be satisfied by each return trajectory: first, the vehicle must reenter within a specified corridor, and second, the vehicle must achieve a desired touchdown point on the earth's surface. It is the second of these
---- -------constraints which causes the most trouble because the specification of a desired touchdown point introduces the factor of time into the problem. In satisfying the requirement of a specific touchdown longitude, the -parameters of importance are the time selected for injection and the return-trajectory flight time. In satisfying the requirement of a specific touchdown latitude, the -parameter of importance is the inclination of the return-trajectory -plane to the earth-moon plane. This parameter is designated by I in figure 1. Note that this parameter I is the inclination of the return trajectory to the earth-moon plane and should not be confused with the inclination of the lunar orbit to the earth-moon plane designated in figure 1 by i. It is well to note here that it is possible to obtain one specific value of the return-trajectory inclination from lunar orbits which have a wide range of values of orbital inclination to the earth-moon plane. To illustrate the importance of the return-trajectory inclination, consider the data shown in figure 2. In this figure is shown the variation of the required return-trajectory inclination I with the angular position of the moon in its orbit to achieve a touchdown latitude of 300. The abscissa is essentially a time scale; that is ., during the lunar month, the angular position of the moon will change through 3600 in the direction shown at the rate of 13 .20 per day. Maximum positive, or northerly, declination of the moon on this scale occurs at an angular position of 900., and maximum negative, or southerly, declination occurs at an angular position of 2700. Results are presented for representative values of the reentry range of 300 , 60O , and 900. Note from these results that for any particular reentry range the returntrajectory inclination which is required changes rapidly with the position of the moon. For example ., consider the results for a 30 0 reentry range. If the return flight is initiated when the moon has an angular position of about 230 0 , then the return trajectory must be inclined 70 0 t o the earth-moon plane. Three days later, when the moon has a position of about 270 0 , the return trajectory must be inclined only 10 0 to the earth-moon plane. Further, the time period during the month when a return flight to this latitude is possible is a function of the reentry range available. For example, with a 300 reentry range a return flight is possible only when the moon is near its position of maximum negative declination. With a 60 0 reentry range approximately half a month is available ., and with a 90° reentry range a return flight can be made at any time during the month. These time intervals place no constraints on the magnitude of the return inclination required. It is desirable ,, however, to utilize return trajectories with low inclinations to the earth-moon plane, not only from a guidance standpoint, but also because of the fact that return flights initiated from lunar orbits having low orbital inclinations may require low return-trajectory inclinations. In order that a touchdown latitude of 30 0 be achieved with reasonable values of the return inclination, the mission should be initiated when the moon is approaching its position of maximum negative declination. This time restriction may be alleviated somewhat if an additional touchdown point is assigned which has a southern latitude of 300. The variations in this case are similar 12
to those shown in figure 2, except for a shift along the abscissa of 180 0 , and reasonable values of the return inclination may now be obtained when the moon is approaching its position of maximum positive declination. In summary, the return trajectory must be such that the vehicle will reenter the earth's atmosphere within a desired corridor, and it must be inclined to the earth-moon plane at some specific angle which depends upon the time of the month selected for the initiation of the return flight and the reentry range available. The problem of the return mission, then, is one of determining how these requirements may be met or, alternately, of determining the values of the returntrajectory inclination which can be achieved from an arbitrarily oriented lunar
It is convenient in a parametric study of this type to utilize the "patchedconic" type of solution (ref. 1), wherein the return vehicle is assumed to be subject only to the attraction of the moon within the lunar sphere of influence and subject only to the attraction of the earth exterior to this region. Thus, the problem becomes a combination of two solutions, one geocentric and the other selenocentric. With this type of approach, it is possible to determine the conditions at the lunar sphere of influence which will satisfy the reentry constraints. When these conditions are known, it is then possible to determine the lunar orbital characteristics which satisfy them. Further, it will be assumed in the discussion which follows that the vehicle exits normal to the lunar sphere of influence. Actually, the angle between the vehicle velocity vector and radius vector at this point is of the order of 3.5 0 , and results obtained from this study, without the use of this assumption, are presented in reference 2. However, by using this assumption here it is possible to obtain a very good physical picture of the phenomena associated with the return flight,
RESULTS AND DISCUSSION
Sphere of Influence Conditions The types of results which may be obtained from the analysis of this problem are shown in figure 3. For orientation purposes, a sketch of the lunar sphere of influence is shown in the upper portion of the figure. The direction to earth and the direction of the moon's motion are indicated by arrows. Two angles which will be referred to are a, or the longitude of a position on the lunar sphere of influence, measured from the earth-moon line, and 71 or the latitude of a position on the lunar sphere of influence, measured from the earth-moon plane. In the lower portion of figure 3 are shown the loci of vehicle exit points on the sphere of influence which result in a satisfactory earth return. The ordinate is the latitude of the exit position, and the abscissa is the longitude of the exit position. Only representative values of the return inclination of
10 0 , 4oO , and 900 are shown. The loci for all other return inclinations will fall between the 900 variations. If the vehicle exits from the sphere of influence anywhere along one of these loci, then the return trajectory will satisfy the reentry-corridor constraint and have the inclination specified. If the exit point has a negative latitude, the vehicle will return to the earth's Northern Hemisphere. If the exit point has a positive latitude, the vehicle will return to the earth's Southern Hemisphere, and the return trajectory will be oriented as shown by the dashed line in figure 1. The implication of the results shown in figure 3 is that in order to return to the earth the vehicle must exit from the trailing edge of the lunar sphere of influence within a rather narrow band defined by latitude limits of approximately ±10 0 . For orientation purposes, this area is indicated by the shaded region in the upper portion of the figure. The parameter which distinguishes one exit point from another along any one of these curves is the vehicle injection velocity increment required to achieve earth return. The minimum velocity increment is associated with exit positions near 800 of longitude. As the exit position is shifted from this point in either direction, the required velocity increment increases. The total mission flight time - that is, the time from injection to reentry - decreases as the exit position is shifted toward the earth-moon line and increases as the exit point is shifted in the other direction. The reason for this increase in flight time for exit points having longitude values greater than 800 is that the vehicle is heading away from the earth when it leaves the sphere of influence. In view of the fact that there will be injection-velocity and flight-time limits imposed upon the return mission, the actual exit area will be smaller than that discussed so far. If the arbitrary, but representative, limits of 3,100 feet per second in the velocity increment and 100 hours in flight time are imposed, the exit area will be confined to the shaded region shown superimposed on the results in the lower portion of figure 3. The left boundary of this region is defined by the velocity limit of 3,100 feet per second and the right boundary, by the time limit of 100 hours. An increase in the allowable value of either of these two parameters will increase the size of this shaded area. For a successful return mission within the defined limits of velocity and flight time, then, the selenocentric, or lunar trajectory, characteristics must be such that the vehicle exits from the sphere of influence within this shaded region. The exact location of the exit point within this region will be determined by the return-trajectory inclination desired and the velocity required to achieve a specific longitude.
Lunar-Orbit Characteristics The required lunar-orbit characteristics may be assessed from an examination of the results shown in figure 4. The shaded region in this figure is the permissible exit area as defined previously. Superimposed on this exit area are the
,.ate traces of three representative lunar orbital planes which have inclinations to the earth-moon plane of 90 0 , 200 , and 6.50. The point to be made here is that since there are any number of arbitrarily inclined lunar orbital planes which intersect this area it is possible to return from a lunar orbit having almost any inclination desired. The only restriction on the value of this inclination is that it cannot be less than the latitude of the desired exit point within the required area. Once the inclination of the orbit is selected, however, it is not possible to choose arbitrarily the nodal position of this orbit, because it must be of such a value that the orbital plane will intersect some desired point in the required exit area. For example, if the desired inclination of the orbit is 20 0 , a return flight is possible for nodal positions having longitudinal values of approximately 800. However, if the nodal position is changed to that shown for the dashed line (approximately 120 0 ), the trace of this orbital plane does not intersect the required exit area, and a return is not possible with this inclination at this nodal position. There are two alternatives available if such is the case: first, an orbital-plane change can be made to bring the node to the proper position, and second, the initiation of the flight can be delayed. Because of the motion of the moon, the trace of any particular orbit in figure 4 will appear to precess from right to left at the rate of about 13.2 0 per day, and, therefore, there will always be some time during the lunar month when the trace of the orbital plane will intersect the desired exit area. One additional point may be noted with relation to this apparent precession of the node across the exit area. If the vehicle is to have a return capability at any time during the rendezvous phase of the mission, then, in general, the stay time will be determined by the time period that the trace of the orbit intersects the permissible exit area. Further, it is to be noted that it is not possible to achieve exit points within the entire permissible area if the lunar orbit has a low inclination to the earth-moon plane. For example, exit points with latitudes greater than 6.50 cannot be achieved with a lunar orbit which is inclined 6.50 to the earth-moon plane. Since the return-trajectory inclination increases with the exit-point latitude, it is apparent that the maximum value of the return-trajectory inclination which can be obtained with an orbit of this type is limited. This factor will introduce an additional restriction on the time period during the month when a return flight is possible. In summary, it has been shown that it is possible to achieve a satisfactory return flight from lunar orbits which have a wide range of values of inclination to the earth-moon plane and that once the inclination of the orbit has been selected the ascending nodal position must fall within a specified region to ensure proper return to earth.
As outlined by Robert H. Tolson in paper no. 1, the use of the lunar-orbitrendezvous technique to achieve landing sites anywhere on the lunar surface necessitates the establishment of lunar orbits having a range of inclinations from 00 to 90 0 , and associated with each inclination in this range is a specific nodal position which is determined by the earth-moon transfer-trajectory characteristics. Similarly, it has been shown in this paper that return flights may be achieved from lunar orbits having inclinations in the range from 0 0 to 90 0 , and that here again a specific nodal position is required for each inclination, depending in this case on the velocity and time selected for injection into the return trajectory. The problem remaining is to determine the relationship between the orientation of the orbit when it is first established and the orientation required for the return flight. This information is of fundamental importance to the lunarorbit-rendezvous technique since the selection of lunar landing sites and the associated stay times should be compatible with both the orbit establishment phase and the return phase of the mission. To illustrate this relationship, consider the results shown in figure 5. First, for orientation purposes, consider the sketch of the moon, with a vehicle in some arbitrary circular orbit, shown in the upper right-hand portion of the figure. The darkened regions in this sketch represent the required range of nodal-line positions in the earth-moon plane for lunar orbits having inclinations between 200 and 90 0 . If this area is viewed from the northern lunar latitudes, the result shown in the major portion of figure 5 will be obtained, where the direction of the moon's motion and the earth are again denoted by arrows. The hatched area lying between 1350 and 1+50 represents the nodal-line positions of those lunar orbits established from a typical earth-moon transfer trajectory and having inclinations between 20 0 and 900 . The stippled area lying between 30 0 and 1000 represents the range of nodal-line positions required for a return flight to either the Northern or Southern Hemisphere of the earth, within the velocity and flight-time limits discussed previously and for the same inclination range, that is, between 20 0 and 900. It is evident that the two areas are not compatible, and, therefore, a return flight cannot be initiated immediately from any lunar orbit having an inclination greater than 20 0 , unless flight times of well over 100 hours are acceptable. Because of the motion of the moon, the nodal line of any orbit established in the hatched area will appear to shift in the direction shown at the rate of about 13.20 per day, and, therefore, after some specified waiting period the nodal line will have the correct orientation so that a return flight is possible. Note that this may require a waiting period from 3 to 4 days. As the orbital inclination is decreased below 20 0 , both regions begin to expand, and below 10 0 they overlap; thus, a return flight may be initiated at any time after orbit establishment subject to the capability of obtaining the desired return-trajectory inclination and achieving a specified longitude.
The discussion of the return-to-earth portion of the lunar missions presented in this paper is intended to serve as a guide in the analysis of the overall manned lunar mission. Within the limits of the assumptions that were made in the analysis, it appears that a satisfactory return flight may be achieved from lunar orbits which have a wide range of values of inclinations to the earth-moon plane, and that once the inclination of the orbit has been selected the ascending nodal position must fall within a specified region to ensure proper return to earth. Further, if the lunar landing site dictates the use of an orbit having an inclination greater than 10 0 , then there must be some specified waiting period, after orbit establishment, prior to the time when a return flight is possible. If an inclination of less than 10 0 is utilized, then it appears that a return flight can be made at any time.
1. Plummer, H. C.: An Introductory Treatise on Dynamical Astronomy. Dover Publ., Inc., 196o. 2. Gapcynski, John P., and Tolson, Robert H.: An Analysis of the Lunar Return Mission. NASA TN D- 1 939, 1963•
THE LUNAR RETURN PROBLEM
MOON MOON'S MOTION
RETURN TRAJECTORY EARTH REENTRY I = LUNAR-ORBIT INCLINATION TO THE EARTH-MOON RANGE PLANE ~^ R ^ - D=ASCEND|NG NODAL POSITION OF LUNAR ORBIT I~RETURN -TRAJECTORY INCLINATION TO THE EARTH-MOON PLANE
TOUCHDOWN LATITUDE = 30 0
RETURNTRAJECTORY INCLINATION, I, DEG 120 r-
MOON'S MOTION 13i20/DAY ^--->
REENTRY RANGE, DEG
- 0 40
120 160 200 240 280 320 360 ANGULAR POSITION OF MOON, DEG DECLINATION
'° ^^ w *** ° °^ n °° ° * " ° w= ^ ° o ~ ^ o ~ *~ * ~ s °~^ ° ° ^ , ~ » ,° ^*^ *^ ^^^ ^ ^ ^° °^ ° ^
° °°^ °° " ~^ e ° -^ ° ° ° ^*, °"
U]C| OF EXIT POINTS ON THE LUNAR SPHERE OF|NFLUENCE
VELOCITY INCREASES ~-----^ VELOCITY INCREASES TIME DECREASES TIME INCREASES 20LATITUDE, 77, D DEG -2O^
20 40 60 80 100 120 140 160 180
LONGITUDE, u,DEG Figure 3
LUNAR-ORBIT CHARACTERISTICS FOR THE RETURN W4|SS|DN
2O LATITUDE 77,O
LONGITUDE, a, DEG
ORIENTATION OF LUNAR ORBITS
20° < i :5 900 MOON'S MOTION NODAL-LINE ORIENTATION FOR ORBIT FOR RETURN--\ ESTABLISHMENT, TO 145* EARTH 1350 M OONS MOTIONI
APPARENT o NODE SHIFT 13.2-/DAY 100
--- ---------------- --20
A SIMPLE GUIDANCE TECHNIQUE FOR LUNAR LETDOWN By Richard Reid
The back-up system proposed herein would allow the pilot of the lunar excursion module (LEM) to predict, during the descending coast phase of orbit transfer, what the altitude error would be at the point where the braking descent was to begin. The system would provide sufficient information to enable the pilot to make a correction by means of a simple thrust maneuver and attain the preselected conditions for the start of the braking descent.
Studies are underway at the Langley Research Center to determine methods of utilizing pilot capability for back-up systems in the lunar mission. (See refs. 1 and 2.) In the investigation of this paper, requirements and accuracies have been determined for a system wherein the pilot of the lunar excursion module (LEM) monitors the descending coast phase of the orbit transfer maneuver. The purpose of this investigation is to develop a system that will enable the pilot to monitor his position and make necessary corrections to assure arrival at a preselected altitude and range angle for the start of the braking descent.
GENERAL CONSIDERATIONS AND PROCEDURES
In this study it is assumed that the Apollo vehicle has established an circular orbit. At a preselected point a thrust maneuver is used to transfer the LEM to a synchronous orbit with an altitude at pericynthion of 50,000 feet for a range angle 0 of 94.40 . In this study the range angle is measured from the orbit-transfer point. From pericynthion the LEM begins the braking descent to the lunar surface.
The magnitude of the error in thrust is the only error parameter considered in this study for the purpose of perturbing the descent trajectory. The error in alinement of the thrust vector is of significance also, but is not used in this illustration. It is expected that the method of correcting orbit-transfer errors outlined in this paper for the range angle of 94.4 0 , where braking to the lunar surface begins, would be valid for errors from various sources. During the coast phase the pilot of the LEM utilizes altitude information at known range angles. The range-angle information could be obtained by the . pilot from an optical sighting on a known lunar landmark or on the Apollo vehicle in its known orbit® The altitude at these angles could be obtained either with a radar altimeter or with optical sightings. This information enables the pilot to determine what his altitude would be at a range angle of 94®40 . If his altitude will not be 21
. ^. sufficiently near 50,000 feet at 94.4 Y a correction is made by applying thrust along the local vertical. The procedure of making a midcourse correction standardizes the lunar-landing maneuver. Since the corrections required will be small, thrust along the local vertical is used for simplicity. This thrust can be applied with the reaction control motors.
This mission profile is shown in figure 1. The optimum maneuver for the LEM orbit transfer is shown by the dashed line. If too much thrust is applied, the LEM enters an "error orbit" which results in an altitude of less than 50,000 feet at 0 = 94.4 0 unless a correction is made. The figure shows a correction at 0 = 600 to place the LEM at 50,000 feet at 0 = 94.40 . If such a midcourse correction is made, the LEM is no longer in a synchronous orbit. Subsequent corrections will return the LEM to the original synchronous orbit if the pilot decides not to land. Figure 2 shows the errors in altitude at 0 = 94.4 0 that would result from thrust errors (expressed as errors in velocity change LV) in the orbit-transfer maneuver. Zero error results in an altitude of 50,000 feet at 94.4 0 . An error of approximately 50 fps in the transfer maneuver would result in the LEM hitting
the lunar surface at 94.40.
Figures 3(a), 3(b), and 3(c) are examples of the charts a pilot would use to determine what his altitude would be at 0 = 94.4 0 by taking altitude measurements at range angles of 30 0 , 45 0 , and 600 , respectively. In addition, these charts show the DU corrections required to reach 50,000 feet at 94.4 0 , by means of vertical thrust, if the pilot determines that there is an error in the orbit transfer. The hatched bands in the figures indicate a ±1 percent inaccuracy in altitude determination and the consequent inaccuracies in altitude at 0 = 94.40 and in correction-velocity requirements. By comparing figures 3(a), 3(b), and 3(c) it is seen that correction requirements are smaller for the smaller range angles, but inaccuracies in altitude determination are greater for the smaller range angles.
It is seen that an error in the orbit-transfer maneuver does not necessarily create an abort situation. By monitoring the coast phase, the pilot can detect errors, establish correction-velocity requirements, and perform a simple thrust maneuver to reach preselected conditions for the start of the braking descent. Corrections can be made that do not jeopardize the mission, since subsequent corrections can return the LEM to the original synchronous orbit.
REFERENCES 1. Queijo, M. J., and Riley, Donald R.: A Fixed-Base-Simulator Study of the Ability of a Pilot to Establish Close Orbits Around the Moon. NASA TN D -917,
2. Houbolt, John C., Bird, John D., and Queijo, Manuel J.: Guidance and Navigation Aspects of Space Rendezvous. Proceedings of the NASA-University Conference on the Science and Technology of Space Exploration, Vol. 1, NASA SP-11, 1962, pp. 353-366. (Also available as NASA SP-17.)
ORBIT TRANSFER \ `' ^SYNCHRONOUS ORBIT
EFFECTS OF VELOCITY ERROR
Of------- — — --
0 10 20 30 40 50 60 70 X 103 ALTITUDE AT 6 = 94.4°, FT
PREDICTION OF ALTITUDE AND CORRECTION VELOCITY
ALTITUDE AT 8= 30 0, FT
305 x103 300 295 290 285 280 275 270 NOMINAL TRANSFER
-20 -10 0 10 20 30 CORRECTION,
2650 10 20 30 40 50 60 70 X 103 ALTITUDE AT 8 = 94.4 0, FT
I I I
PREDICTION OFALTITUDE AND CORRECTION VELOCITY
ALTITUDE AT 8=45°, FT 220 X103
_ _ _ __
AV CORRECTION, FPS 12.5
170 i 1600 10 20 30 40 50 60 70 X103 ALTITUDE AT 8=94.4°, FT Figure 3 (b)
PREDICTION OF ALTITUDE AND CORRECTION VELOCITY
ALTITUDE AT 8 = 60*, FT
140, x 103 -30
800 10 20 30 40 50 60 70 x 103
ALTITUDE AT 8=94.4*, FT
VISUAL SIMULATION STUDY OF LUNAR HOVERING, TRANSLATION, AND TOUCHDOWN By Maxwell W. Goode
A pilot's ability to control hovering, translation, and touchdown on the lunar surface has been investigated by means of a three-degree-of-freedom analog simulation oriented toward contact flight. The pilot controlled the position of his vehicle in the vertical plane with a throttle and a side-arm controller. The results of this investigation indicate that the hovering, translation, and touchdown portions of a lunar landing are within the capability of a human pilot. There is, however, a control coupling between vehicle attitude and thrust that is not inherent in V/STOL vehicles operating in the earth gravity field. This coupling can cause difficulty and perhaps warrants further study. If the landing point is overshot, generally as a result of the coupling problem, a correction by backing up should be avoided if possible. Pilots preferred to fly a line-of-sight type of maneuver to the landing point from the hovering condition. A maximum angular acceleration of 0.5 radian/sec 2 seems adequate for a rate command system. A horizontal acceleration capability of 2 to 3 feet/sec2 near the surface appears desirable.
This investigation was a preliminary study of a pilot's ability to control hovering, translation, and touchdown on the lunar surface. In all probability the actual landing will be performed by a human pilot flying contact. This is based on the premise that final selection of the touchdown point will be made by the pilot. A capability of hovering and translation in the immediate vicinity of the lunar surface is a logical requirement to afford the pilot an opportunity to survey the surface, select the final touchdown point, and maneuver his vehicle to that point. It was assumed for this study that the landing vehicle had been brought to the hovering condition in the immediate vicinity of the surface after a deorbit braking maneuver and that the preselected landing site was unacceptable for a landing. It was the task of the pilot to select and translate to another point for touchdown.
DESCRIPTION OF APPARATUS
The simulation used in this study was a three-degree-of-freedom system oriented toward contact flight. The pilot controlled vehicle attitude and position F< 27
in the vertical plane with a throttle and a side-arm controller. The throttle controlled the rate of fuel consumption of a fixed rocket engine with a maximum initial thrust-weight ratio of 0.4g. The attitude was controlled by a proport tional rate command system. Figure 1 is a picture of the simulator showing the throttle, the side-arm controller, and the display presented to the pilot. Altitude was displayed along the Y-axis and horizontal distance along the X-axis. Vehicle position and attitude were indicated by the meter in the upper right corner of the display, shown as a closeup in figure 2. The position indicator was a pointer only and did not make a visible trace of the runs. Vehicle attitude in the vertical plane was indicated by deflection of the needle from the vertical. A fuel-quantity indicator was also included. This simulation was oriented toward contact flight in that the pilot had to judge his acceleration, velocities, and displacements from the motions of the display. The background of the display was blank, with only the surface and desired landing points included. The success of a landing was determined by the touchdown conditions. The established limits for touchdown were: horizontal velocity, less than 5 ft/sec; vertical velocity, less than 10 ft/sec; touchdown angle, less than 15 0 ; and miss distance, less than 40 feet. RESULTS Figure 3 is a graphic presentation of several landing trajectories from a 1,000-foot initial altitude and a 1,000-foot translation distance. An initial altitude of 500 feet and a translation distance of 600 feet were also used. An analytical study (reported in ref. 1) indicated that a ballistic-type trajectory was approximately 15 to 20 percent more economical than trajectories requiring continuous thrust. The fuel expenditure for the entire maneuver, however, represents only 2 to 3 percent of the initial hovering weight of the vehicle. Trace 0 is a ballistic trajectory. To fly this trajectory the pilot pitched the vehicle forward and applied maximum thrust for a short time. Hey then shut off the engine and coasted. As the vehicle approached the surface in the vicinity of the landing point, the pilot pitched the vehicle back, applied maximum thrust to brake, and made the landing. This trajectory had two distinct disadvantages: (1) There were too many parameters, such as the pitch angle, the firing time, and the correct braking maneuver, that had to be judged correctly in order to capitalize on any fuel economy, and (2) the vehicle was not under positive control at all times. Trace O2 is a modified ballistic trajectory. In this one the pilot pitched the vehicle forward, and after a few seconds he applied maximum thrust and then shut off the engine. In the touchdown region he pitched the vehicle back and applied maximum thrust again to brake and make . the landing. This trajectory had the same disadvantages as the true ballistic trajectory.
Trace ® is a helicopter-type approach. The pilot reduced power considerably and pitched the vehicle forward. After a desired horizontal velocity was obtained he rotated the vehicle to a vertical attitude and started arresting his vertical velocity, still coasting horizontally. As he approached the landing point he pitched the vehicle back to arrest the horizontal velocity and made the landing. This trajectory had a serious disadvantage in that it required braking at low altitude with a high pitch angle. When the horizontal velocity was arrested the pilot did not have a chance to correct any miss distance before the landing, and in some cases he did not attain a vertical attitude before touchdown. Trace O is the trajectory that showed the most promise and is the one that was used in the simulation. Generally, it was a line-of-sight flight path to the desired touchdown point. The piloting procedure for this trajectory was: (a) Pitch the vehicle forward approximately 30 0 to 350. (b) Reduce power slightly to fly a near line-of-sight path to a point estimated to be 100 feet above the desired touchdown point. (c) As the halfway point is reached, pitch the vehicle back approximately
30 0 to 35 0 to start braking.
(d) Manipulate the attitude of the vehicle when approaching the touchdown point to arrest the horizontal velocity. (e) Add sufficient power to arrest the vertical velocity and make the landing. This trajectory was under positive control at all times and could be flown for various initial altitudes or translation distances. The braking altitude before touchdown was sufficient to allow corrections in miss distance or vehicle attitude before touchdown. The speed with which the maneuver was performed was left to the discretion of the pilot. A series of preliminary runs was made with trajectory Q to determine an acceptable attitude-control system. For this series the pilot was given a rate command system with a maximum angular acceleration of 1 radian/sec 2 and a maximum angular velocity of 1 radian/sec. Figure 4 is a curve of the attitude-control power distribution for this series of runs. There are two points of particular interest: (1) The predominant braking point at 0.5 radian/sec 2 or less, used in 75 percent of the runs, and (2) the infrequent use of maximum available power. It was noted that the only time maximum power was used was in the turn-around maneuver at the halfway point, which was not critical. This established the criteria for attitude control power for the actual tests: maximum angular acceleration of 0.5 radian/sec 2 and maximum angular velocity of 1 radian/sec. For fuel with a specific impulse of 315 seconds, a fuel weight of approximately 3.3 percent of the initial hovering weight of the landing craft allowed the pilot to land safely in all cases with translation distances up to 1,000 feet. It is also interesting to note that the time required for translation and touchdown 29
m x .:x
was approximately 1 minute, and the fuel consumed was about the same as for an equivalent period of hovering. Figure 5 shows the distribution of fuel usage as a function of translation distance for two typical runs. Approximately the same total fuel was required for the two runs, and the curves indicate that the distribution of fuel usage was about the same. Starting and braking to a landing consumed a large portion of the fuel - 15 to 20 percent for starting and 35 to 45 percent for braking and landing. Thus a relatively large change in translation distance would require only a small change in total fuel. One of the most significant results of this study was the disclosure of a control coupling between vehicle attitude and thrust. Generally speaking, attitude controlled the translational conditions, and thrust controlled the altitude conditions; but the relatively large pitch angles involved caused pronounced cross coupling. To help explain the source of the coupling problem, figure 6 is an illustration of the force vectors of a rocket-propelled landing vehicle. The thrust for hovering on the moon will be 1/6 of that required on earth because of the 1/6 gravity term. To translate with a fixed rocket-engine system in the earth gravity field, a pitch angle of 5.7 0 is required to produce a horizontal acceleration of l/lOg with no appreciable change in thrust to maintain altitude. However, in the lunar gravity field a pitch angle of 31 0 will be required with an increase of 17 percent in thrust to maintain altitude. At this pitch angle approximately 50 percent of any thrust change is realized in horizontal acceleration as compared with approximately 10 percent for the earth case. This coupling manifested itself in ballooning or overshooting the landing point. It was, generally, better to choose another suitable landing point rather than to try to correct an overshoot by backing up, especially when very close to the surface. The coupling was not necessarily uncontrollable, but suggests a restriction of pitch angle close to the surface. Figure 7 shows the distribution of maximum horizontal acceleration below an altitude of 100 feet. In most of the runs the maximum horizontal acceleration was 2 to 3 ft/sec2 or less. Horizontal acceleration was achieved by tilting the vehicle in this study; however, if horizontal-firing engines were considered as a solution to the coupling problem, engines of the 1,000-pound-thrust class would be required to produce an acceleration of about 3 ft/sec2 for a vehicle weighing about 10,000 earth pounds.
This study of piloted control of hovering, translation, and touchdown on the lunar surface has yielded the following conclusions; 1. Pilots preferred to fly a line-of-sight type of maneuver from hovering to the desired landing point. 2. The effects of a control coupling between attitude and thrust can cause difficulty and perhaps warrant further study.
3- If the landing point is overshot, a correction by backing up should be avoided if possible.
4. An angular acceleration of 0.5 radian/sec t seems adequate for attitude control. 5. A horizontal-acceleration capability of 2 to 3 feet/sec t close to the surface appears desirable.
1. Sivo, Joseph N., Campbell ,, Carl E., and Hamza, Vladimir: Analysis of Close Lunar Translation Techniques. NASA TE R-126, 1962.
ATTITUDE AND POSITION INDICATOR
Figure 2 32
1,000 TRANSLATION DISTANCE, FT
ATTITUDE-CONTROL POWER DISTRIBUTION
PERCENTAGE OF RUNS AT OR BELOW 100 r-
MAXIMUM ANGULAR ACCELERATION USED, V, RADIANS/SEC 2
PERCENT FUEL luu
4 8 2 6 TRANSLATION DISTANCE, FT
lo x 10,
FORCE VECTORS FOR ROCKET-PROPELLED LANDING VEHICLE EARTH (g) HOVERING
-1 t- 53*
y 3 10
TO MOVE 100 FEET IN ABOUT 12 SECS, q/to
g= 32.2 FT/SEC2 Figure
L-2o 34 -6
DISTRIBUTION OFMAXIMUM HORIZONTAL ACCELERATION BELOW 100 FEET
60 PERCENTAGE OF RUNS AT OR BELOW *i 40
MAXIMUM HORIZONTAL ACCELERATION USED, 'i, FT/SEG 2
Page intentionally left blank
5. LANDING CHARACTERISTICS OF A MODEL
OF THE LUNAR EXCURSION MODULE By Ulysse J. Blanchard
An experimental investigation has been made of some lunar-landing characteristics of a 1/6 -scale dynamic model of a lunar excursion module having multiplepoint leg-alighting-gear systems. Symmetrical four-point and five-point gear systems and an asymmetric four-point system were investigated. The alightinggear legs incorporated a yielding-metal strap for energy dissipation. The landing tests were made by launching a free model onto an impenetrable hard surface (concrete) and onto a crushed-pumice overlay of various depths. Landing motions and acceleration data were obtained for a range of touchdown speeds, touchdown attitudes, and landing-surface conditions. Landing characteristics of a type of the lunar excursion module were satisfactory over a considerable range of touchdown parameters for the surface conditions investigated. The best gear orientation for landing was two legs forward. Compared with the four-point gear, the five-point system improved overturn stability slightly during landings in pumice dust. Considerable improvement in overturn stability was obtained with the asymmetric four-point gear having the two long legs extended forward in the direction of horizontal flight.
Two important items for overall success of the Apollo mission are a soft and stable landing of the lunar excursion module on the lunar surface and the subsequent launch of the spacecraft from the lunar surface. The alighting gear must arrest and stabilize the module and provide a stable platform on a variety of probable surface conditions. Multiple-point leg systems with simple nonviscous low-rebound shock absorbers are attractive with regard to these general requirements. In this paper are discussed the results of an investigation of a preliminary version of the lunar excursion module employing a multiple-point leg-gear system with yielding-metal shock absorbers and several contact-point arrangements. Impact acceleration and behavior were determined with a 1/6 -scale dynamic model over a range of touchdown speeds and attitudes. The landing-surface configurations investigated included a smooth hard surface, various depths of dust overlay, and a ledged surface.
A a F g I Z m t V a
area acceleration force acceleration due to gravity moment of inertia length mass time velocity angular acceleration gravitational ratio geometric model scale factor stress of yield strap angular velocity
A o w
MODEL AND APPARATUS
The 1/6-scale dynamic model used for the investigation is shown in figure 1. Simulated full-scale spacecraft height was 185 inches, basic landing-gear diameter was 222 inches, and initial center-of-gravity height above groundline was 108 inches® The basic landing gear was a symmetric four-point arrangement, as shown in figures 1 and 2(a). Each of the four legs consisted of three struts mounted so as to form an inverted tripod. The upper strut telescoped during impact, and the lower V-strut was a hinged unit which served to guide and stabilize the tripod. Impact energy was absorbed by yielding a pure nickel strap with the telescoping strut. The landing pads were ball-joint mounted. Vertical working stroke for the full-scale gear was approximately 2 feet and was designed to give a maximum load of 2 earth g units at an impact speed of 15 ft/sec. Modifications to the basic landing-gear arrangement are illustrated in figure 2. Figure 2(b) shows the asymmetric four-point arrangement in which the forward pads were extended in the direction of horizontal flight® The radial distance to the center of the offset pads was 33 percent greater than that for the
basic gear arrangement. Figure 2(c) is a symmetric five-point arrangement in which the radial distance to the center of the pad was the same as for the basic gear arrangement. The scale relationships pertinent to the earth model tested are shown in table I. For geometrical scaling the characteristic length was varied as the model scale factor T. A 1/6 -scale model was used inasmuch as this was suitable for meeting construction, size, and weight requirements for test purposes. The same yield-strap material was assumed for the 1/6 -scale model and the full-scale vehicle; hence, the stress was held 1 to 1, and thus exact structural scaling of shock-absorber forces was provided. The gravitational ratio R is dictated by the fact that the force of the earth's gravity is six times that of the moon, requiring that accelerations experienced by the model be six times that resulting on the moon. With these three relationships fixed, other pertinent scale relationships follow from laws of physics for a dynamically scaled model. The investigation was conducted by launching the model as a free body with the pendulum device illustrated in figure 3. Desired landing attitudes were obtained by adjustment of a support-release mechanism mounted on the launch platform. Horizontal speed was obtained by presetting the initial height of the platform above "dead center," and vertical speed was obtained by adjusting the free-fall height of the model. At dead center the model was released, and it impacted upon one of the surfaces illustrated in figure 3. The landing surfaces included a flat hard surface, a ledged hard surface, a flat hard surface with crushed-pumice overlays of various depths, and a ledged hard surface with a crushed-pumice overlay at the lower level. The ledge simulated a 2-foot dropoff. The pumice simulated an average full-scale particle size of 400 microns. Estimated bearing strength of the loosely compacted pumice during model tests was about 2 lb/sq in. with a bulk density of about 40 lb/cu ft. In general, these surface conditions were similar to those of the lunar surface model of the Manned Spacecraft Center.
RESULTS AND DISCUSSION
All data presented are full-scale values. Impact data presented are converted to lunar values in terms of the earth's gravity.
Landing Impact Typical accelerations measured at the center of gravity of the module with the basic gear during landings on a flat hard surface are shown in figure 4. Maximum full-scale normal, longitudinal, and angular accelerations are plotted against horizontal speed for two touchdown pitch attitudes ( - 15 0 and 0 0 ), as shown by the sketches, and for two gear orientations ,(one leg and two legs forward). Vertical speed for these data is constant at 10 ft/sec. Resultant flight-path angles varied from 90 0 to 330 . The sliding coefficient of friction on the hard surface was about 0.4. Results for a pitch attitude of 150 and for
landings astride the ledge were about the same as for those at a pitch attitude of -15 0 . As expected, for a constant-force shock absorber, there was little change in maximum acceleration values over the entire speed range. Accelerations for all three configurations during hard-surface landings were within the limits shown on this figure, and no overturning occurred. Typical accelerations obtained during landings on a flat surface with pumicedust overlay are shown in figure 5. Maximum normal, longitudinal, and angular accelerations are plotted against variation of horizontal speed for pitch attitudes of ±15 0 and for dust 5 and 21 inches deep. Again, vertical speed is constant at 10 ft/sec. These landings were made with two legs forward. Acceleration characteristics were generally similar to those for hard-surface landings. Negative longitudinal acceleration (or drag) increased by a factor of 2 or 3 to about 17 g units with increase in horizontal speed. It was estimated that the coefficient of friction was 0.7 to 1.0 in pumice. Drag forces also decreased righting moments, as indicated by the reduced angular accelerations at a pitch attitude of -150 . These factors resulted in the overturning of the module in the direction of the flight path, as indicated by the solid symbols. The sample data in figures 3 and 4 show that the maximum landing accelerations were small and did not vary a great deal over a wide range of landing conditions. Maximum normal acceleration was 2g units, maximum longitudinal acceleration was 1^ units, and ,g maximum angular acceleration was 12 2 radians/sec2.
Landing Stability A comparison of overturn characteristics of the three configurations is provided in figure 6 for various pitch attitudes and horizontal velocities. These characteristics were obtained for simulated landings in a 21-inch pumice overlay at a constant vertical speed of 10 ft/sec. These landings were made with two legs forward. Conditions where overturning occurred are shown by the solid symbols. The symmetrical four-point configuration overturned at a horizontal speed of 10 ft/sec or greater, depending upon touchdown attitude. The symmetrical fivepoint configuration increased overturn stability slightly, as can be seen by comparing the circular and square data symbols at a pitch attitude of -15 0 and a horizontal speed of 10 ft/sec. Here the four-point configuration overturned and the five-point configuration remained upright. With the asymmetric four-point gear, there is a significant improvement in overturn stability, as can be seen by comparing open triangles with the solid circles and squares at a horizontal speed of 15 ft/sec and at pitch attitudes from -15 0 to 15 0 . The symmetric four- and five-point configurations overturned, whereas the asymmetric configuration remained upright. A summary of stability and motion characteristics illustrated in motion pictures of several typical test runs is presented in table II. During hard.-surface landings with horizontal speed, a more steady platform was obtained during and after slideout by landing with two legs forward. Also, landing with one leg forward in the pumice overlay resulted in some directional instability and overloading of the lower V-strut of the forward leg. 4o
z 9 2 }
Results of an experimental investigation have shown that the landing characteristics of a type of lunar excursion module having multiple-point leg-gear landing systems and yielding-metal shock absorbers were satisfactory over a considerable range of touchdown parameters. The best gear orientation for landing was two legs forward. The asymmetric gear arrangement - that is, , two long legs forward - improved overturn stability considerably. Final design of the landing gear is greatly dependent upon the detailed surface characteristics of the lunar landing site, of which little is currently known.
TABLE I.- SCALING LAWS = Geometric model scale factor; Quantity Length Stress (yield strap) Acceleration Area Force Mass Velocity Time Inertia Angular velocity Angular acceleration
Gravitational ratio, Scale factor
Full-scale lunar model
A2 ?12 ?,2/p
Pa A2A ?2F (A2/p)m fp_A V
F/a fa-1 V/a ml2 1/t 1/t2
FP _x V/P
CP 77 P/?\
TABLE II.- SUMMARY OF CHARACTERISTICS OF TYPICAL TEST RUNS
Vertical speed, ft/sec
Horizontal speed, ft/sec
Flight-path angle, deg
Pitch attitude, deg
Gear orientation, number of legs forward
Symmetrical four point
Flat, hard Flat, hard
Very little rebound. Undamped pitch oscillation between leading and trailing pads about two side pads during slideout. Unsteady platform. Two trailing pads impact initially on ledge followed by two leading pads impacting on lower level. Stable smooth slideout. Initial impact on trailing leg. Side pads missed ledge and essentially all energy absorbed by leading leg. Leading-leg shock strut bottomed. Stable.
Ledged, hard (2-ft ledge)
Ledged, hard (2-ft ledge)
5 6 7 8
Flat, pumice overlay ( 5 -in. overlay) Flat, pumice overlay ( 5 -in. overlay)
10 10 10
10 15 15
Overturned. Short slide through pumice and small shock-absorber stroke. Very nearly overturned as vehicle veered to left. Lower V-strut of leading leg buckled and module overturned. Impact similar to hitting solid obstacle. Stable. Increased horizontal displacement through pumice and increased shock-absorber stroke. Stable.
Flat, pumice overlay (21-in. overlay) Flat, pumice overlay (21-in. overlay)
Asymmetrical four point
Flat, pumice overlay ( 5 -in. overlay)
Ledged, pumice overlay at lower level (2-ft ledge, 21-in. overlay) Ledged, hard (2-ft ledge) Ledged, pumice overlay at lower level (2-ft ledge, 5 -in. overlay)
Symmetrical five point
Module righted itself after slideout. Stable platform. Stable.
12 -F" UI
(a) SYMMETRICAL 4-POINT ARRANGEMENT.
RECTI ON OF MOTI ON
— (b) ASYMMETRICAL 4-POINT ARRANGEMENT.
°~ * v ^ o ^"
° n ° ° wo v ^ p ~^^ °~
° v ~ ° " + °" w °°°
m ^ ^* ^ °~° ° ° o °^ n ° ° ^ ^ ^
" " ^ ^
^°^ *" ~ ~°°
"p ° ^ ° ° °°
MAX. ACCEL -------- NNORMAL,2 g units O 2 LONG., gunits 0 -2 ANGULAR, RADIANS
HARD-SURFACE LANDINGS SYMMETRICAL 4'P0|NTGEAR » ONE LEG FORWARD N TWO LEGS FORWARD o-____ ^o _^—_^±^^^l5`__^^^1`__
O 5 lO 15 HORIZONTAL SPEED, FT/SEC
v a °°° p^ ° °" °° v ° *~* r ^° ° y *,^ ^ pn ° *°° ^° °° o~ °°» ° o ° ^ ^ ^^° ° ^ ^ ^ °
` _ DUST—SURFACE LANDINGS SYMMETR| CAL 4'PO|NTGEAR STABLE OVER- DEPTH, TURN |N. o m 5 ^ ^ 21 l5
^^ NORMAL 2 gunib O 2 LONG., gunits 0 -2 20 ANGULAR. RADIANS SEC -20 O
o l} —~
lO 15 0 5 HORIZONTAL SPEED, FT/SEC
DUST — SURFACE LANDINGS
20 o ^ STAOLE PITCH AT[ ' DEG o o u v«E«TURN o SYM.4-P[ m SYN. 5-PT A ASYhl4'P[
- 0 o o
5 f5 10 HORIZONTAL SPEED, FT/SEC 20
LUNAR TOUCHDOWN OPERATIONS AS AFFECTED BY ABORT CONSIDERATIONS By Gary P. Beasley
A preliminary analytical study was conducted to determine the boundary or "deadman" curves for combinations of altitude and descent rate for which aborts from the final letdown phase of a lunar landing are possible. The study took into consideration various lunar-excursion-module parameters and descent conditions at the time of abort. The results of the study established the minimum altitude threshold for aborting the landing and defined the region in which aborts are not possible. INTRODUCTION The final letdown phase of the lunar landing mission (the region from hover to touchdown) is an extremely critical region from an abort standpoint due to the short times and altitudes available for aborts. Because of this phase's critical nature, it is important to define the boundary or "deadman" curves of rate of descent and altitude for aborts in this region, that is, to determine the combination of altitude and rate of descent for which aborts are possible. Because of this need, an analytical study has been started to determine the boundary curves for this region, a wide range of lunar-excursion-module (LEM) characteristics and conditions at the time of abort being taken into account. SYMBOLS F,Fl ,F2 ,F3' F4 reaction-control engine thrust at various locations FA thrust of abort engine h altitude rate of descent altitude at time of abort rate of descent at time of abort moment arm
ho ho I
tR response time
earth weight pitch angle
0 0 pitch angle at time of abort 0 pitching acceleration VEHICLE CHARACTERISTICS AND OPERATIONAL CONSIDERATIONS OF THE STUDY The LEM control parameters taken into account in the study are shown in figure 1 and consist of the thrust capability of the reaction-control system (RCS) engines (which could be throttled), the moment arm at which these engines fired, and the thrust capability of the main abort (ascent) engine. These quantities determine the reaction-control acceleration and the main deceleration capability available. The pertinent conditions at the time of abort are the pitch angle, rate of descent, and altitude. Along with the control parameters and conditions, the study also included some operational considerations, one of these being that the pitch angle be zero or near zero before staging occurs and the abort (ascent) engine fires. This statement is not meant to imply that the abort engine must be vertical before staging and firing but was assumed for efficient use of abort engine thrust. This assumption also presents the more severe case to be studied. Other operational considerations were the decision time of the pilot and the staging time of the LEM. It might be noted here that for the study an abort was considered successful if the rate of descent of the LEM was halted before the LEM hit the moon. BOUNDARY CURVES FOR AN ASSUMED LEM CONFIGURATION In order to provide a first look at the type of results this study will produce, some boundary curves were prepared for an assumed nominal LEM configuration. The nominal LEM parameters to be discussed and the boundary curve for this configuration are shown in figure 2. As indicated in the figure, the maximum thrust level of the RCS engines was considered to be 75 pounds per engine at a moment arm sufficient to provide a pitching acceleration of 5 0 per second squared by using differential throttling of two engines. The thrust-to-earth-weight ratio of the abort engine was assumed to be 0.6 and the initial pitch angle of the LEM was assumed to be 400 at the time of abort. This appears to be a rather large pitch angle for the LEM to have at such low altitudes, but large angles such as this might be encountered when establishing translation velocities as discussed in paper no. 4. Other considerations assumed for the nominal LEM configuration were that: (1) the LEM was required to be perpendicular to the local vertical before the abort engines fired and (2) a 3- second combined pilot reaction and LEM staging time was necessary for the abort procedure. When the LEM configuration and initial conditions as well as the operational considerations mentioned are taken into account, the study provides the boundary curve for this case as indicated in figure 2.
The abscissa in figure 2 is the altitude of the LEM at the time of abort and the ordinate is the rate of descent of the LEM at the time of abort. The area to the right of the boundary curve represents combinations of rate of descent and altitude, at the time of abort, for which aborts are possible. Also shown in the figure is the curve for combinations of rate of descent and altitude for free fall of the LEM to the moon's surface, an acceptable vertical impact velocity of 20 feet per second being assumed. The important result for the configuration assumed, as indicated in figure 2, is that the two curves define the region in which aborts are not possible. For example, for altitudes below 250 feet, an abort cannot be completed safely before hitting the moon's surface regardless of the rate of descent the LEM has at the time of abort. Thus, means of shrinking the no-abort region must be developed in order to provide abort capability during the whole letdown phase.
EFFECTS OF CHANGING PARAMETERS
To indicate the effect on the boundary curves of changes in the LEM parameters and conditions, a preliminary study of some changes and their consequences is shown in figure 3. The first curve on the right in figure 3 is the nominal boundary curve discussed in figure 2. Also shown in figure 3 is the free-fall curve mentioned in figure 2. The first two curves to the left of the nominal curve show the effect on the nominal boundary curve of changing the abort engine thrust capability from 0.6g to l.Og and decreasing the lag time from 3 seconds to 1.5 seconds. These changes provide little increase in abort capability. The third curve indicates the effect of having a pitch acceleration of 28.60/sec2 (0.5 radian per second squared) which was found to be desirable from a pilot's standpoint in the simulation study discussed in paper no. 4. This increase in pitch acceleration decreases the minimum abort region at h = 0 to 100 feet. The last two curves are concerned with the problem of the pitch attitude of the LEM at the time of abort. The first assumes that the abort engine of the LEM is staged and fired at 400 to the vertical until the rate of descent is halted rather than erecting the LEM and then firing. The second curve represents the case where the pitch angle is zero at the time of abort. Both of these curves appear to provide abort or survivable free-fall capability throughout the entire letdown phase. The maneuver which includes staging immediately and firing the abort engines while at the same time pitching the LEM back to 0 = 0 provides a boundary curve close to the 0 0 = 0 curve. Results from a piloted simulation study of lunar landings (paper no. 4) indicated that the successful combinations of h and h which occurred tended to fall to the right of the 0 0 = 0 curve. In general, the combination of h and A tended to be closer to the 0 0 = 0 curve at altitudes near 400 feet than at the lower altitudes. Thus, in case of abort situation, abort near an altitude of 400 feet would have been more critical than an abort at a lower altitude.
Results of a preliminary study have indicated that it is impractical to reduce the lag time to below 1.5 seconds or increase the thrust level of the abort engine because of abort consideration alone since little abort capability is derived from large changes in these parameters. It can also be concluded that even though relatively large changes in the no-abort region are obtained by changing the pitch acceleration, it still does not provide the abort capability required. The results also show that in order to provide abort capability down to zero altitude, the vehicle must be made capable of firing the abort engine while tilted or, if this is not possible, it must be kept erect, a condition which might dictate the need to provide translation rockets on the LEM. Combinations of these changes will also provide the abort capability necessary.
LEM VARIABLES INVESTIGATED
NOMINAL LEM CONFIGURATION
F = 75 LB PER RCS ENGINE
0 = 5o PER SEC 2
e =4do T
t R = 4 'SEC 3
h , h
FPS 40 —
NO-ABORT REGION FREE
-0 0. 0 6
0 LF FALL
SAFE-A BORT 10 REGION
EFFECTS OF CHANGING LEM VARIABLES — — o --- --- 2
5 SEC 80 L LEM 10 0
60 ho, FPS 40
DEVELOPMENT OF A SIMULATOR FOR STUDIES OF
SELF-LOCOMOTION OF MAN UNDER REDUCED GRAVITY By Donald E. Hewes and Amos A. Spady, Jr.
A technique for simulating reduced gravity for studies of the ability of man to walk, run, jump, and perform other locomotive tasks has been developed and proven to be practical and useful. Preliminary tests under conditions of simulated lunar gravity have shown that subjects dressed in lightweight street-type clothing experienced some difficulty when attempting to move about and climb stairs and ladders rapidly but could jump to vertical heights of about 12 feet. This technique appears to be useful in the design and development of space and lunar suits, space-station and lunar-housing facilities, and auxiliary locomotive devices such as a rocket-propelled jump pack.
Shortly after completing the lunar landing, the astronauts will have the task of climbing out of their vehicle and moving about the lunar surface to carry out exploration and scientific measurements. Although there is little doubt that they will be able to walk, run, jump, carry loads, or perform other tasks under the condition of reduced gravity in one way or another, there is essentially no information as to how well they will be able to do so, what the proper suit design should be, or what auxiliary devices may be required to assist them in performing their tasks. One of the main reasons for the lack of useful information, which is a serious problem confronting researchers in undertaking meaningful studies of the mechanics of self-locomotion, is the lack of a practical method for simulating reduced gravity with equipment that permits test subjects to move about freely over the considerable distances and extended periods of time necessary to perform the various locomotive tasks. The techniques that have been used for zero gravity, or weightlessness, studies, such as, immersion in water or riding in an airplane flying nearly a Keplerian trajectory, are not readily applicable to these studies. Consequently, efforts were made to develop a practical technique for simulating the reduced-gravity field.
DISCUSSION OF THE SIMULATOR
The principles on which the simulator technique was based are illustrated in figure 1. It was observed that any desired reduced , gravity could be achieved by inclining the test subject relative to the vertical gravity vector. In the case of lunar-gravity simulation, the subject would be inclined 9.50 from the horizontal. The body is supported in this attitude by a series of cables located at
the indicated points. In performing most of the self-locomotive activities under normal conditions, the body members move primarily in parallel planes; consequently, the test subject is able to perform the same activities in a more or less normal manner even though the body members are restricted by the cables. A somewhat oversimplified illustration of this point is that of a person walking down a very narrow hall. Figure 2 illustrates the general layout of the equipment that was developed. The test subject is supported by a series of small cables attached to a lightweight crossbar which is in turn attached to a trolley which is free to move along a monorail. A test operator or, possibly, a servo-controlled drive unit is used to keep the trolley and, consequently, the cable directly over the test subject as he moves either forward or backward. An inclined walkway runs parallel to the monorail track and represents the surface of the moon or the floor of the space station. The displacement of the walkway from directly beneath the track establishes the inclination angle of the test subject, the supporting cable, and, consequently, the magnitude of the simulated reduced gravity. The case of zero gravity, or weightlessness, is produced with the walkway directly beneath the track or without the use of any walkway. It should be noted that this technique is useful for studies of the mechanics of self-locomotion or other dynamic problems but is not suited to physiological or psychological studies of approximate weightlessness inasmuch as the internal organs are still subject to the earth's gravity. A photograph of the suspension system that was used in the development of a mock-up of the simulator is shown in figure 3. The test subject's body is supported at the head, the upper torso, the buttocks, and the calf of each leg by individual suspension cables attached to the crossbar shown in figure 2. In some cases, depending on the type of task the subject wished to perform, the arms were also supported just below the elbow joint. Support of the lower leg proved to be the most difficult problem but was achieved by use of a lightweight metal tubular frame that supported the leg from behind and permitted freedom of movement for the upper leg. The frame was supported by one of the suspension cables. The mock-up of the simulator, part of which is shown in figure 3, utilized a monorail system which had an elevation above the floor of about 40 feet and permitted a total cable length of slightly less than this distance. A small trolley was used for this mock-up but for simplicity was not controlled by an operator or a drive system to keep it directly over the test subject at all times. Consequently, the test subject had to drag the trolley with him as he moved back and forth along the walkway. A 16-foot walkway was used for these preliminary tests but was considered too short for practical testing purposes.
TESTS AND RESULTS
Several test subjects of various physiques and dressed in normal street clothes were used to perform a series of locomotive tasks consisting of walking, running, jumping vertically, and climbing ladders, poles, and stairs both with
and without the simulator. Motion pictures of the body movements for the conditions. of earth gravity and simulated lunar gravity were taken; also, the subjects were asked to make comments on their ability to perform under both conditions. In general, the subjects had little difficulty in walking and running at slow paces but had trouble in starting and stopping when trying to move at fast paces in the simulated lunar gravity. The subjects were able to jump vertically about 2 feet from a standing position in the earth gravity and could jump the equivalent of about 12 feet in the simulated lunar gravity; however, they experienced considerable difficulty in maintaining their proper attitude during the jump and aften lost their balance by the time they landed. In no cases, however, did they sustain injuries in falls from these heights. The cable length of about 37 feet proved to be too short for these jumping tests and subsequent climbing tests because of the change in simulated gravity resulting from the change in cable angle as the subject increased his distance from the walkway. A cable length of at least twice this length is recommended to minimize the gravity gradient. Another problem associated with the mock-up equipment was the friction drag and mass of the overhead trolley which interfered somewhat with the subjects' movement along the walkway. The suggested use of a manually or servo-operated trolley is recommended to alleviate this problem. The subjects had no difficulty in climbing stairs and ladders at various paces under earth gravity but experienced difficulty in keeping the legs properly coordinated when using a fast pace in the simulated lunar condition. The climbing of a vertical pole could be accomplished much more easily in . the lunar gravity than in the earth gravity and even could be performed easily with the use of only one hand in the simulator. The test subjects could perform easily with no practice a number of gymnastic feats including hand stands and backward and forward flips in the simulated lunar gravity that normally would be performed in the earth gravity only by a gymnast. In all cases, the subjects required only a few seconds to become accustomed to the facility and were able to perform the body movements normally without loss of orientation or feeling of discomfort. There were no cases of any serious or sustained aftereffects. CONCLUDING REMARKS In conclusion, the developed technique for simulating reduced gravity appears to be a practical means for studying many aspects of the self-locomotive capabilities ofman working in the lunar gravity or some other reduced-gravity field such as would be encountered in a rotating space station. This technique should be very useful as a practical tool in the development of more efficient lunar and space suits, which in their present form seriously limit the wearer's capabilities. Also, results of extended studies in which this technique is utilized appear to be useful in the design and development of efficient space-station and lunar-housing facilities, practical erection and fabrication techniques, and
useful auxiliary locomotive devices such as jump packs to assist the lunar explorers. Finally, these test results should be helpful in planning the efficient logistic support for the lunar-exploration missions®
MAN SELF-LOCOMOTION STUDIES
TAKE-OFF, RENDEZVOUS, AND DOCKING
8. SOME ASPECTS OF MAN'S VISUAL CAPABILITIES IN SPACE
By Jack E. Pennington SUMMARY Several studies have been conducted to determine a pilot's capabilities to utilize visual information for rendezvous, docking, and navigation. The results show that visual control is efficient in many areas, and in other areas visual information can be used in secondary or backup control techniques. This paper attempts to summarize the results of these studies which relate to the Apollo mission. INTRODUCTION A number of studies of man's capabilities to control rendezvous have been made under both fully instrumented and visual-control conditions. These studies (see, for example, refs. 1 to 5) show that manual control of rendezvous under conditions for which visual references can be used is simple and effective. Studies of the terminal and docking phases of rendezvous showed a need for basic information concerning man's visual capabilities in a space environment. This paper summarizes the results of subsequent studies of man's visual capabilities in several areas of space operations which relate to the Apollo mission. SYMBOLS
separation distance, ft closure rate, ft/sec visual angle, deg
8 Subscript min
One technique for control of rendezvous is to bring the angular rate of the line of sight of the target vehicle in an inertial reference system to zero and
then to control closure rate to effect a safe rendezvous. In simulations utilizing this technique with visual references for background and target, it was found that pilots could detect the target and establish the intercept course. However, at the time these studies were made, no data were available to indicate to what accuracy the pilots could null the line-of-sight rate. Therefore, for this information, a separate study was made to determine the smallest angular rate that a human pilot could be expected to detect. (See ref. 6.) The same basic equipment used in the rendezvous simulations was used for this visualacuity study. An important parameter in the detection of angular rate is the angular separation between the target and the nearest reference star. The range of separation values between the target and the background used in this visualacuity study was from zero separation (or superimposed condition) to an angle of separation of 60 milliradians. For this superimposed condition the pilot could immediately detect the rate of 0.1 milliradian per second. Results of these visual-acuity tests made with angular separations between the target and the nearest background star of 12.5 and 34.0 milliradians are shown in figure 1. This figure shows that at a separation distance of 12.5 milliradians, less than 10 seconds is required for the pilot to detect the desired rate of 0.1 milliradian per second that was set forth in the pilot-control study. As shown in the figure, the ability had deteriorated somewhat at a separation of 34.0 milliradians. This reduced ability indicates that the target must be within 12 to 15 milliradians of a reference star in order that the angular rate of 0.1 milliradian per second be detected. A study of star charts has shown that, on the average, a pilot can expect to have a visible star (sixth magnitude or less) within 2 0 of the target. This means that the pilot may find it necessary to delay the line-of-sight correction until the target is in close proximity to a visible star or, alternatively, to use an optical aid either to superimpose the target on a star or to magnify and make more stars visible. For instance, use of a 3 -inch telescope would permit an average density of 4 to 16 stars (11th magnitude or brighter) per square degree. This density would place the target within 12 milliradians of a star and permit detection of 0.1 milliradian per second. In reference 5 a study was made of a pilot's ability to determine the range and range rate from visual sightings. The analysis and simulation results indicate that it should be possible to determine range and range rate from the angular rate of the target. This technique depends upon development of an optical device which has not yet been demonstrated. Naturally, the first objective in rendezvous is to recognize and track the target. In future missions a beacon will be used for rendezvous practice and possibly for target identification. In order to study the ability of an astronaut to acquire and track a flashing beacon and to measure his night adaptation level throughout the orbit, Langley Research Center (LRC) and Manned Spacecraft Center cooperated in developing such a beacon and included it in the MA -9 flight. The beacon's trajectory prevented its being seen during part of the first orbit after release, but on succeeding orbits Major L. Gordon Cooper periodically compared the beacon intensity with selected stars or with an onboard standard source. These data are still being evaluated.
DOCKING In the visual-docking maneuver the rate of change of size could be used to determine the rate of closure between two vehicles. A study of this problem has been completed and published in reference 7. The study was primarily concerned with vertical descent to the lunar surface, and in the simulation a projection of the lunar surface was servo driven in a closed-loop system for closure cues. The pilot applied a braking thrust to stop the apparent closure velocity; the thrust voltage was fed to an analog computer and then to the landscape-projector drive system. The results may be applied to the visual-contact portion of the docking. Figure 2 (taken from ref. 7) is an example of the results obtained for one observer over the range of visual angles considered. The results, shown as a ratio of closure rate S to distance S, define (for this particular test subject) the boundary of this SIS ratio. This threshold was based on a reply time of 2 seconds dictated by time lags inherent in the test procedure. The figure is of interest because it shows that the maximum perception of closure occurs at visual angles, as subtended by the target outline, from 70 0 to 900 . This boundary agrees with an analytical derivation of the relation between S and S. For this S fell between 0.013 and 0.016. These results can be related to docking; the pilot should be able to judge the closure rate to about 0.15 feet per second from a distance of 10 feet, a value which agrees with results of preliminary visual-docking simulation studies conducted at LRC in which closed-circuit television was used. The results of these studies are discussed in paper no. 9 by Byron M. Jaquet and Donald R. Riley. A test has been conducted inside the U.S. Navy 2,800-foot-long hydrodynamic model basin located at Langley Air Force Base, Va., to see whether the pilot can accurately estimate the separation distance to a target of known size with no cues except the apparent size of the target. After a period of dark adaptation, subjects were asked to estimate the range of several models of known size placed at random distances. The models used are shown in figure 3. They included three disks, a triangle, three cylinders scaled to 1/5, 110, and 120 the size of the Agena target which will be used in the Gemini docking, and a multicolor mylar balloon like the one used in Lieutenant Commander M. Scott Carpenter's orbital flight. Models were painted both flat white and fluorescent orange so that color effects could be investigated. Figure 4 shows the average distance judgments of several observers for various configurations. The solid line in the figure represents perfect estimates. Beyond 500 feet, average estimations were better than expected but with a tendency towards overestimating the range of the large objects and underestimating the range of the smaller target objects. All average estimates (except for the balloon) were fairly accurate for distances from 500,to 0 feet. Although the average estimations for most models were accurate, the individual estimates varied widely, and in many cases the average was accurate because individual estimates "compensated." When the standard deviation of individual estimations was considered, it became apparent that estimates were reasonably accurate (20- to 25-percent error) within ranges from 300 to 500 feet for a receding model, depending on its size, but estimates were accurate only within 61 subject, it was found that a representative value for the closure threshold
the last 50 to 75 feet for an approaching model. An example of the standarddeviation results obtained is shown in figure 5 for the 1-foot disk. True range (separation distance) is plotted against both the average percent error and the standard-deviation error. An average overestimation is indicated by a positive percent error, and an underestimation, by a negative percent error. The points designated by diamonds show estimates made during a test in which the model was moving away (receding) from the observers, and the circles represent estimates made for the approaching disk. At all distances beyond 300 feet the subjects overestimated the range. Within 300 feet the average estimates were close to the true range, which could be expected from figure 4. The shaded region on the standard-deviation plot was drawn primarily to show the trend of the data for the receding disk, but it does include most of the data points within that region. The maximum limit is not exact but was chosen at about 275 feet because additional tests would be necessary to define the curve precisely beyond that point. The standard-deviation error for the approaching disk also has a trend, if not a definite curve. The initial errors started high and then increased as the range became less to about 200 feet, then began to decrease until they were fairly reasonable within the last 50 feet. Individual estimates were apparently made by adding an apparent change in distance to a previous estimate rather than making a completely new estimate at each test distance. Thus, a large initial range error for an approaching model would still appear in the estimate until the object was within 50 to 100 feet of the observer.
Tests indicate that a simple technique utilizing the pilot's visual acuity, such as recognizing a definite shape or spacing between lines, could greatly extend the region of reliable estimates for an approaching object. As stated in paper no. 23 by Howard G. Hatch, Jr., one of the objectives of the six-degree-of-freedom simulator at LRC is to study the effects of target lighting and glare on docking control. In order to do this, a battery of 500watt spotlights has been constructed. These spotlights will simulate sunlight conditions in the docking maneuver. This study is not yet complete.
A series of tests has been made to determine how accurately a navigator could manually record star-transit times for measuring space angles from a rotating space vehicle. It was determined that the variance in timing errors was, in general, inversely proportional to the rate of star travel. Both vehicle rotation and magnification by the viewing instrument influence this rate. Results show that with an apparent rate of rotation of 40 per second and a vehicle rotation of 0.040 per second the average observer tested could measure an angle with a standard-deviation error of 4.3 arc-seconds. The worst observer tested had a standard-deviation error of 4.9 arc-seconds (at the same rates) in measuring a space angle. A study is now in progress to develop a navigational scheme based on this technique.
Another visual-navigation study involved the simple lightweight optical device described in paper no. 12 by Alfred J. Meintel, Jr. This device can be used to provide roll and pitch attitude and altitude, and its accuracy is currently being investigated. CONCLUDING REMARKS Studies in the area of man's visual capabilities in space have been conducted at the Langley Research Center. Results show that there are many areas throughout the Apollo mission in which man has visual-control capabilities. It must be kept in mind that in all cases the optimum use of man and machine is necessary for maximum efficiency and reliability. This integration must be made carefully in order not to hinder the pilot in his prime responsibility of judgment and control.
1. Brissenden, Roy T., Burton, Bert B., Foudriat, Edwin C., and Whitten, James B.: Analog Simulation of a Pilot-Controlled Rendezvous. NASA TN D -747, 1961. 2. Brissenden, Roy F., and Lineberry, Edgar C., Jr.: Visual Control of Rendezvous. Paper No. 62-42, Inst. Aerospace Sci., Jan. 1962. 3. Beasley, Gary P.: Pilot-Controlled Simulation of Rendezvous Between a.Spacecraft and a Commanded Module Having Low Thrust. NASA TN D -1613, 19634. Pennington, Jack E.: Effects of Display Noise on Pilot Control of the Terminal Phase of Space Rendezvous. NASA TN D-1619, 1963. 5. Lineberry, Edgar C., Jr., Brissenden, Roy F., and Kurbjun, Max C.: Analytical and Preliminary Simulation of a Pilot's Ability to Control the Terminal Phase of a Rendezvous With Simple Optical Devices and a Timer. NASA TN D-965, 1961. 6. Brissenden, Roy F.: A Study of Human Pilots' Ability to Detect Angular Motion With Application to Control of Space Rendezvous. NASA TN D- 1+98, 1962. 7. Lina, Lindsay J., and Assadourian, Arthur: Investigation of the Visual'Boundary for Immediate Perception of Vertical Rate of Descent. NASA TN D -1591, 1963-
TYPICAL ANGULAR -RATE PERCEPTION
188.8.131.52.0RATE OF OBJECT MOTION MILLIRADIANS/SEC .8.6.4 .2k 0 5 10 15 20 TIME TO IDENTIFY OBJECT MOTION, SEC DIRECTION OF MOTION o RIGHT 0 RIGHT d LEFT LEFT INITIAL SEPARATION DISTANCE, MILLI RADIANS 12.5 34.0 12.5 34.0
THRESHOLD OF VELOCITY PERCEPTION
.03 S/S FPS/FT .02
o INCREASING ANGLE q DECREASING ANGLE Ot
Figure 2 64
L-62-5658 Figure 3 DISTANCE JUDGMENT
16 X 102 ° ° o q °
00 q °
0 q q °
12 ACTUAL RANGE, FT
° 12-IN. DISK 0 9.25-IN. DISK ° 6-IN.DISK 0 3.35-IN. DISK ® 30-IN.MERCURY BALLOON
q ° ° q m 4 q®
ESTIMATED RANGE, FT
Figure 4 65
O o Disk receding Disk approaching
I - FOOT
^ f0 L
20 O 0
0 O O 0
O -20 -40
80 0 70
0 ^^ O 0 O
10 O 0 100 200 300 400 500 600 700 800 900 1,000 1,100 1,200 1,300
True range, feet Figure 5
FIXED-BASE GEMINI-AGENA DOCKING SIMULATION By Byron M. Jaquet and Donald R. Riley
Preliminary results of a fixed-base analog simulation study of the GeminiAgena docking indicate that a pilot can achieve successful docking of the Gemini with the Agena when using the rate-command attitude mode, which is the proposed primary docking mode. With the direct attitude mode, it appears that successful docking can be achieved although considerably more training will be required than with the rate-command mode. Successful docking can be achieved under various lighting conditions but further study is required for docking on the dark side of the earth.
In order to insure the success of the lunar-orbit rendezvous concept in the exploration of the moon, it is necessary to develop techniques for rendezvous and docking. Project Gemini has been undertaken to provide experience and knowledge in rendezvous and docking procedures, long-duration space flight, maneuvering performance, extravehicular operations, and near and distant observation of a satellite under various lighting conditions. This information will be applicable to other missions in which the rendezvous concept is used. After the rendezvous phase, during which the Gemini spacecraft has been maneuvered to a range relatively close to the Agena, the astronauts will maneuver the spacecraft by using visual cues obtained from observation of the attitude-stabilized Agena through the spacecraft windows. Both the Gemini spacecraft and the Agena vehicle are then in approximately a 161- nautical-mile circular orbit about the earth. This paper presents preliminary results of a Gemini-Agena visual-docking study which is currently in progress. The study employs a visual-docking simulator of the fixed-base type.
DESCRIPTION OF SPACECRAFT
Figure 1 depicts the two vehicles in the range where the Gemini spacecraft can be maneuvered by using visual cues obtained from observation of the Agena through the spacecraft windows. The Gemini spacecraft consists of the reentry vehicle and a maneuvering unit. Located in the maneuvering unit are eight attitude-control engines and eight translation-control engines, all of which use hypergolic fuels. All engines are located behind the center of gravity, which is located between the heat shield and the astronauts. The heat shield is located at the section where the maneuvering unit is joined to the reentry vehicle. Because of the rearward location of the engines with respect to the center of gravity of the spacecraft, severe coupling occurs between vertical
and lateral translation-control inputs and the pitch and yaw spacecraft motions. This effect is discussed subsequently in the present paper. For docking, the Agena has a 5- foot-diameter shock-mounted ring on the front which serves to channel the Gemini nose to the Agena coupler for latching and rigidizing the two vehicles. The V-shaped slot in the docking ring and the indexing bar on the Gemini provide the necessary roll positioning for the latching mechanism. For the simulation, the Gemini spacecraft inertias and weight corresponded to a one-half fuel load condition. Table I presents pertinent Gemini information used in the simulation.
DESCRIPTION OF SIMULATOR
General Arrangement Figure 2 is an artist's sketch of the visual-docking simulator. The simulator consists of analog-computer equipment combined with a USAF F -151 gunnery trainer which has been adapted for the present study. Included in the gunnery trainer was a closed-circuit-television pickup and projection system. Complete six-degree-of-freedom motion is obtained from three angular degrees of freedom of a small-scale model of the Agena, translation of the Agena model in front of the TV pickup camera, and the elevation and azimuth motion of a two-axis mirror located directly over the pilot's head. A 20-foot-diameter, spherical, projection screen houses a full-size two-man wooden mock-up of the Gemini spacecraft. Computer equipment associated with the gunnery trainer determines the position of the Agena model on the range bed, the proper aspect of the model, and the proper azimuth and elevation angles of the mirror. The analog computer solves the six-degree-of-freedom equations of relative motion.
Mock-Up, Control Systems, and Hand Controllers The internal arrangement of the Gemini mock-up with a view of the Agena through the window is shown in figure 3. Also shown are the basic flight instruments, which include a Project Mercury angular rate and attitude instrument, a relative-range instrument, and a range-rate instrument. The instruments are used for pilot training and familiarization and are not used on data-taking flights. The pilot has his right hand on the attitude controller and his left hand on the translation controller. Both controllers are manipulated with the fingers. In the operational Gemini spacecraft, current plans call for controllers of the fullhand-grip type. For translation control, the pilot uses an on-off type of controller and can maneuver the Gemini fore and aft, up and down, and left and right by corresponding controller deflections. Full thrust is provided when the controller deflection exceeds a deadband. For attitude control, the pilot manipulates a three-axis hand controller and has the option of selecting one of three attitude-control modes: (1) rate command, providing an angular rate, above a 0.1 0 /sec deadband, which is proportional 68
to controller deflection; (2) direct, providing full angular acceleration; and (3) pulse, in which a small impulse is generated to produce a small angular acceleration. The three attitude modes become activated when the controller deflection exceeds a deadband. The rate-command mode presently provides maximum rates of 80 /sec in roll and 40 /sec in pitch and yaw. SCOPE OF INVESTIGATION
Utilizing the visual-docking simulator, the capabilities of the pilot in performing Gemini-Agena docking are currently being investigated for various initial conditions from a maximum range of about 325 feet up to the point of contact. Fuel requirements, the time required to complete docking, contact velocities, and vehicle attitudes and positions are of primary concern in this phase of the study. The effects of the various attitude-control modes, targetlighting conditions, different docking approaches, visual aids, and controlsystem failures are also being investigated.
RESULTS AND DISCUSSION
The data presented in the following section were obtained with the ratecommand attitude mode with one NASA research pilot and two engineers as the simulator pilots. The Gemini initially is located 250 feet behind, 100 feet below, and 75 feet to the right of the Agena. When closing maneuvers are initiated, the Gemini approaches the end of the Agena on which the docking ring is mounted. For the data presented herein, initial relative velocities are zero, but they are being varied in the study. The Agena is assumed to be perfectly stabilized with respect to the local vertical and is fully lighted along the line of sight. Other lighting conditions and the direct attitude mode are discussed in subsequent sections of this paper.
Fully Lighted Agena
Rate-command attitude mode.- In figure 4 the front and top views of the docking ring are shown. In the top view is shown the contact plane in which the docking flights are terminated, after which the computer is interrogated for the various items of interest. Acceptable end conditions require that the indexing bar on the Gemini be alined with the V-slot in the docking ring as projected on the screen by the TV projection system. Plotted in the figure are points for the three simulator pilots. These points represent the vertical and lateral displacements of the center line of the Gemini with respect to the center of the docking ring of the Agena. Also shown is the design tolerance for these displacements, a circle of 12 -inch radius. Out of a total of 69 simulator flights, 94 percent were within the design tolerance. Note that 49 flights were in the shaded area. The research pilot had only one flight outside of the design tolerance. With more training, all flights should be within the tolerance.
The design tolerance for roll, pitch, and yaw angles at contact is ±100. Except for one flight by each of the three simulator pilots, in which yaw angle was greater than -10 0 , all flights were within the design tolerance. The lateral and vertical contact velocities are shown in figure 5. All flights, except one by the research pilot, were within the design tolerance of 0.5 ft/sec. Note that 42 flights were within the shaded area. With the ratecommand attitude mode, the Gemini angular rates at contact were within or near the 0.1 0 /sec deadband of the control system with a few exceptions where an attitude-control input was applied at or just prior to the instant of termination of a flight. As noted previously, the maximum commanded rates were 80/sec in roll and 40 /sec in pitch and yaw. During a typical docking flight, the pilot generally used about 40 /sec to 50 /sec in roll and 2 0 /sec to 30 /sec in pitch and yaw. The longitudinal contact velocities are summarized in figure 6 in the form of a bar graph indicating the number of flights within various velocity ranges for each of the simulator pilots. None of the flights had contact velocities greater than 1.0 ft/sec, the design tolerance being 1.5 ft/sec. It is interesting to note that no specific instructions regarding low contact velocities were given to the simulator pilots. However, they knew the design limit and that a closing velocity should exist in order that latching of the two vehicles could be completed. The total amount of fuel used, including attitude-control and translationcontrol fuel, is presented in figure 7 as a function of flight time. The NASA research pilot and the two engineers used essentially the same amount of fuel. The trained research pilot, however, flew more consistently and in less time than the engineers. The research pilot took about 2 to 5 minutes to complete a docking flight. The three flights by the research pilot with fuel above30 pounds were cases in which he realized that he would not dock properly and he backed off to try again. Except for the six cases by the research pilot (shown by the.filled-in circular symbol), no special effort was made to conserve fuel. Also, no specific time allotment was given to complete a docking run. In figure 8, the translation fuel and the attitude fuel used are shown as a function of flight time. In the research pilot's attempts for minimum fuel usage, he effected a greater saving in translation fuel than in attitude fuel.
Direct attitude mode.- A number of flights have been made with the direct attitude-control mode by the research pilot and two engineers. All three simulator pilots had a number of hours of practice flights prior to the data-taking flights. These were made with the Agena attitude stabilized with respect to the local vertical and fully lighted along the line of sight. Using the direct attitude mode, the piloting task was considerably more difficult than that with the rate-command mode because of the strong coupling between vertical and lateral translation-control inputs into the pitch and yaw motions. This coupling is caused by the rearward engine locations with respect to the spacecraft center of gravity. With the direct attitude mode, it appears that a longer training period is required to attain a high percentage of flights within the tolerances. The rate-command mode incorporates a feedback loop which activates attitude engines that reduce the angular rates caused by translation engine operation to 70
the 0.10 /sec deadband. In order to determine whether the difficulty in flying with the direct mode was actually due to the coupling, several direct-mode flights were made with an uncoupled system. For these flights, the translation engines fired through the center of gravity and the moments were produced in couples about the center of gravity. With this control configuration, the piloting task was much easier. Although redesign of the Gemini is not suggested, this means is used to indicate the source of difficulty in flying with the direct attitude mode.
Use of Visual Aids With Direct Attitude Mode Astronauts Schirra and Grissom made several docking flights using the ratecommand attitude mode. Astronaut Schirra stated that, when nearly alined, he could not see beyond the edge of the docking ring when at relatively close ranges and that he would like to know where the rear of the Agena was located. Because of this situation, it is difficult to determine attitudes at close range. In order to provide a means of "seeing" the after end of the Agena, the visual aids shown in figure 9 have been tried. Two-inch-diameter posts extending 30 inches above and to the left of the line of sight were mounted on the rear of the cylinder on the Agena. Also shown in figure 9 is a grid, indicating lateral and vertical tolerances, which is located on the coupling face within the Agena. The grid helps to accentuate the inner circle of the docking ring and has been used on all flights. With the posts located on the Agena, it was found that flying with the direct attitude mode was easier and required less time and that consistent flights within the design tolerances would be made. The research pilot felt that the visual aids improved the operation of the direct mode but that it still was not as easy to fly with as the rate-command mode without visual aids.
Agena Lighting Conditions A few flights condition shown in Agena which are in successful docking have been made with the rate-command mode for an Agena lighting figure 10. All areas which are shaded represent regions of the shadows because of the position of the sun. Consistent and could be made under this lighting condition.
Another lighting condition, corresponding to docking on the dark side of the earth, has been investigated briefly. For this condition, running lights were located on the Agena. Three were located within the rim of the docking ring. Two of these were located near the peaks of the V-slot and the other was at the bottom edge of the vertical center line. Four lights were located on the end of the Agena cylinder tangent to the surface — two vertically and two horizontally. The docking-ring face was illuminated slightly so that the outline was just visible. With this arrangement, flying was more difficult than with the previous lighting conditions, but docking could be accomplished. One difficulty encountered was the lack of attitude information available from observation of the target when the bottom light on the docking ring was obscured by the Gemini nose. More study is required for this lighting condition. 71
With the rate-command attitude mode, which is the proposed primary docking mode for the operational spacecraft, successful and consistent docking can be achieved between the Gemini spacecraft and the Agena vehicle. With the direct attitude mode, without visual aids on the Agena, fewer flights were within the design tolerances. With visual aids on the Agena, flying with the direct mode becomes easier. Also, docking can be accomplished under various Agena lighting conditions, although more study is required for docking on the dark side of the earth.
TABLE I GEMINI CONFIGURATION USED IN SIMULATION Thrust along longitudinal axis, lb . . . . . . . . . . . . . Thrust along lateral axis, lb . . . . . . . . . . . . . . . Thrust along vertical axis, lb . . . . . . . . . . . . . . . Rolling moment about center of gravity, ft-lb . . . . . . . Pitching moment about center of gravity, ft-lb . . . . . . . Yawing moment about center of gravity, ft-lb . . . . . . . . Weight, lb . . . . . . . . . . . . . . . . . . . . . . . . . Center-of-gravity location, full-scale station, in. . . . . Distance from center of gravity to roll nozzles, ft . . . . Distance from center of gravity to pitch and yaw nozzles, ft Specific impulse of fuel, sec . . . . . . . . . . . . . . . Moments of inertia, slug-ft2: Roll . . . . . . . . . . . . . . . . . . . . . . . . . . . Pitch . . . . . . . . . . . . . . . . . . . . . . . . . . Yaw . . . . . . . . . . . . . . . . . . . . . . . . Products of inertia, slug-ft2: Pitch-yaw . . . . . . . . . . . . . . . . . . . . . . . . Roll-pitch . . . . . . . . . . . . . . . . . . . . . . . . Roll-yaw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . +200, . . . . . . . . . . . . . . . . . . . -187.94 ±99.62 ±100 ±186.88 ±397.30 ±397.30 6,623 115.35 3.738 7.946 320 1,270.0 4, 062.5 4, 066.0 -12.8 44.2 90.8
. . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
^^ w °° ^ ° o °* ~°~ ^ ° ^^ ^ °
° ~° ^" ° °^° ^ ^° m o° ° p °° ^ °p °°° r ° °^ ~ °~ ^ ^ ° ° °
INTERNAL ARRANGEMENT OF GEMINI MOCP -UP
VERTICAL AND LATERAL DISPLACEMENTS AT CONTACT INITIAL RELATIVE VELOCITIES, &
TOTAL NUMBER OF FLIGHTS, 69
» RESEARCH PILOT o ENGINEER A ENGINEER D
)ES I GN TOLERANCE
m VERTICAL DISPLACEMENT, 0 |N. 0 0 30^ -30 -20 - 0 0 0 20 0 LATERAL DISPLACEMENT, |Ki
0 26 1 01
VERTICAL AND LATERAL VELOCITIES AT CONTACT
INITIAL RELATIVE VELOCITIES, 0;
TOTAL NUMBER OF FLIGHTS, 69
o RESEARCH PILOT
q ENGINEER A
O ENGINEER B
-.6 -.4 -•2 VERTICAL
VELOCITY, 0 .2
DES I G TOLERANCE 83 PERCENT OF FLIGHTS WITHIN ^ 0.2 FT/SEC
\ I (016 C3 O ^\ q O / O 20 42 -.6 -.4 -.2 0 .4 .2 .6 LATERAL VELOCITY, FT/SEC
LONGITUDINAL VELOCITY AT CONTACT INITIAL RELATIVE VELOCITIES, 0; TOTAL NUMBER OF FLIGHTS, 69 1.5
DESIGN TOLERANCE ® RESEARCH PILOT
® ENGINEER A M ENGINEER B PERCENT OF FLIGHTS ;.^
CONTACT VELOCITY, FT/SEC
'° 5 ^,.
NUMBER OF FLIGHTS
VARIATION OF TOTAL AMOUNT OF FUEL USED WITH FLIGHT TIME INITIAL RELATIVE VELOCITIES, 0
30 0 0 q
OF FUEL USED, LB 10
aO p 00q O q 0 0 ap 0 oRESEARCH PILOT b *RESEARCH PILOT, ATTEMPTS Oi $•p FOR MIN. FUEL a • p p O q ENGINEER A *ENGINEER B 2 4 6 FLIGHT TIME, MIN Figure 7
VARIATION OF TRANSLATION AND ATTITUDE FUEL WITH
INITIAL RELATIVE VELOCITIES, 0 40 0
TRANSLATION FUEL, LB
30 2010 0
0 00 0 0 ^ q
p O^O O O©p 00
q q^,,^ •p0 OO
0 RESEARCH PILOT
• RESEARCH PILOT, ATTEMPTS FORM] N. FUEL q ENGINEER A
O ENGINEER B
ATTITUDE FUEL, LB
10 0 1 2 3
7 6 4 5 FLIGHT TIME, MIN
^1 1^ I
O O O ppjp O O 8 9 10
VISUAL AIDS ON AGENA
DIAM. POSTS LOCATED - END OF CYLINDER
3.7" BEHIND FRONT .ANE OF DOCK I NG R I NG
GR I D ON COU PLI NG FACE
LARGE SQUARES 1 REPRESENT ±121 TOLERANCES IN CONTACT
D I S PLACEMENTS
AGENA LIGHTING CONDITION
^` xP o ' ^ °¢ y ,^.:>7^^ ^ i.
T ^;^ ^¢^
10. THREE-DEGREE-OF-FREEDOM FIXED-BASE SIMULATION OF PILOT-CONTROLLED LUNAR TRAJECTORIES FROM LIFT-OFF TO RENDEZVOUS By Charles P. Llewellyn
A three-degree-of-freedom simulation of pilot-controlled lunar trajectories from lift-off to rendezvous with an orbiting space station has been made. The results of this planar study have shown that a pilot can visually determine his launch time and effectively manually control both vehicle attitude and mainengine cut-off to arrive at the proper altitude and position to successfully initiate and complete a rendezvous maneuver. It has also been shown through the use of three trajectories, having coast angles of 24 0 , 90 0 , and 180 0 , and a loiter maneuver that a launch window of as much as 5 minutes may be realized.
The purpose of this simulator study is to investigate the ability of a man to perform a successful and reasonably efficient lunar take-off and rendezvous given a few simple instruments and utilizing certain visual cues. It is also the aim of this study to find a means whereby the launch window may be increased and the problem of on-time launch may be minimized. In general, the pilot's task is to visually determine his time of launch, manually control his pitch attitude to follow a predetermined launch trajectory from lift-off to space-station altitude, and rendezvous with the orbiting station. The primary displays used by the pilot are vehicle angular pitch rate, time, vehicle altitude above the lunar surface, and station range and range rate.
Some of the equipment used in the simulation may be seen in figure 1. This figure is an artist's sketch showing the screen, cockpit, and some of the equipment such as the star-field and lunar-horizon projector and the satellite projector. A stereo tape recorder to simulate noise of the main engine and the reaction motor was also used. Shown in figure 2 is a view of the lunar horizon, star field, and satellite as seen by the pilot. The instrument panel and pilot can be seen in the foreground.
PROCEDURES As mentioned previously, the pilot's task is to launch from the lunar surface and rendezvous with an orbiting space station. This task can be divided into three phases as shown in figure 3: the boost phase, the coast phase, and the rendezvous phase. In the lift-off or boost phase, the pilot visually determines his lift-off time by observing the position of the station against the star-field background or by determining the station elevation E above the lunar horizon with the aid of a simple device such as an etching on the cockpit canopy or some hand-held device. After lift-off, the pilot controls his attitude with the reaction control jets and attempts to follow a predetermined pitch program by pitching the vehicle at two constant pitch rates. An acceleration command system was used. The pilot, in addition to following the pitch-rate program, monitors altitude and time to be sure he is on the nominal ascent trajectory. At a particular time in the boost phase the pilot commands main-engine cut-off. The time for main-engine cut-off and the pitch-rate program are chosen on the basis of the desired coast angle from booster burnout (20 miles) to apogee of the ascent trajectory, or station altitude (100 miles). The approach taken in this study to increase the launch window was to use trajectories involving different coast times from booster burnout to apogee of the ascent trajectory. Three different trajectories requiring three different pitch programs and three different cut-off times were chosen. These trajectories correspond to coast angles, or transfer angles, from booster burnout to apogee of 240 , 900 , and 1800 . The pilot was required to perform each of the pilot programs from memory. For each of these trajectories, there is an associated nominal launch time which will result in an interception of the station at apogee of the ascent trajectory. Of these three trajectories, the earliest launch time is that time associated with the 240 transfer, and the latest is the time associated with the 1800 transfer. The time difference is about 4 minutes. Launch time for the 90 0 transfer occurs about 2 minutes later than that for the 240 trajectory. The relative position of the station and launch vehicle for each of these times is indicated in figure 3. The procedure, then, is to have the pilot launch the vehicle at the time associated with the 240 transfer trajectory. If for some reason he is delayed, he may wait about 2 minutes and use the 90 0 transfer trajectory, or if he is delayed longer, he can wait about 2 more minutes and use the 180 0 trajectory. Thus, the pilot has at his disposal, with the use of these three different trajectories, a leeway of about 4 minutes of launch window. During the coast phase, the pilot must monitor the time history of his altitude and determine any gross deviation from the nominal of this parameter. Since the coast phases range from about 10 minutes to about an hour, depending upon the transfer chosen, the pilot has ample time in which to make any corrections. In general, however, corrections are made only during the last few minutes of the coast phase.
During the rendezvous phase of the mission, the pilot uses visual cues from the station and star field and, primarily, the range and range-rate instrument displays to initiate and complete the maneuver. Certain of the techniques used during this phase are dependent upon the particular trajectory the pilot is lying principally because of the difference in closing velocity at apogee for the three different ascent trajectories. This closing velocity ranges from about 1, 400 ft/sec for the 24 0 transfer to about 106 ft/sec for the 180 0 transfer. For the purpose of this study, the mission was terminated when the ferry vehicle was within 3 miles of the orbiting station with a closing velocity of about 10 ft/sec.
Prior to discussing some of the results of this simulator study, some comments on the difficulty of the pilot's task are necessary. It was found that with only a brief period of training the pilot could easily accomplish the mission even with low-precision instruments. For example, altitude could be read to an accuracy of only about 3,000 feet and angular pitch rate to a little better than 0.010 per second. The pilot experienced no difficulty in manually controlling the attitude of the vehicle and following fairly closely any one of the three nominal trajectories. Print-out data of the important trajectory parameters of velocity, altitude, and flight-path angle at booster burnout indicate some deviation from the nominal of these quantities; however, these deviations were small enough that the pilot was able to bring his vehicle to proper altitude and position to successfully initiate and complete a rendezvous maneuver. For the 90 0 trajectory, some runs were made where off-nominal thrust conditions of f2 percent were simulated. The pilot had an indication of this condition either from a thrust-level indicator or an axial accelerometer. He compensated for this off-nominal condition by adjusting his cut-off time by the same percentage his thrust was off. It was found that this off-nominal condition had little effect on the pilot's ability to efficiently carry out the mission.
Shown in figure 4 are the results of a few typical runs. The ratio of total fuel used in the simulation to the fuel required to perform a perfectly executed mission is shown plotted against increment in launch time from the nominal ontime launch of the 24 0 transfer trajectory. These points are indicated by the circles. Again, it can be seen that there is a margin in launch time of about 2 minutes between the 24 0 and 90 0 trajectories and about 4 minutes between the 240 and 1800 trajectories. As can be seen from the figure, the pilot was able to perform the mission without appreciable fuel cost. The highest points for each transfer represent the earliest runs, and the trend downward gives some indication of the pilot's learning ability. The relatively few points for the 1800 transfer is due to the fact that this particular trajectory has only begun to be f 81
avestigated. It is expected that with experience the pilot will be able to nprove these results. Also included in the figure and indicated by the crosses are some typical esults where the pilot chose to launch early as shown by the negative time scale. a the runs where the pilot launched early in order to minimize the problem of onime launch, he followed the same nominal pitch program, but when he arrived at pogee of the ascent trajectory, he went into a loiter maneuver. During this aneuver he orientates his vehicle, using the station's position as a cue, and :crusts vertically with short bursts of his main engine to maintain his altitude, ,king advantage of his closing velocity to minimize the time required for renezvous. As can be seen in figure 4, this procedure results in a safety factor C at least 1 minute from on-time launches at little expense in the fuel cost. Dr clarity, the early launches are shown only for the 240 transfer. Preliminary esults from the 90 0 and 1800 transfers indicate a similar trend, but because of ery low closing velocities, this increment in launch time may not be as much as minute from the time of on-time launches for these particular trajectories.
In conclusion, the results of this planar study have shown that a pilot can isually determine his launch time to within a few seconds, effectively manually Dntrol vehicle attitude to main-engine cut-off, and arrive at the proper altiade in position to successfully initiate and complete a rendezvous maneuver. It is also been shown through the use of three trajectories and a loiter maneuver nat a launch window of as much as 5 minutes can be realized.
LUNAR TAKE-OFF AND RENDEZVOUS SIMULATOR
L-63-1500 Figure 2 83
°° ^"^ ^ =° ° ^^ °~ ~^ ^° = °°^ ~ °°° ~ ° ^
~ - ~w * ~° ~°^ °°^ °°
LUNAR TAKE — OFF AND RENDEZVOUS T RAJECTORY 4
^o 1mr ~
l BOOST PHASE l COAST PHASE l RENDEZVOUS PHASE Figure 3
FUEL RATIO PLOTTED AGAINST INCREMENT IN
^^ 1.06"~ «ON-T|MELAUNCH +[ARLY LAUNCH C24-TRANSFER)
FUEL RATIO ) -
f t + 8 + + o + +o ^ R n 4
2 O INCREMENT |N LAUNCH TIME, MIN
11. MANUAL CONTROL OF A LUNAR LAUNCH By Lindsay J. Lina SUMMARY A preliminary fixed-base simulation study was made to investigate a pilot's capability in manually controlling a launch from the moon's surface to a safe lunar orbit with a minimum of instrumentation. In this study the data displayed to the pilot and his controls were purposely degraded in accuracy to simulate simple backup instruments and control. The results of the study indicate that, by following a predetermined pitch program and by making corrections in pitch angle near the burnout of a constantthrust engine, the pilot control technique was adequate to achieve a safe chasing orbit at an altitude near 50,000 feet. INTRODUCTION This investigation is part of a general research approach to determine simplified manual piloting techniques and minimum instrumentation requirements for backup of the primary guidance system of a lunar excursion module (LEM). This preliminary study was made to determine these requirements with the assumption that the pilot is manually controlling a takeoff from the lunar surface or from a hovering condition and desires to reach a safe chasing orbit at an altitude near 50, 000 feet. DESCRIPTION OF THE SIMULATION The control panel for the fixed-base simulation is illustrated in figure 1. Vehicle pitch attitude was displayed on the indicator and was varied by the control knob shown in figure 1. In all cases of control, the pilot could not adjust the attitude angle more accurately than ±2 0; this inaccuracy could be used to simulate errors due to sensing accuracy, pilot control, and the omission of vehicle dynamics. Altitude was displayed on the control panel by a digital voltmeter to the nearest 1,000 feet to approximate the accuracy that may be possible with a simple optical backup instrument, such as the one described in paper no. 12. The timer shown in figure 1 was used to aid the pilot in following a predescribed pitch program, which was displayed to the pilot on a placard mounted near the controller, and to indicate the approach of, and the time of, engine 85
cutoff. Engine cutoff was made with the switch shown on the left of figure 1. The timer used in the simulation could be replaced in an actual vehicle by a bodyfixed integrating accelerometer mounted on the thrust axis. An analog computer was used to solve the equations of motion for the verticalplane maneuver used. A continuous-thrust engine was assumed with an initial thrust-to-weight ratio of 0.4 and a specific impulse of 315 seconds. PILOT CONTROL TECHNIQUE The pilot control technique used in this study was to have the pilot follow a predescribed pitch-time program from takeoff to near cutoff of the constantthrust engine. Prior to engine-cutoff time, the pilot would monitor altitude and adjust attitude to hold altitude as nearly constant as possible with the degraded altitude display. This procedure was followed in order to hold the flight-path angle to near zero at burnout. Actual altitude at burnout was not considered critical, and the pilot was allowed ±10,000 feet from the nominal altitude. Burnout time was selected to give an excess velocity of 100 fps over circular-orbital speed in order to permit the required maneuvering near burnout and still achieve excess velocity over circular-orbital speed for a safe insertion condition. Such a burnout would allow sufficient time to monitor the orbital characteristics and to apply small velocity corrections, if required later. PREDETERMINED PITCH PROGRAM The pitch program that was used by the pilot as a guide for the launch maneuver is shown in figure 2. This three-step pitch program allows manual control with simple attitude changes. To follow the program, the lunar excursion module takes off from the 90 0 resting attitude by thrusting vertically. In an actual launch, the resting attitude might be slightly off from the vertical, but the pilot would attain the vertical at lift-off and apply vertical thrust for 8 seconds or until nearby obstacles are cleared. The pilot then pitches down to 550 from the local horizontal. At 85 seconds he again pitches down to 50; that is, with a small upward component of thrust. The optical device described in paper no. 12 would also be useful in judging vehicle-attitude angles for the pitch program to be followed. Although the nominal trajectory should be on a path to orbit at 50,000 feet, the requirements for the simulation allowed the pilot a deviation of ±10,000 feet of altitude. However, the altitude rate must be brought to nearly zero at the time preset for burnout. RESULTS AND DISCUSSION Some results of the launch simulations are shown in figures 3 and 4. These plots show the variations of velocity with altitude. Although the plots are not
trajectories, the flight-path angle at orbital injection is indicated in the proper direction by the slope of the line to an exaggerated degree. The practice launches shown in figure 3 were made by an inexperienced pilot who did not try to level out at an altitude of 50,000 feet nor did he try for a final altitude closer than about ±10,000 feet. He did, however, make an effort to end with zero vertical velocity and, thus, zero flight-path angle. Except for his first run which has a downward slope (labeled "Run 1" in fig. 3), all his trials ended within ±120 of the horizontal. With the pilot initiating engine cutoff with a timer, the runs ended with a ±50 fps error in the desired velocity. This error includes the following: (a) Pilot reaction and timing error in initiating engine cutoff (b) Errors produced by the pilot control attitude deviations from the nominal 50 pitch angle for insertion corrections (c) Errors due to deviations from the nominal altitude (d) Errors due to the analog computer drift The use of a body-fixed integrating accelerometer, with an early cutoff of the high-thrust engine followed by a small thrust input, such as applying the thrust of the reaction motors, would produce less error in item (a) and would permit the pilot to adjust for (c). The introduced accelerometer instrument error would also be less than (d). It should also be noted that the body-fixed integrating accelerometer would automatically account for any variation of engine thrust during the launch. The errors produced by (b) are due to the technique used. An analysis of data shows that this error contributes approximately ±10 fps to the total insertion speed. An experienced pilot made the simulated launches shown in figure 4. His first run (labeled in fig. 4) was the most inaccurate ' ending at 0.90 upward. The remainder of his launch trials ended at less than 0.25 0, in the upward direction. The runs all ended at an altitude near 50,000 feet. A typical time history of one of the launches is shown in figure 5. This figure shows the pilot's command inputs of vehicle pitch angle and how he relates them to what he sees on the altimeter. At the top of the figure is shown the actual altitude as recorded in the simulation; in the center of the figure is altitude as displayed to the pilot in 1,000-foot intervals. The scale has been expanded to give a better indication of the altitude display near the orbital altitude. In addition to seeing altitude displayed . digitally, the pilot has a sense of the time interval between the altitude jumps. The pilot's selections of pitch-attitude angle are shown at the bottom of figure 5. It may be seen that the pilot follows the pitch program by timing the steps in a mechanical fashion. At about 21 minutes after the start of the launch, the pilot no longer follows the program and begins to use his own judgment. Realizing that he is rapidly approaching the orbital altitude of 50,000 feet, he Omb
selects a reduced attitude angle and awaits further developments. When leveling out becomes apparent by the reduced rate of altitude change, the pilot anticipates loss of height and increases the attitude angle in two steps. Then, as the altimeter actually indicates a descent, he increases the pitch angle again. Now, as the altitude rate is nearly zero, he reduces attitude angle to avoid climbing, but then increases attitude angle and then decreases it. The pilot becomes aware of the approach of the end of the run as his changes in attitude angle become more frequent. If the final flight-path angle were allowed to be always biased upward, an increased period for adjustment to avoid hitting the moon would be available. CONCLUDING REMARKS The results of simple simulations of pilot control of launch from the moon's surface to a chasing orbit at an altitude near 50,000 feet indicate the feasibility of a simplified launch technique. The technique utilizes a pitch-control program and requires the pilot to adjust the pitch-attitude angle of the vehicle to bring the altitude rate to nearly zero as orbital velocity is approached. Simple instruments not dependent on the primary guidance system may be adequate for use. Instruments that are being considered for this use are a bodyfixed integrating accelerometer, a timer, and a simple optical device. The simulated launches indicate that, with practice, a pilot may be able to attain orbit with deviations of ±50 feet per second in velocity and ±1/20 in flight-path angle from the nominal values. These errors do not appear excessive, when it is considered that final errors in launch can be corrected after the launch has been nominally completed. With restart capability of the launch rocket motor, or perhaps even with properly directed thrusting of the reaction-control jets, corrections can be made in the long period of coasting that follows the launch. An increased period for adjustment to avoid hitting the moon would be available if the final flight-path angle were always biased toward the upward direction.
THREE - STEP PITCH PROGRAM FOR LUNAR LAUNCH
PITCH ATTITUDE, DEG 60
60 120 180 240 300 360
PRACTICE LAUNCHES WITH PILOT CONTROL
60 RUN I 50 40 ALTITUDE, FT 30 20 10 0 1 2 3
6 x 10'
LUNAR LAUNCHES WITH PILOT CONTROL ..
50 40 ALTITUDE, FT 30
10 0 1 2 3 4 VELOCITY, FPS 5 6x 103
TIME HISTORY OF LUNAR LAUNCH
ACTUAL 50x 103 ALTITUDE, 25 FT x 103 INDICATED 48 ALTITUDE, 46 FT 44 90 PITCH ATTITUDE, 60 DEG 30 0 60 120 240 180 TIME, SEC —— 0
Page intentionally left blank
F o . f f^FF?SY,.^^
12® AN OPTICAL DEVICE FOR OBTAINING ATTITUDE AND ALTITUDE By Alfred J. Meintel, Jr. SUMMARY The concept of a simple optical device which could be used as a backup to onboard instrumentation on the lunar excursion module (LEM) is described. The instrument presents vehicle altitude and attitude information on a screen which could be located on the vehicle instrument panel. A mockup of the instrument has been constructed and static accuracies are presented. INTRODUCTION The successful accomplishment of launch and abort trajectories is dependent on information supplied by the onboard instrumentation. It is proposed that a manual backup to this instrumentation system could be supplied by the use of an uncomplicated optical device. The principle of the instrument is to present right and left peripheral information on a screen which could be located on the vehicle instrument panel. When used for the lunar mission, the images on the screen would be of the right and left lunar horizons. The position of the horizon images with respect to a screen reticle would represent attitude and altitude with respect to a local vertical. The instrument appears suited as a backup system for determination of local vertical with respect to the moon or earth by using the horizons as a reference. The instrument could be used as an attitude reference to orient the vehicle to a plane in space by using known stars as a reference. The use of the instrument as applied to the lunar mission is as a backup LEM altitude and attitude reference in performing a lunar launch as described in paper no. 11.
change in altitude pitch angle roll angle yaw angle
DESIGN AND OPERATION
A mockup of the simple optical device is shown in figure 1. The instrument acts as two cameras with ground-glass viewers located with their lenses directed 1800 to each other. The components and operation of the instrument are shown in figure 2. The lenses are located at a distance from the ground-glass screen equal to their focal length. This arrangement allows all objects at a distance of 100 feet or more from the lenses to be imaged on the screen. The mirrors are introduced between the lenses to redirect the light rays 90 0 so that the images are focused on the screen. In this manner, both right and left images can be viewed side by side for comparison. If the lenses and mirror combination are located in the vehicle, then for any vehicle motion there is a corresponding change in the position of the horizon images on the screen. The manner in which the images move for a certain vehicle motion is dependent on the orientation of the instrument in the vehicle. For a lunar mission, the instrument should be oriented so that the right and left lunar horizons are imaged on the screen because these should offer high contrast on both the earthlit and the sunlit sides of the moon. Also, with this orientation, chances of imaging the sun when moving from the earthlit side to the sunlit side of the moon are remote. Therefore, in this paper, the instrument is assumed to be positioned in the vehicle so that the pilot in a normal aircraft would view the screen in the instrument panel while looking along the thrust axis of the vehicle through the forward window. However, for the LEM, which seats the pilot at 900 to the thrust axis, the screen view must be optically rotated 900. For the normal aircraft instrument orientation, the screen views for various vehicle maneuvers are shown in figure 3. Altitude is shown as a displacement of the imaged horizons from a zero line. The displacement of the images is directly related to the depression angle of the horizons with respect to the center line of the lenses. The actual amount of screen displacement is equal to the tangent of the horizon-depression angle times the focal length of the lenses used. The depression angle can be determined from the geometric relationship between altitude and horizon-depression angle for the moon. Therefore, a suitable reticle may be superimposed on the screen and altitude may be read directly. A roll maneuver is shown by a separation between the two horizon images in figure 3. This separation represents twice the vehicle roll angle and is due to the fact that as the vehicle rolls, one horizon is above the center line of the lenses and the other is below the center line by the same angle. The amount of separation is equal to twice the tangent of the vehicle roll angle times the focal length of the lenses. The screen view for a vehicle pitch maneuver is shown as a rotation of the images about the center line. For a vehicle pitch of 1 0 , each image rotates 10 from its initial position. Determination of vehicle yaw or change in yaw angle requires the tracking of a recognizable object such as a known star pattern. Because the yaw heading
x could be determined as accurately by other means, the instrument is not recommended for this parameter. A general reticle which could be used with the instrument ure 4. The altitude scale on the reticle is for the moon. It the reticle that the amount of screen displacement for a given decreases as altitude increases. This is due to the geometric between horizon-depression angle and altitude of the moon. is shown in figmay be noted on altitude change relationship
The radial lines on this reticle are loo increments in pitch altitude; smaller increments could be used to allow more accurate positioning of small pitch angles. The tick marks on the vertical center line are 1 0 increments in roll attitude. When this general reticle is used, a local vertical should be established by bringing roll and pitch attitude to zero and then reading the altitude. This procedure is not a requirement, but simultaneous inputs complicate the interpretation of the information on the screen. SIMPLIFIED LUNAR LAUNCH A special screen reticle for a simplified lunar launch is shown in figure 5. This reticle would be used on the instrument for attitude and altitude information to accomplish a piloted lunar launch as described in paper no. 11. The pilot would perform a simple three-step pitch-program launch by obtaining altitude and pitch attitude from the instrument while holding the roll angle to zero by matching the two horizon images at the vertical center line. The LEM would take off vertically, and the view on the screen would show the horizons parallel and along the vertical center line. After 8 seconds timed by a clock, the pilot would pitch the vehicle over to 550 . He would accomplish this by bringing the horizon images over to , conform to the lower 550 lines. The altitude at this time is 300 feet as shown by the tick mark. The pilot would then hold the images parallel to the 550 lines until they match the second 550 lines (the altitude at this time is 17,000 feet). The pilot would then pitch the vehicle over to 50 by matching the images to the 50 lines. The pilot then maneuvers through small pitch-angle changes about the 50 lines as described in paper no. 11 so as to obtain zero altitude rate at about 50,000 feet. The reticle in figure 5 is presented only to show the principle; more reticle lines can be added for greater reading accuracy. The actual size of the scale used in this paper is 2.9 inches total distance from 0 to 60,000 feet, and there is 0.25 inch of displacement between 40,000 and 50,000 feet. The reticle used employs 20-inch focal-length lenses.
DISCUSSION The simple optical instrument discussed in this paper shows promise as a backup attitude and altitude instrument for use with the lunar excursion module. The reading accuracy of the instrument is directly dependent on the focal length of the lenses used. The reticles in this paper were designed to use 20-inch focal-length lenses. This focal length was chosen by using 2-inch-diameter lenses and a ratio of focal length to lens diameter of fl10. It seems reasonable that this f-number will give a clear usable image on the screen for the lunar lightintensity values now available. This f-number may become smaller when actual design parameters such as cockpit lighting and field of view are considered. Static accuracies for the instrument using 20-inch focal-length lenses are as follows; at 50,000 feet, altitude can be read to 500 feet; pitch angle can be held to a given angle to within 1/2 0 ; and yaw can be read to 1/40 . The accuracies of the instrument under actual conditions, when vehicle dynamics and horizon roughness are considered, still remain to be determined. CONCLUDING REMARKS The concept of a simple optical device which could be used as a backup to onboard instrumentation on the lunar excursion module is described. The instrument presents vehicle altitude and attitude information on a screen which could be located on the vehicle instrument panel. A mockup of the instrument has been constructed. Accuracies of the instrument under actual conditions are still to be evaluated.
^" rvn w ^ v ^° ° x * °°° ^^ * m ^ °^° ^ °°° ^° ° ^ °"^ "^ °° ^° ^ °° ^~
SCHEMATIC OF SIMPLE OPTICAL INSTRUMENT
IR R0R /
LENS LEFT OBJECT
| RIGHT OBJECT
/ LEFT IMAGE
SCREEN VIEWS FOR VEHICLE MANEUVERS
SCREEN RETICLE FOR SIMPLE OPTICAL INSTRUMENT
SIMPLIFIED LUNAR-LAUNCH RETICLE
Page intentionally left blank
13. ANGULAR SEPARATION OF APOLLO AND LUNAR EXCURSION MODULE AT LANDER TOUCHDOWN By James L. Williams and L. Keith Barker
An analytical investigation has been made of the angular separation of the service module and the lunar excursion module (LEM) at lander touchdown for several values of thrust-weight ratios and LEM landing ranges. Two types of landing maneuvers were used: one type used a synchronous transfer orbit and the other a direct-descent gravity turn. All landing trajectories utilized two thrusting periods - one to deorbit and another for braking to the lunar surface with zero velocity. Descent maneuvers were initiated from parkingorbit altitudes of 43.5 and 86.9 international nautical miles. The results of this study have indicated that increasing parking-orbit altitude from 43.5 to 86.9 international nautical miles reduced the separation angle for a given thrust-weight ratio and landing range. For landings from synchronous transfer orbit of increased altitude, LEM was ahead of the Apollo command service module at LEM touchdown, and the separation angle was such that direct ascent to Apollo could be accomplished with reasonable values of thrustweight ratio and of angular range traveled to rendezvous®
Abort of the landing phase of a lunar mission could be caused by a number of reasons® In most cases it would probably be sufficient for LEM to abort to a low-altitude parking orbit and wait for the proper phase relationship before performing a transfer to the Apollo command service module. However, in some situations it might be necessary to abort the landing and perform a direct return to Apollo. This could be required, for example, if a major failure occurred in the LEM life-support system® If such a failure occurred shortly after the Apollo -LEM separation, rendezvous techniques could be used to return LEM to Apollo, and the return problem appears to be simple. However, if abort occurs during the final braking maneuver, both the separation distance and the relative velocity of the vehicles are large, and a direct return to Apollo might be difficult. Therefore, it appears that the most critical time for a direct abort exists at or near LEM touchdown on the lunar surface. The factor of primary interest in this situation appears to be the angular separation between Apollo and LEM. Therefore, an analytical investigation was made to determine the effect of total landing range and thrust-weight ratio on the angular separation of Apollo and LEM and to examine descent maneuvers which result in separation angles that permit direct ascent with reasonable thrust-weight ratios. This paper presents some of these results.
SYMBOLS Any consistent set of units may be used. In this report it is assumed that 1 international nautical mile = 1.85200 kilometers. F/Wo AL AG initial thrust-weight ratio, where weight is in terms of earth pounds total landing range over lunar surface, deg angular separation of LEM and Apollo command service module at LEM touchdown, positive when Apollo is ahead of LEM (see fig. 1), deg ANALYSIS For this investigation it was assumed that a circular Apollo parking orbit had been established and that all landing maneuvers were initiated from this orbit. Two types of landing maneuvers were used in this study. In one type of landing maneuver, the LEM first established an elliptic orbit having the same period as the parking orbit by thrusting in a direction almost along the radius vector. The descent maneuver associated with the synchronous transfer orbit is shown in figure 1. Final braking for the landing maneuver (gravity turn) was initiated from pericynthion. A constant thrust level was used throughout. In the second type of landing maneuver examined, a small amount of retrothrust against the velocity vector was used to initiate descent. The LEM vehicle then coasted to a lower altitude where braking thrust was applied to perform a soft landing. The deorbit and braking phases were both accomplished with the same constant thrust level, with thrust applied against the velocity vector. Typical trajectory characteristics are illustrated in figure 2 for the direct gravity-turn descent. The total landing range over the lunar surface for the direct gravity-turn descent is primarily a function of thrust-weight ratio and length of deorbiting thrusting time. Landing maneuvers from Apollo parking orbits were initiated from altitudes of 43.5 and 86.9 nautical miles. RESULTS AND DISCUSSION The effect of thrust-weight ratio F/Wo on the separation angle at LEM touchdown can be seen in figures 3 and 4 for landings initiated from parkingorbit altitudes of 43.5 and 86.9 nautical miles, respectively. In these figures the separation angle 09 is plotted against total landing range 8L for several values of thrust-weight ratio. Results are also presented for descents from synchronous transfer orbits. For comparative purposes, results for impulse calculations are also shown in figures 3 and 4. Two main points of interest can be seen from a study of these figures. First, for a direct descent, the separation angle 102
,LA depends strongly on thrust-weight ratio and LEM landing range. In general, a high thrust-weight ratio and long landing ranges tend to keep the separation angle low. Second, increasing the altitude of the Apollo parking orbit tended to reduce the separation angle for given values of thrust-weight ratio and landing range. The information shown in figures 3 and 4 can be used to indicate approximately the conditions required for direct ascent. Some preliminary calculations have shown that the ascent maneuver is almost a mirror image of the descent phase for thrust-weight ratios equal to or greater than 0.60. Therefore, the direct gravity-turn descent trajectories can be used as direct ascent trajectories to rendezvous. As an example, assume that LEM used a synchronous orbit for descent from an orbital altitude of 43.5 nautical miles. A study of figure 3 shows that the Apollo command service module is ahead of LEM by about 4 0 at LEM touchdown. In order to make a direct ascent to Apollo, LEM must gain 4 0 . A study of the gravity-turn trajectories shown in figure 3 indicates that a combination of very high thrust-weight ratio (F/Wo = 2.00) and the 180 0 Hohmann transfer would be required for direct ascent to the Apollo command service module in the parking orbit with altitude of 43.5 nautical miles. This situation of high thrust-weight ratio and long angular range is alleviated to some extent if the orbital altitude of Apollo is increased. The results for a parking-orbit altitude of 86.9 nautical miles are presented in figure 4. As mentioned earlier, the most obvious effect of the higher orbital altitude is to reduce the separation angle for a given thrust-weight ratio and angular range. Now from this altitude a synchronous landing trajectory places LEM about 20 ahead of Apollo at the time of LEM landing. It is possible to make a direct abort by using smaller values of ascent thrust-weight ratio. For example, a combination of an ascent thrust-weight ratio of 0.64 and angular range traveled over the lunar surface to rendezvous of about 105 0 permits direct ascent to the Apollo command service module. The curves of figure 4 can also be used to determine permissible hovering time if direct ascent to rendezvous is to be accomplished. For the following example it is necessary to remember that the Apollo command service module is moving at an angular rate of about 3 0 per minute. Now, consider a synchronous orbital descent (fig. 4); then LEM is seen to be ahead of Apollo by 2 0 at touchdown. If LEM hovers 1 minute, the Apollo will be ahead of LEM by l o . It can also be seen from figure 4 that LEM can make a direct ascent by using a combination of a thrust-weight ratio of 0.64 and angular range traveled over the lunar surface to rendezvous of about 1600. CONCLUDING REMARKS An analytical investigation of the angular separation of the Apollo command service module and the lunar excursion module (LEM) at larder touchdown has indicated that increasing the altitude of the Apollo parking orbit from 43.5 to 86 .9 international nautical miles reduced the separation angle for a given thrustweight ratio and landing range. For landings from synchronous transfer orbits of
increased altitude,, LEM was ahead of the Apollo command service module at LEM touchdown ., and the separation angle was such that direct ascent to Apollo could be accomplished with reasonable values of thrust-weight ratio and of angular range traveled to rendezvous.
LANDING MANEUVER FROM SYNCHRONOUS TRANSFER ORBIT
TRANSFER ORBIT--\ DEORBIT COAST
DIRECT-DESCENT LANDING MANEUVER FROM PARKING ORBIT
°^ °~» ~ *°^ ^ ~^ °° ~`° vm ^^° ^"
SEPARATION ANGLE BETWEEN LBN AND APOLLO ATLEMTOUCHDOWN PARK |NG-UR8|T ALT |TUUE~43.5 NAUTICAL MILES F lh
"0 (SYNCHRONOUS ORBIT) ` wo
&4] ~~ -- - -^ 0.64 ----l^ ----2.0O ^^``_ /MPULSE
SEPARATION ANGLE, 48 ' 0BC
` -8|- ^ 'IV0
TOTAL LANDING RANGE, 8 L' DEG
SEPARATION ANGLE BETWEEN LEPN AND APOLLO ATLEK8TOUCHDOWN PARK |NG-0R BIT ALTITUDE ~Wi4 NAUTICAL MILES
5EPARAT|0NANGLEzB 0. A8,
0.43 0 64
00 `—IMPULSE -l6 O 90 TOTAL LANDING RANGE, 8 DEG l8O
14. SIMPLE ABORT SCHEME FOR SYNCHRONOUS ORBITS By G. Kimball Miller, Jr., and L. Keith Barker
In this analytical study it was desired to investigate a simple abort scheme for the lunar landing trajectory which is dependent upon a minimum of automatic equipment. The abort scheme investigated is dependent only upon line-of-sight measurements to the lunar horizon and a timer, which are used to place the landing vehicle in a safe, nonimpact trajectory should an abort occur during the lunar descent.
INTRODUCTION Abort capability throughout all phases of the lunar mission is desirable for man's flight to the moon. One of the more critical phases of the mission from abort considerations is the lunar landing. In examining just what the requirements would be for an abort during the lunar landing to some safe nonimpact trajectory, the following factors were considered: Before starting any braking maneuver, the landing vehicle is in some orbit around the moon. Braking thrust is applied during the landing maneuver to reduce the vehicle velocity and altitude, and for a nominal maneuver this is accomplished by a prescribed thrust and pitch program. If an abort situation occurs during the landing, it should be possible to return nearly to the original orbit by making the abort path a "mirror" image of the landing path, as indicated in figure 1. It was therefore decided to examine the landing path from the synchronous orbit to determine some of its characteristics to see how they might be used or mechanized for abort situations.
h hp y
altitude above lunar surface, ft pericynthion altitude, ft flight-path angle, deg angle between vehicle thrust axis and line of sight to lunar horizon, deg initial thrust angle for two-angle abort plan, deg secondary thrust angle for two-angle abort plan, deg angular range over lunar surface, ft 107
elapsed time along gravity-turn descent to abort point, sec abort thrusting time, sec thrust, lb initial weight, earth pounds
proportionality constants vehicle velocity
CHARACTERISTICS OF NOMINAL LANDING TRAJECTORY It was assumed that an Apollo circular parking orbit of 80 nautical miles had been established about the moon and that the lunar excursion module (LEM) had been injected into a synchronous orbit having a pericynthion altitude of 50,000 feet® Thrust was then applied against the velocity vector at the pericynthion of the synchronous orbit and a gravity-turn descent was made. An initial thrust-weight ratio of 0.485 and a specific impulse of 305 seconds were used to perform the landing® Some of the more interesting characteristics of the landing path are shown in figure 2® This figure shows altitude above the lunar surface h and the flight-path angle y as functions of range from landing initiation. Note that the flight-path angle and altitude change only slightly over most of the landing range. Since the landing was a gravity turn, this meant that the vehicle attitude, relative to the local vertical, also remained nearly constant® Further, this also implied that the angle between the vehicle thrust axis and the lunar horizon also remained nearly constant over most of the landing range. This proved to be the case, as shown in figure 3® Figure 3 shows the angle rc between the vehicle thrust axis and the lunar horizon plotted as a function of range from landing initiation® Note that K changes very slightly over a large portion of the landing range. Therefore, it appeared that the lunar horizon might be a useful reference in developing a simple abort plan with the mirror-image concept described previously.
PRIMARY ABORT PLAN
Technique The abort procedure, using the mirror-image concept, is illustrated in figure 4. At any point along most of the powered descent, the vehicle velocity has been decreased by an amount AV by retro-thrust at essentially a constant K. Thus it appeared that a logical abort procedure would be to rotate the vehicle to attain an angle k with respect to the other lunar horizon, and to thrust sufficiently long to make up the LV which had been removed during the braking phase. In this preliminary study, it was assumed that vehicle staging and rotation of the vehicle occurred instantaneously at abort initiation. The initial thrust-weight
ratio at abort was 1.2, with a specific impulse of 305 seconds. This value of thrust-weight ratio is somewhat higher than the value presently contemplated for the LEM; however, the basic technique should not depend strongly on thrust-weight ratio. The thrusting period required to make up the AV is given approximately by the equation shown in figure 4. The thrusting period to F necessary to s impart the required velocity increase is approximately equal to the product of the elapsed braking time and the ratio of F/Wo at landing to F/Wo at abort.
Results Use of the abort procedure just outlined did not result in a mirror image of the landing trajectory; however, this was to be expected because of the simplified abort procedure. The procedure did, however, result in the establishment of nonimpact orbits when used over most of the landing range. Some of the results of this study are shown in figure 5. Figure 5 indicates the minimum pericynthion altitudes resulting from use of the ones-angle abort plan as functions of the thrusting angle K and range from landing initiation to the abort point R. Also shown is the elapsed time from landing initiation to the abort point, tA,o° For example, if abort occurs about 600,000 feet downrange from landing initiation (or 133 seconds after the landing maneuver has started) and if a thrust angle K of from -1 0 to 2 0 is used, the pericynthion altitude of the abort trajectory will be at least 30,000 feet as indicated by the shaded area. This simple one-angle thrust concept results in pericynthions above 30,000 feet if abort occurs in about the initial 70 percent of the landing range. However, as the figure indicates, the one-angle concept results in pericynthion altitudes which become undesirably low if aborts occur after about 70 percent of the landing range is covered. For example, if an abort occurs about 650,000 feet downrange from landing initiation, the minimum pericynthion altitude decreases to 20,000 feet and if abort occurs a little farther downrange, the minimum pericynthion altitude decreases to 10,000 feet. Aborts from farther downrange will be discussed subsequently. With respect to the satisfactory one-angle aborts discussed thus far, the resulting trajectories intersect the circular parking orbit as indicated in figure 6. At the first intersection, the lander leads Apollo by from 15 0 to 200. At the second intersection, the separation angle is less than 40 . Therefore, the trajectory resulting from the simple abort plan appears to place the lander in a situation suitable for rendezvous®
MODIFIED ABORT PLAN
As mentioned previously, aborts from farther downrange than about 70 percent of the nominal landing range using the simple one-angle abort plan resulted in undesirably low pericynthion altitudes; therefore, modifications of this abort plan were examined for the remaining 30 percent of the landing range. A plan which worked very well involves the use of two constant angles between the thrust vector and the lunar horizon.
Technique At abort initiation, thrust is directed 50 0 (k l = - 50) below the lunar horizon for a specified fraction of the elapsed time along the landing trajectory. The vehicle is then rotated to a more shallow angle K2 for a different fraction of the elapsed time. The thrusting periods can be expressed as follows:
50 0 = atA o s
Kl = -
\tA ' F 1 K2 = PttAo where a and P are constants depending on the value of F/W o at landing and the value of F/Wo at abort. For the aborts studied in this investigation a had a value of 0.10 and R a value of 0.32+. Results Results obtained by using these equations for aborts occurring downrange of 70 percent of the landing trajectory are presented in figure 7. This figure showE the minimum pericynthion altitudes resulting from the two-angle abort plan, where the initial thrust angle K l is -50 0 , presented as functions of secondary thrust angle r 2 and range from landing initiation to the abort point. Also shown is the elapsed time from landing initiation to the abort point. Notice that going from the one-angle to the two-angle concept resulted in higher pericynthion altitudes for the latter part of the landing trajectory. As an example of the use of this figure, assume that an abort occurs about 800,000 feet downrange from initiation of landing (or about 220 seconds after the landing maneuver is initiated). The procedure, based on the plan and given values of a and Q of this study, would be as follows Stage the vehicle, and simultaneously rotate to point the thrust axis 50 0 below the horizon. Thrust at this attitude for a period of 0.10 times the 220 seconds, that is, 22 seconds, and then rotate the thrust axis to about72o above the horizon for a period of 0.32+ times the 220 seconds, or 71.28 seconds. This procedure should result in an orbit having a pericynthion of at least 30,000 feet. Although it appears somewhat awkward to use these values to determine thrusting times, there are several methods of mechanizing simple timing devices to indicate the proper thrusting periods at the various attitudes. CONCLUDING REMARKS Work on the abort plan is still continuing, and the results presented so far are preliminary. However, the abort plan does appear to be promising since it is simple and does not depend upon elaborate equipment. 110
wo n° ^ °** p ~ "^ o ^^° ^^» °"° o ^ ^ ° ~ * " " v ° "° °°° ° y " .0 ^ ° ° o ^°
* °^ m ^° p ,, ° °
° °°* °° ^ ^^ ~ °°^
°m v° ° ° °°
ILLUSTRATION OF MIRROR — IMAGE ABORT PLAN
ABORT I NG
GRAVITY — TURN CHARACTERISTICS
20 O FL|GKFPATHANGUE, -20 y, DEG -40 -vv -80— ~ 20 x 104 ALTITUDE, h. FT 0 O lO 20 }O 40 0 ^ 0 ^ ^xl0 RANGE FROM LANDING |N|T|AJ|0N. R. FT
VARIATION OF THRUST ANGLE K ALONG GRAVITY-TURN DESCENT
K 40 20 0
THRUST ANGLE, K, DEG -20 -40 -60 -80
80 go x lo 10 20 ' 60 L LL 30 40 50 L 70 ' RANGE FROM LANDING INITIATION, R, FT
F igure 3
ILLUSTRATION OF THRUST ANGLE
) (/ Wo A BORTI NG
^m ° °w ^ ^°, ^ y ^ ° ^° , ^ ~^°
°^ "^ p n °^ * ** v ^ ^ u ° " w n ° " ~ ° ~~ e ° o v ° ^ °° °+* ,
°x wp e ~ *p
NUN|M0N PERICYNTHION ALTITUDES OF ABORT ORBITS ONE-ANGLE CONCEPT U H P' ^(<>>|<^^^ ^10,00 4 ] 2 l THRUST ANGLE, K, DEC » -1 -2
NOMINAL TOUCHDOWN POINT
RANGE FROM LANDING INITIATION TO ABORT POINT, R, FT
160 200 280
ELAPSED TIME FROM LANDING INITIATION TU ABORT POINT, t
INTERSECTION OF ABORT TRAJECTORIES WITH APOLLO ORBIT SECOND INTERSECTION POINT (LAN0ERLEADS BY LESS \
CIRCULAR PARKING ORBIT FIRST INTERSECTION POINT (LANDER LEADS APOLLO BY 150-200) ABORT ORBIT LANDING TRAJECTORY ABORT PO I NT
= °,° °
~ , ~
MINIMUM PER|[YNTH|ON ALTITUDES OF ABORT ORBITS
0 V 8 SECONDARY THRUST ANGLE, 7 OL | h~R P ^l0'0OO > 20,000 * >_}0,N0
^ 64t8 7276 80 84 88 92x, 4
RANGE FROM LANDING INITIATION TOABORT POINT, R, FT
3[C ELAPSED TIME FROM LANDING INITIATION T0 ABORT POINT, t )^ A. Figure 7
15. ABORT CONSIDERATIONS FOR THE LUNAR EXCURSION MODULE By David B. Middleton SUMMARY A study has been made of returning the lunar excursion module (LEM) to Apollo after an abort from the powered-descent portion of the lunar landing trajectory. The technique developed is first to establish the LEM in a low circular orbit of about 50,000 feet and then use an efficient orbital transfer back to Apollo. A family of safe orbits was selected and the study was limited to those orbits in which the LEM requires less fuel than for a synchronous-orbit transfer and rendezvous. Once the LEM is in the 50,000-foot "chasing orbit," the particular transfer orbit that the pilot chooses is governed by the time he decides to make the transfer and thus the phase angle between the two vehicles. The limits in the study allow up to an hour of "transfer window" to initiate this return trajectory to Apollo and the technique can be easily modified to extend the window. The abort technique considered in this study standardizes the procedure for returning to Apollo after an abort from any point on the powered-descent trajectory. The technique also offers the following additional advantages: (1) It allows a time flexibility for the pilot to evaluate his situation before committing LEM to the rendezvous trajectory. (2) The transfer windows are longest for aborts from the most critical part of the landing trajectory, that is, near touchdown. (3) Execution of the final transfer is a simple task in that the pilot has only to match the injection velocity with the proper phase angle. (4) All the orbits considered are safe in that they never bring the LEM closer than 50,000 feet to the moon. (5) None of the transfer orbits require more fuel than is required for the synchronous-orbit transfer and rendezvous. INTRODUCTION Investigations are being made of the procedures and requirements for returning the lunar excursion module (LEM) to Apollo after an abort from the lunar landing mission. Figure 1 shows the Apollo/LEM mission profile. The studies, at present, are being made by assuming the 80- nautical-mile circular Apollo orbit and injection of the LEM into an equiperiod or synchronous orbit having a 50,000-foot pericynthion. At pericynthion, where the LEM begins its powered descent to the lunar surface, it leads Apollo by about 80, its flight-path angle is zero, and its velocity is about 190 fps greater than circular velocity.
Direct return to Apollo is apparently the most difficult from the powereddescent part of the mission because of the rapidly changing velocity, flight-path angle, and phase angle between the two vehicles. Nevertheless, in certain emergencies where the return time is of primary consideration, it may be necessary to use this method and there are a number of studies underway to develop the associated operational techniques. However, in abort situations where the time constraint is less severe, other techniques of return are available. The purpose of this paper is to consider and develop one such abort technique. DISCUSSION After the decision to abort from anywhere along the powered-descent trajectory, the LEM is injected into a low circular orbit of about 50,000 feet. Then, after the pilot has had more time to evaluate the situation, he initiates an efficient transfer back to Apollo. Injection into the low circular orbit is a relatively easy task. In a simulation study (paper no. 11), the simplicity of manually controlling the take-off from the lunar surface into a 50,000-foot cir-. cular orbit is demonstrated. Also, the technique presented in paper no. 14 seems applicable as a simple means of reaching the 50,000-foot orbit from any point along the powered-descent trajectory. Since the circular velocity in the 50,000-foot orbit is greater than that in the Apollo orbit, the 50,000-foot orbit is referred to as the "chasing orbit." After the pilot establishes the LEM in the chasing orbit, he has a whole family of safe trajectories to choose from for his orbital transfer back to Apollo. Figure 2 shows some of these orbits which are generated in the following manner: At some point in the chasing orbit, the pilot adds an incremental velocity AV of about 99 fps at zero flight-path angle and the result is a Hohmann transfer ellipse up to the Apollo orbit. This, of course, is the most efficient method of transfer. If the pilot adds velocity in excess of this 99 fps, he will generate orbits having apocynthions greater than 80 nautical miles and each orbit will cut the Apollo orbit in two places. When this excess reaches 91 fps, he will attain an orbit which has a period equal to the Apollo orbit. Now, for any of these orbits to be appropriate for rendezvous, the pilot must pay particular attention to the phase angle between the two vehicles at the time of injection into the transfer orbit. For example, the Apollo must be about 0 92 ahead of the LEM for initiation of a Hohmann transfer, about 80 behind for a transfer using an equiperiod orbit, and somewhere in 0between to use one of the intermediate orbits. This phase-angle spread of 172 , representing about 51 minutes, can be termed a "transfer window." Also, any transfer using the equiperiod orbit will be referred to as a "synchronous-orbit transfer." In this study, the LV required for the synchronous-orbit transfer is used as an upper limit and thus only the use of orbits between this transfer orbit and the Hohmann transfer is investigated. It is also of interest to note that 116
the AV required to rendezvous with Apollo from the Hohmann transfer orbit is about 97 fps and that required to rendezvous from the synchronous orbit is about 373 fps. Thus, rendezvous from any of the intermediate orbits is bounded by these two values. Since the phase angle is such an important parameter in this study, it should be mentioned that it can be measured with a simple optical device such as a sextant since an error as large as t o would only cause a miss-distance of about 15 nautical miles. A simulation study (ref. 1) has shown that rendezvous can still be easily accomplished and with small fuel penalty. Once the LEM is in the chasing orbit, the particular transfer orbit that the pilot chooses is governed by the time he decides to make the transfer; that is, at any given time, he can choose the intermediate orbit appropriate to the phase angle at that time, or he can wait for the phase angle associated with a particular transfer orbit that he would like to use. Figure 3 shows how these determinations can be made. The ordinate shows values of the Apollo lead or lag angle at the moment the LEM enters the 50,000-foot chasing orbit following abort from any point along the powered-descent trajectory. This angle '^S8c is plotted as a function of the time of abort after initiation of the powered descent at pericynthion. The upper horizontal line in the figure defines the suitable phase angle for Hohmann transfer and the lower line defines the suitable phase angle for synchronous-orbit transfer. The waiting time in the chasing orbit between these two conditions (transfer window) is about 51 minutes, as mentioned previously. If abort occurs before the beginning of the powered descent, the pilot does nothing except continue to coast around in the original synchronous orbit for the prescribed meeting with Apollo. But for abort at all later times, the same initial task is required; that is, the pilot must get into the chasing orbit and then decide on the best final transfer. The curve in figure 3 shows that for abort anytime during the first 125 seconds, the phase angle for synchronous-orbit transfer will be favorable as soon as the LEM enters the chasing orbit. At the beginning of hover the Apollo is nearly overhead, but will lead the LEM by about lo o by the time the LEM can get into the chasing orbit. Thus an abort from this part of the trajectory will put the LEM in the chasing orbit a couple of minutes before the phase angle is favorable for a Hohmann transfer. However, if the pilot decides not to transfer at this time, he still has nearly an hour of transfer window to make one of the intermediate transfers. To make a Hohmann transfer after an abort following several minutes of hovering will require a wait of about 8 2 minutes in the chasing orbit for each minute spent hovering. The curve in figure 3 between 125 and 325 seconds corresponds to the phase relationship as the LEM enters the chasing orbit after the Hohmann phasing has passed but before the synchronous-orbit phasing has arrived. In this region the pilot may initiate one of the intermediate orbits shown in figure 2. As noted previously, in order to generate these orbits, the pilot has to put in some LV in excess of that required for the Hohmann injection; but, of equal importance, 117
the pilot must know how to match this AV with the proper phase angle. Figure 4 shows the correlation of phase angle and excess AV. The curve labeled "2nd intersection" indicates the amount of velocity the pilot must add to LEM at injection for a given phase angle. Since the intermediate orbits have characteristics similar to the synchronous one, rendezvous normally would take place at the second intercept. (See fig. 2.) If the pilot, however, is interested in trying the rendezvous at the first intercept, the curve in figure 4 labeled "1st intersection" indicates that the phase angle has to be within of the Hohmann phasing or the limiting AV for
synchronous-orbit transfer will be exceeded. So this curve is considered only if the Hohmann phasing has just been missed but the pilot still wants to return 11 quickly. Incidently, this range could be extended by using other than zero flight-path angle at injection, but then the transfer orbit would have a pericynthion lower than 50,000 feet and the pilot would be initiating an unsafe orbit. Where rendezvous is made at the second intercept, there is an attractive feature with regard to the detection and correction of injection-velocity errors. The change in apocynthion altitude is very sensitive to injection-velocity errors; for each foot-per-second error the change in apocynthion altitude is nearly 5,000 feet. This relation is nearly linear over the range of injection velocities under consideration. So, if the altitude is measured with a radar having an accuracy within 1 percent over this range, the.injection error can be determined within less than 1 fps. Once the magnitude of the error is known, appropriate energy adjustments can be made at once to LEM or Apollo which will effect a new rendezvous position. The types of adjustments considered are as follows: (1) Add AV (energy) to the LEM in such a manner that the transfer-orbit pericynthion is not reduced, (2) add energy to Apollo but never remove any, because any removed will have to be put back in subsequently for the return trip to earth, or (3) use a combination of types (1) and (2). Typical altered trajectories and new rendezvous positions are shown in figure 5. Instead of correcting an injection-velocity error, the following maneuvers might be considered: (1) When the LEM reaches apocynthion, it can add enough AV tangentially to go into a circular orbit at the apocynthion altitude and (2) Apollo can initiate an appropriate orbital transfer up to the LEM; thus, the original pattern is reversed. Such an extension of the abort technique need not be reserved for error cases, but may be used to advantage as a regular procedure for any abort from the powered-descent trajectory. In figure 3 it is apparent that for aborts during the first 125 seconds, a synchronous-orbit transfer is called for as soon as the LEM enters the chasing orbit. However, with the modification just introduced, the LEM would be allowed to wait some time in the chasing orbit and then initiate one of the intermediate orbits shown in figure 2. Upon reaching apocynthion, the LEM would go into a circular orbit and gain a favorable phase relationship for Apollo to make a Hohmann transfer up to LEM. The total fuel used by the two vehicles in this modified maneuver is less than the
total for a synchronous-orbit transfer. In addition, the energy added to Apollo is not wasted since it would have to be added anyway for the return to earth. CONCLUDING REMARKS The abort technique which has been considered for the lunar excursion module (LEM) seems to offer the following advantages: 1. It allows a time flexibility for the pilot to evaluate his situation before committing LEM to the rendezvous trajectory. 2. The "transfer windows" are longest for aborts from the most critical part of the landing trajectory, that is, near touchdown. 3. Execution of the final transfer is a simple task in that the pilot has only to match the injection velocity with the proper phase angle. 4. All the orbits considered are safe in that they never bring the LEM closer than 50,000 feet to the moon. 5. None of the transfer orbits require more fuel than is required for the synchronous-orbit transfer and rendezvous.
REFERENCE 1. Brissenden, Roy F®, Burton, Bert B., Foudriat, Edwin C., and Whitten, James B.: Analog Simulation of a Pilot-Controlled Rendezvous. NASA TN D -747, 1961.
APOLLO/LEM MISSION PROFILE
– POWERED— DESCENT TRAJECTORY
FAMILY OF TRANSFER ORBITS CONSIDERED
o = RENDEZVOUS WITH APOLLO TYPICAL INTERMEDIATE ORBIT SYNCHRONOUS ORBIT—\q CHASING ORBIT
APOLLO ORBIT HOHMANN TRANSFER ORBIT
^ rF ^' a
APOLLO/LEM PHASE RELATIONSHIP AS LEM ENTERS CHASING ORBIT AFTER ABORT
PHASE ANGLE FOR HOHMANN TRANSFER APOLLO 10 LEADS 5
Ae c , USE INTERMEDIATE ORBIT
DEG APOLLO LAGS
1 HR PHASE ANGLE FOR SYNCHRONOUS-ORBIT TRANSFER
200 300 400
TIME FROM PER ICYNTHION OF ABORT, SEC
CORRELATION OF PHASE ANGLE AND EXCESS AV
EXCESS AV REQUIRED
FOR SYNCHRONOUS-ORBIT INJECTION PHASE ANGLE FOR HOHMANN TRANSFER
10 1ST INTERSECTION
2ND INTERSECTION PHASE ANGLE FOR SYNCHRONOUS-ORBIT
TRANSFER -10 0 20 40 60 80 100
AV IN EXCESS OF HOHMANN INJECTION VELOCITY, FPS
°~^ p ^^ ' *=° " °° ^» w , ^ ^°^ r
TYPICAL ALTERED TRAJECTORIES AND NEW RENDEZVOUS POSITIONS
— LBNTRANSF[R ORBIT WITH INJECTION ERROR
TYP| CAL ALTERED v^ ro^/r,roorcc
NEW RENDEZVOU S \Vll POSITIONS
16. PENETROMETER TECHNIQUES FOR LUNAR SURFACE EVALUATION By John Locke McCarty, Alfred Gm Beswick, and George W. Brooks
The penetrometer technique for evaluating characteristics of remote targets is discussed as a means for studying the lunar surface structure. A history of the concept is presented together with possible applications of the technique to both manned and unmanned lunar spacecraft® In addition, basic considerations for an optimum penetrometer design are given, and research study areas in penetrometer development are outlined.
Reliable data on the nature of the lunar surface structure is urgently needed for engineering purposes as well as to satisfy scientific curiosity® The hardness, bearing strength, and texture of the surface of the moon is of particular importance to the successful accomplishment of any manned lunar landing phases of the Apollo mission. With this view in mind, a general study was undertaken several years ago to evaluate the possibility of relating the physical characteristics of target surfaces to acceleration time histories observed during impact on those surfaces of accelerometer-equipped projectiles or penetrometersm The results of this study indicated that certain physical characteristics of a target material could indeed be defined from the impact acceleration time histories® Thus, it appeared that penetrometers would serve as a means for evaluating characteristics of the lunar surface prior to a manned landing. From this general study, consideration was given to the applications of the penetrometer technique to various lunar spacecraft: first, as an experiment aboard unmanned vehicles and, more recently, as a sounding device in support of manned vehicles. The purpose of this paper is to summarize briefly the general impact research study conducted at the Langley Research Center and to discuss the possible application of the penetrometer technique to both manned and unmanned spacecraft.
a amax t
acceleration maximum acceleration encountered during impact arbitrary time
tr rise time for peak acceleration in acceleration time history F'F.. 123
total time for acceleration time history weight of projectile
GENERAL IMPACT STUDY
The general study was aimed at evaluating the impact characteristics of penetrometers on target materials which were representative of the different categories of impact - that is, elastic, plastic, penetration, and combinations of these - and were not intended to be representative of anticipated lunar surface media. The scope of these impact tests is presented in table I and includes the impact of various metallic projectiles, over two velocity ranges, on a wide variety of target materials with measurements taken of the impact accelerations and characteristics of the resulting craters. The results of these impact tests, presented in reference 1, indicated that several characteristics of the acceleration time histories are of significance in defining the target material® These characteristics are noted in the time history in figure 1 and include the magnitude of the peak acceleration amax, the time required to reach that acceleration (designated by tr for rise time), the total duration of the pulse t T , and the overall shape of the pulse. Figures 2 and 3 are presented to illustrate how a remote target such as the lunar surface can be evaluated from a knowledge of certain of these characteristics. In figure 2, the ratio of the measured peak acceleration to the impact velocity is plotted as a function of total pulse time for a 2-inch-diameter steel hemispherical penetrometer impacting on a variety of the test materials which ranged in surface hardness from sand to concrete. The important facts shown by these data are as follows. If a projectile, or penetrometer, such as this is impacted on the lunar surface at a known velocity, a good idea of the nature of the surface may be obtained either by measuring the peak acceleration or the total pulse time since the magnitude of these characteristics is dependent upon the target material. Hence, it appears that the hardness of the lunar surface can be described in terms of the hardness of accessible earth materials from a knowledge of either of these impact characteristics. With the additional knowledge of the pulse shape, a great deal more can be learned about the nature of the lunar surface structure, including the existence of possible soft dust layers® To demonstrate this capability, typical acceleration time histories are presented in figure 3 for impacts on several target materials used in the general investigation® Both the acceleration and time scales in this figure are normalized: acceleration on the basis of the maximum acceleration encountered during impact and the time with respect to the total pulse time. Acceleration time histories are presented for an impacting body striking an elastic material such as concrete for which the pulse shape is nearly symmetrical about a mean vertical line which passes through the peak; a plastic material such as lead for which the restitution increment of the pulse shape approaches a straight vertical line; and two-layer configurations which involve penetration of the projectile into the impacted medium, the class of targets generally anticipated in lunar experiments. The pulse shapes illustrated for the layer configurations resulted from impacts into loose peat moss on balsa and into a layer of ground portland cement on soil 124
and indicate the ability of the penetrometer technique to distinguish the layer characteristics of target materials. Impacts into dust-like media, which are presently being evaluated, yield pulse shapes which are characterized by two acceleration peaks as illustrated in the portland cement portion of the layer configuration in the lower right-hand corner of figure 3. The initial peak is believed to be associated with the rapid compression of the target material ahead of the partially imbedded projectile which, upon expansion, forms the crater and the second peak results from the drag forces tending to decelerate the projectile. From this brief review of results of the general study it can be concluded that much information on the nature of the lunar surface can be gained from complete acceleration time histories recorded during the impact of penetrometers on that surface.
APPLICATIONS TO SPACECRAFT
Consideration was given to the application of the penetrometer technique as an experiment to evaluate characteristics of the lunar surface and as a means for sounding that surface in support of manned landings. Two applications are briefly discussed; one wherein the technique is adapted to payloads already situated on the lunar surface and the other having application to spacecraft positioned above the lunar surface. Figure 4 presents one conception of a penetrometer payload designed to operate as an experiment on the lunar surface. The operational technique of this concept consists of launching, from the payload, spherical projectiles equipped with omnidirectional accelerometers to impact on the lunar surface. Signals from the penetrometers pass to the payload either through a telemetry link or, as shown in figure 4, through a trailing wire and are then transmitted to a receiving station either on earth or onboard an overhead spacecraft. A penetrometer payload based on this concept could be readily adapted to the Ranger spacecraft; moreover, a more extensive evaluation of the lunar surface would be afforded through the use of larger, more sophisticated, spacecraft such as the Surveyor type. In either case, penetrometer experiments performed from payloads situated on the lunar surface demand the successful landing of the payload on a surface of unknown character in an attitude compatible for experimental operations. Applications of the penetrometer technique to spacecraft above the lunar surface avoid many of the demands placed upon the lunar-based experiment. The operational procedure for such applications is illustrated in figure 5. In this application, the penetrometers are released in free fall from a payload - or a manned spacecraft - during a retroburning stage of the descent trajectory. In this application, the penetrometers are simply self-contained acceleration-measuring telemeters which transmit the developed impact information to the overhead spacecraft. For unmanned missions, the spacecraft serves as a relay station which conditions the information for retransmission to earth or to a parent lunar orbiter. However, should the spacecraft be manned, the penetrometers act as surface sounding devices and this impact information is presented to the astronaut for his immediate use.
PENETROMETER DESIGN CONSIDERATIONS
The application of the penetrometer technique as shown in figure 5 was evaluated in some detail at the Langley Research Center (LRC) for the Ranger spacecraft. During this evaluation, several basic requirements were established for lunar penetrometer design which are appropriate to all spacecraft whether manned or unmanned. The characteristic requirements believed necessary for application in any penetrometer system are as follows. First, it is readily apparent that the optimum and practical penetrometer would contain an omnidirectional accelerometer which would require no specific orientation during flight or during impact; thereby, attitude stability problems and elaborate deployment techniques from spacecraft would be eliminated. In addition, such penetrometers would be operational despite local surface slopes which would tend to affect penetrometer attitude upon impact. Second, the range of measurable accelerations should be adequate to accommodate surface materials having bearing strengths which extend from that of fine dust to beyond that required to support landing vehicles. For manned lunar landers, it is anticipated that three orders of magnitude would be required on the acceleration scale to comply with this requirement. Third, complete acceleration time histories are required to describe adequately the characteristics of the surface, particularly in view of the fact that the lunar surface may consist of layers of materials of different hardness. Fourth, penetrometer dispersion is necessary in order to secure significant samplings of a designated lunar surface area, this dispersion being limited, of course, to the length of trailing wire available should that technique be employed or to the line-of-sight requirements of telemetry systems. Fifth, the body and the encapsulated components must be sufficiently rugged to remain operational throughout all anticipated impact loads. Any less rugged structure may lead to an erroneous interpretation of the data. The final design consideration pertains to the requirement that the penetrometers must be amenable to earth calibrations which define the instrument acceleration response characteristics to various earth materials.
FUTURE PENETROMETER RESEARCH
The penetrometer characteristics given in the previous section refer to an optimum system having application in any penetrometer experiment or sounding device. However, the present state of technology imposes practical limitations on what can be achieved. Consideration of these factors has been the guideline for the LRC lunar penetrometer research program which is continuing in the following areas®
Further research involving basic impact tests will include an evaluation of the impact characteristics of various penetrometers into rubble media of different shape, size, and density and into dust media having a wide range of bearing strengths. In addition, an instrumentation development effort is continuing to perfect omnidirectional accelerometer and antenna systems, to proof-test and extend the capability of LRC developed unidirectional penetrometer systems, including the possible application of these systems to achieve omnidirectional capability, and to extend further telemetry components and circuitry performance for application to penetrometers.
Basic impact research has indicated that the penetrometer technique is a practical system capable of evaluating characteristics of a remote target such as the moon. Application of this technique to lunar spacecraft is appropriate to unmanned exploratory missions and as a sounding device for landing-site evaluation on manned missions.
1. McCarty, John Locke, and Carden, Huey D.: Impact Characteristics of Various Materials Obtained by an Acceleration-Time-History Technique Applicable to Evaluating Remote Targets. NASA TN D-1269, 1962.
SCOPE OF IMPACT TESTS
TEST VELOCITY RANGE
TARGET CONCRETE LEAD BALSA SOD SAND (00 GRADE) PEAT MOSS CEMENT DUST PEAT MOSS AND SAND MIXTURE SOIL CONCRETE PEAT MOSS CEMENT DUST LAYERS: PEAT MOSS ON BALSA CEMENT DUST ON SOIL
TEST VARIABLES (PROJECTI LE)
LOW RANGE (5-40 FPS)
*ACCELERATION TIME HISTORY o PENETRATION DEPTH •CRATER DIAMETER
HIGH RANGE (100-900 FPS)
MASS NOSE SHAPE
FOR 100 TO 240 FPS: ®ACCELERATION TIME HISTORY *PENETRATION DEPTH FOR 250 TO 900 FPS: *PEAK ACCELERATION *PENETRATION DEPTH
CHARACTERISTICS OF IMPACT ACCELERATION TIME HISTORIES
SUMMARY OF IMPACT DATA FOR A STEEL HEMISPHERE
PROJECTILE (W = I LB) Kz::) CONCRETE SURFACE .10,000 — MAX. ACC. IMPACT VEL. 1,000— BALSA SOD ^ LEAD
10 t T' msec
GENERAL SHAPES OF IMPACT TIME HISTORIES
a a max
a a max
OPERATING TECHNIQUE FOR LUNAR-BASED PENETROMETER EXPERIMENT
OPERATING TECHNIQUE FOR LUNAR PENETROMETER
':i ^ ^
rF k z
^ ^b ^b
17. PROPOSED SURVEYOR LANDING EXPERIMENT By Sidney A. Batterson
A description is given of a proposed landing experiment to be carried on the Surveyor spacecraft during touchdown on the lunar surface. Instrumentation is discussed which will be used to obtain data on the vehicle motions during landing and on the physical properties of the moon surface. This proposed experiment could be carried on all seven flights of the Surveyor flight program scheduled to begin during the last quarter of 1964.
The design of vehicles capable of making soft landings on the moon requires more information about the lunar surface than is now available in order to predict the ground forces that will be developed during the landing impact. As a means of obtaining lunar surface information as well as vehicle motion and attitude data, an experiment has been proposed for use during the lunar landings of the Surveyor spacecraft. Since the landing system and the initial landing velocities of Surveyor are somewhat similar to those of the lunar excursion module (LEM), the description of the proposed experiment should be of direct interest. At the present time, this proposal is only in the formative stage. Since the first Surveyor flight is scheduled for the last quarter of 1964, the results of the experiment could be available in time to be applied to LEM. The purpose of this paper is to describe the proposed experiment and indicate the general scope of the investigation.
DESCRIPTION OF VEHICLE
A photograph of the Surveyor spacecraft is shown in figure 1. It is composed essentially of a space frame which houses the main retro-engine and the scientific experiments. The space frame is supported on three landing gears, two of which can be seen in figure 1. Each landing gear consists of a tripod arrangement of two lower legs which pivot about a hinge line located on the main frame. The upper leg is the shock absorber and it pivots about a point also located on the main frame. Ground loads are transmitted to the landing gear through the footpads attached at the end of each landing gear. The three footpads can be enclosed by a circle of approximately 13 feet in diameter. Weight of the vehicle at lunar touchdown will be approximately 600 pounds. The structure is designed to withstand maximum initial contact velocities of 20 feet per second vertically and 7 feet per second laterally.
Figure 2 is a schematic drawing of the spacecraft and shows the instrumentation provided for the proposed landing experiment® A total of six accelerometers will be mounted at primary structural joints in the main space frame. The accelerometers will be oriented so that their outputs will define the six translational and rotational motions of the spacecraft during the landing impact. Strain gages will be mounted on each leg of one of the landing gears to measure the forces developed during the landing. This same landing gear will also be equipped with an instrument to measure the telescoping displacement of the shock absorber. In effect the landing-gear instrumentation converts this one landing gear into a three-component dynamometer capable of measuring both the magnitude and direction of the applied ground force and the motion of the landing-gear footpad. Although it would be desirable to have each of the three landing gears equipped in a similar way to measure the forces and motions, this is not possible because of present limitations of the available telemetry bandwidth. Since the telemetry bandwidth must be divided among the instruments in accordance with their individual response requirements, a few remarks on bandwidth as related to the experiment might be in order. From the standpoint of frequency response, the accelerometers are more critical than the landing-gear instruments. Landing tests have been made at the Hughes Aircraft Company under simulated lunar gravity with a full-scale dynamic model of Surveyor. The results of these tests, which are reported in reference 1, indicate that 8 cycles per second is the lower limit of acceptable frequency response for the accelerometer data. It is therefore planned to transmit the accelerometer data on channels having a 10-cycle-per-second bandwidth. The remaining bandwidth will be divided equally among the three strain gages and the shock-absorber displacement measuring instrument. At present it appears that these data will be transmitted on channels having an upper response limit of about 8 cycles per second. Obtaining an accurate reference to the lunar vertical axis still remains a problem. A study is being made to develop a method for establishing an accurate gravity reference after the landing has been completed. However, a gravity reference established prior to the landing impact would be desirable if the landing is not completely successful.
At the present time the Surveyor flight program consists of a total of seven flights. The first four flights will be engineering test flights in which the primary emphasis will be placed on the operational phases of the equipment® The last three flights will be scientific test flights, during which instruments will be soft-landed on the moon to gather basic scientific and engineering data. The proposal is that this landing experiment be carried on all flights of the Surveyor program in order to obtain maximum information about the lunar surface. The first of these flights is scheduled to take place during the last quarter of 1964.
The main objective of this paper has been to describe briefly a proposed experiment to be used with Surveyor during a lunar landing. The results to be obtained from this experiment will consist of the translational and rotational motions of the vehicle, the motion of one footpad, and the magnitude and direction of the resultant load applied to this footpad by the lunar surface.
1. Deitrick, R. E., and Jones, R. H.; Surveyor Spacecraft System - Touchdown Dynamics Study (Preliminary Report). SSD 303OR (JPL 950056), Hughes Aircraft Co., Jan. 1963.
SOLAR PANEL 2 I% HIO" GAIN //'ANTENNA
THERMALLY CONTROLLED COMPA9,TMENT TV
SOFT SOFT LANDING TECHNOLOGY *SURVEY VARIOUS LANDING
SURFACE SAMPLER IK 0MNI ANTENNA
oMEASURE PHYSICAL .&
SURVEYOR LANDING INSTRUMENTATION
SHOCK-ABSORBER DISPLACEMENT INSTRUMENT
Figure 2 '
18. RADIO- FREQUENCY SIGNAL ATTENUATION BY PLASMAS
OF ROCKET EXHAUST GASES
By Duncan E. McIver, Jr. SUMMARY A series of ground and flight experiments are being conducted at the Langley Research Center to investigate the causes of rocket-exhaust-induced radiofrequency signal attenuation. The results, which can be related to liquid- or solid-propellant rocket motors, indicate that the factors which determine attenuation are the electron density and collision frequency, the electromagnetic signal frequency, and the size of the exhaust-free jet. Analysis of flight results reveals that attenuation can result from either an exhaust "surface effect" or a it effect." A surface-effect attenuation, which occurred below 250,000 feet and affected only telemetry signals, was eliminated by the injection of small amounts of decomposed hydrogen peroxide on the exhaust surface. A film from an in-flight camera experiment shows that the surface effect is associated with a luminous region (identified as afterburning) on the exhaust surface. A volumeeffect attenuation, which occurred above 350,000 feet, was severe enough to cause complete loss of telemetry and command-destruct capability at the launch site and to seriously degrade the radar tracking signals. Ground tests, in which small rocket motors were fired in a 60-foot-diameter vacuum sphere and wherein the exhausts were probed with microwaves, showed that the exhaust electron density increases with increasing combustion temperature. Since the Surveyor lunar research vehicle may experience attenuation of its Doppler velocity and altimeter radar signals during landing, measurements of the plasma properties of the vernier control engine exhausts are being conducted at the U.S. Naval Research Laboratory, Washington, D.C. Preliminary results indicate that there will be no problem from a single motor exhaust. However, the intersection of multiple exhausts and exhaust interaction with the lunar surface may be problem areas. If-radar data are considered necessary for the Apollo LEM (lunar excursion module) landing maneuver, data from these studies should be carefully examined and extrapolated for the LEM. INTRODUCTION Attenuation of Doppler velocity and altimeter radar signals by the exhausts of landing motors is one of the operational problems that may confront a lunar landing vehicle such as the lunar excursion module (LEM) of Apollo. Attenuation of radio-frequency OF) signals is caused by free electrons in the exhaust-free jet. These electrons are usually created by ionization processes in the high temperature of the combustion chamber. The source of the ionization is normally attributed to trace amounts of alkali metals which contaminate the propellant (ref. 1). Ionization from shocks in the exhaust may also contribute to the electron level (ref. 2). At low ambient pressures the exhaust 135
expands and intercepts the line of sight from the spacecraft to its target, and RF signals traversing the volume of the exhaust suffer attenuation. Other interference effects such as reflection, refraction, and noise may also be present. The zone of interference is determined by the size of the exhaust, which in turn depends on several parameters (ref. 3) but primarily on ambient pressure. Additional exhaust electrons may be ionization reactions, and afterburning. trated layer of electrons on the exhaust signal loss. Either this surface effect degradation of RF energy. generated by ablating nozzles, These processes tend to create surface which may be the prime or the volume effect can cause photoa concenregion of serious
The interaction of electromagnetic energy with the exhaust plasmas of both solid- and liquid-propellant rocket motors is complex; however, the degree of interaction depends on the electron density and the electron collision frequency along the signal path and the electromagnetic signal frequency. If the exhaust electron density is sufficiently below the critical value associated with a given radar frequency, the interference will be slight (refs. 1 and 4).
THE LANGLEY ROCKET EXHAUST PROGRAM
The Langley Research Center (LRC) rocket exhaust program includes the investigation of the exhaust RF signal interference problem on the Scout, NASA solidpropellant research vehicle. The second- and third-stage rocket motors exhibit exhaust interference effects from which considerable data have been accumulated and analyzed. Ground tests are being conducted to investigate fundamental problems and to examine specific propellants. The prime facility used for these tests is a 60-foot-diameter vacuum sphere at LRC, in which small rocket motors are fired at simulated altitudes. Microwave probes are employed to determine the exhaust plasma properties - electron density and collision frequency. Langley is also making attenuation measurements on the exhausts of fullscale, statically fired motors. One object of these tests is to look for scaling relationships between the full-scale motor and smaller motors used in the vacuum facility. The data from the tests are compared with theoretical descriptions of the exhaust in an attempt to construct an adequate model of the exhaust plasma for theoretical predictions. As part of the program, Langley is conducting piggyback experiments on the Scout vehicle. A recent test was the flight of a recoverable camera which photographed the exhaust of the second-stage motor during a period of RF signal
1 F3 f. FA C2
attenuation. (See ref. 5.) Other experiments include the measurement of exhaustinduced separated flow and the in-flight addition of various materials to reduce or eliminate exhaust attenuation. (See ref. 6 for a discussion of separated flow.) The broad objective of the rocket exhaust program, besides developing an understanding of the problem, is to be able to predict the degree of exhaust interference for a specific mission, such as Apollo LEM, and if severe to be able to offer practical solutions. Some aspects of this program are discussed in detail in the following sections.
Scout Second-Stage Motor Figure 1 illustrates the attenuation problem on the Scout second-stage motor (ref. 7), which is an example of the surface effect. The signal strength record obtained at the NASA Wallops Station (launch site) shows that marked VHF telemetry signal attenuation begins at about 190,000 feet (second-stage motor ignition occurs at 130,000 feet) and is intermittently eliminated with the actuation of small hydrogen peroxide control jets located near the second-stage nozzle. Since the exhaust gases are rich in unburned hydrogen and carbon monoxide and react with the atmosphere, afterburning on the exhaust surface is assumed to be the cause of attenuation. Recovery from attenuation is caused by the quenching action of the control-jet exhausts which are rich in water vapor. The mass flow of the small control jets is only one-seventieth the mass flow of the second-stage motor. These results suggest a possible solution to exhaust-induced attenuation, the external injection of materials. Figure 2 illustrates a recoverable camera which was used with Scout to investigate this phenomenon further (ref. 5). The camera, which was located at the top of the second-stage motor about 15 feet from the exhaust, was in operation from lift-off to burnout of the second-stage motor and was then ejected for recovery. The film clearly shows the formation of a luminous region on the exhaust surface at the onset of attenuation. The intermittent disappearance of a portion of this luminous region coincides with control-jet firing and signal recovery. The results of this experiment appear to substantiate the afterburning-surface-effect concept.
Scout Third-Stage Motor Shown in table I is a range summary for one of the Scout third-stage motors which uses a propellant similar to that used in both the Minuteman and Polaris missiles (ref® 8). This table illustrates how severe the exhaust attenuation problem can be. Telemetry and command-destruct capability (vHF and UHF) was lost and radar track was marginal at the launch site. These functions were recovered by the use of a down-range station at Bermuda, which does not "look through" the
exhaust. The LRC station, which was at a vehicle-to-tracking-station aspect angle less than that for Bermuda (measured with respect to the vehicle longitudinal axis), recovered partial telemetry data. The severe attenuation was anticipated since the combustion temperature of the third-stage propellant is higher than that of the second-stage propellant. (Ionization increases with temperature.) Because of the extreme altitude of operation, afterburning would not be expected and attenuation is probably a volume effect®
The more fundamental parameters of the exhaust problem are studied in ground tests® These tests examine the sources of ionization, investigate different propellants, determine the effects of changing ambient pressure, and test schemes for the reduction of radio-frequency attenuation. The primary facility for these studies is a 60-foot-diameter vacuum sphere at LRC® As shown in figure 3, this sphere presently can be operated at pressures as low as 10 -4 mm Hg. This is a static facility, that is, there is no evacuation during motor operation; however, pressure rises are slight when small rocket motors are used® Small solid-propellant motors (usually 2 pounds of propellant) using operational propellants are fired at various ambient pressures and the exhaust jets are probed with microwave beams to determine their plasma properties. Several frequencies are employed to analyze the exhausts since interference varies with frequency. The measurements obtained are attenuation, phase shift, and plasma noise (microwave radiation generated by the exhaust plasma), from which electron concentration and collision frequency are determined® Data at the bottom of figure 3 were obtained from three different propellants. As expected, the maximum electron density increases with increasing chamber temperature. These experimental data are in reasonable agreement with predicted ionization levels based on reported alkali metal contamination of the propellant (ref, 1). The major problem associated with the extrapolation of the experimental data from small motors to large motors is the determination of accurate scaling laws® The discussion in reference 2 has indicated that, for liquid-propellant motors, attenuation should scale as the square root of the thrust® This scaling law has not been verified for solid-propellant motors. Tests are being conducted on full-scale motors in an attempt to develop scaling relationships.
THE SURVEYOR EXHAUST PROBLEM The Surveyor landing maneuver is illustrated in figure 4. (See ref. 9 ®) The main retrorocket, which uses a solid--propellant rocket motor similar to the Scout third-stage motor, will be ignited about 280,000 feet above the lunar 138
F F G? 6
r '^i '2x (q
surface and will burn out at about 27,000 feet. For initial missions, plans were made to design around the possible signal interference from the main retrorocket by increasing vernier control responsibility. This decision added several pounds of vernier-engine fuel. At main retrorocket burnout time, the spacecraft Doppler velocity and altimeter radar systems will take over and control the rest of the flight with the vernier engines. This is the period during which radar signals may encounter interference from the vernier-engine exhausts. There are three engines in the vernier control system, one mounted on each of the Surveyor landing legs, and each is throttleable to yield from 30 to 104 pounds of thrust. The propulsion system utilizes a liquid bipropellant of monomethylhydrazine (fuel) and nitrogen tetroxide (oxidizer). The chamber temperature is about 5,200 0 F. Two radar antennas are employed on the spacecraft to generate the Doppler velocity beams and the single altimeter beam. The operating frequencies are about 13 kmc. The Doppler beams are oriented about 20 0 with respect to the thrust axes, whereas the altimeter beam will probably look straight down. As a result of the possible exhaust attenuation problem on the Surveyor, tentative plans were made to test the Surveyor landing system with operational vernier engines and radar in the 60-foot-diameter vacuum sphere. At present, tests are being conducted at the U.S. Naval Research Laboratory (NRL) by W. W. Balwanz of NRL and H. Shuman of Hughes Aircraft Company to determine the attenuative properties of the propellant for use in the vernier engine. The results of these tests will determine whether the Langley experiments are necessary. Preliminary results from the NRL tests indicate that there is no radarinterference problem from a single vernier-engine exhaust. This result is in agreement with several predictions. (See ref. 10, for example.) The tests are continuing, however, to determine the maximum allowable alkali metal contamination of the propellant. Two phenomena that may create attenuation problems are shown in figure 5. The first, shock ionization, which results from the intersection of the multiple vernier exhausts, may create regions which will interfere with radar transmission. Because of the location of the antennas, radar signals must traverse these intersection regions. Preliminary results from the NRL tests indicate that shock ionization in the exhaust of the vernier engine may be more severe than for other liquid propellants. The second is the interaction of the exhaust with the lunar surface. The temperature behind the standing shock wave, at the lunar surface, approaches the combustion temperature and may create an ionized layer which would interfere with radar signals. This latter effect would occur only when the spacecraft is close to the moon surface and may not be serious. APOLLO LEM Requirements for radar during the landing phase of the LEM do not seem firm at this point, primarily because of the presence of a pilot. If radar is
considered necessary, all rocket-exhaust-attenuation knowledge, especially the results from the NRL tests, should be reviewed for extrapolation to the LEM. Since the propellant planned for the LEM (a 50-50 combination of hydrazine and unsymmetrical dimethylhydrazi.ne as the fuel and nitrogen tetroxide as the oxidizer) has a similar combustion temperature and composition to that of Surveyor, the attenuation effects of the smaller motor operated at NRL (about 100 pounds thrust) might be scaled to the larger LEM motor (about 10,000 pounds thrust). (See ref. 2.) According to the scaling laws of reference 2 the attenuation may be higher by a factor of 10. Another problem peculiar to the LEM vehicle is the possible enhancement of ionization in the exhaust due to the ablation of the rocket nozzle. The ablative material, which is needed to cool the nozzle, should be carefully examined as a possible electron source. The problem areas of multiple exhaust intersection and exhaust interaction with the lunar surface also apply to the Apollo LEM.
The study of rocket-exhaust-induced RF signal attenuation at LRC has been reviewed and some ground and flight tests have been discussed. This brief review of the problem of exhaust attenuation shows that the electron density and collision frequency along the signal path and the exhaust size determine the degree of interaction for a given radar system. The sources of electrons appear to be ionization processes in the combustion chamber (probably caused by alkali metal contaminants), afterburning, shock ionization as a result of the intersection of multiple exhausts, and ionization from ablation-cooled nozzles. Depending on the relative strength of the various processes, the interaction may be either a surface or a volume effect. If radar data are considered necessary for the Apollo LEM landing maneuver, existing data, especially those from the tests being conducted at NRL for Surveyor, should be carefully examined and extrapolated for the LEM. The areas of multiple exhaust intersection and interaction with the lunar surface need to be examined as sources of radar attenuation.
1. Calcote, H. F., and Silla, H.: Radar Attenuation in Solid Propellant Rocket Exhausts. Bull. 18th Meeting JANAF-ARPA-NASA Solid Propellant Group, Vol. III, June 1962, pp. 3-50. 2. Balwanz, W. W., and Stine, P. T.: The Plasmas of Missile Flight - Analysis of Polaris A2X Signal Absorption. NRL Rep. 5636, U.S. Naval Res. Lab., July 24, 1961. 3. Love, Eugene S., Grigsby, Carl E., Lee, Louise P., and Woodling, Mildred J.: Experimental and Theoretical Studies of Axisymmetric Free Jets. NASA TR R-6 1 1959. (Supersedes NACA RM L54L31 by Love and Grigsby, RM L55J14 by Love, RM L56G18 by Love, Woodling, and Lee, and TN 4195 by Love and Lee.) 4. Altshuler, S., Moe, M. M., and Molmud, P.: The Electromagnetics of the Rocket Exhaust. GM-TR-0165-00397, Space Technology Labs., Inc., June 15, 1958. 5. McIver, Duncan E.I. Jr.: Study of the Effects of a Rocket Exhaust on RadioFrequency Signal Attenuation by the Use of a Recoverable Camera on the NASA Scout Vehicle. NASA TM X-888 1 1963. 6. Falanga, Ralph A., Hinson, William F., and Crawford, Davis H.: Exploratory Tests of the Effects of Jet Plumes on the Flow Over Cone-Cylinder-Flare Bodies. NASA TN D-1000, 1962. 7. Sims, Theo E., and Jones, Robert F.: Rocket Exhaust Effects on Radio Frequency Propagation From a Scout Vehicle and Signal Recovery During the Injection of Decomposed Hydrogen Peroxide. NASA TM X-529, 1961. 8. Fetner, E. M., and Hoffman-Heyden, A. E.: R-F Interference by Solid Fuel Rocket Exhausts. AFMTC-TR-61-10 (Contract No. AF 08(606)-3 1+13), Air Force Missile Test Center, U.S. Air Force, Sept. 1, 1961. 9. Miller, Barry: Surveyor Vital to Manned Lunar Program. Aviation Week and Space Technology, Vol. 77, no. 15, Oct. 8, 1962 ; pp. 77-83. 10. Molmud, Paul: Part II - Estimates of Interference Due to Vernier Exhaust To Be Encountered by Radar and Altimeter Systems During a Lunar Landing. [Preprint] 2586-62, American Rocket Soc., Oct. 1962.
TABLE I RANGE SUMMARY FOR SCOUT THIRD-STAGE MOTOR
ALTITUDE OF OPERATION, 360,000 TO 700,000 FT
WALLOPS TELEMETRY NO DATA (z50 DB LOSS) TRACK RADAR MARGINAL
PARTIAL DATA COMPLETE DATA (40-50 DB LOSS)
(40 - 50 DB LOSS) COMMAND
20° - 35°
SCOUT SECOND-STAGE MOTOR
ANTENNA ^^-TO LANGLEY (54°)
SECOND-STAGE MOTOR CONTROL JETS
^-TO WALLOPS (40)
VHF SIGNAL STRENGTH, 0
CONTROL r JET L OPERATION ON------ --------OFF --J
SCOUT CAMERA EXPERIMENT
CAMERA FIELD OF VI
Figure 2 60-FOOT VACUUM SPHERE
MINIMUM PRESSURE = 104
MICROWAVE INSTRUMENTATION 0 10 KMC 0 20 KNIC MEASUREMENT ATTENUATION PHASE SHIFT HIGHLY ALUMINIZED SOLID-PROPELLANT MOTORS (PROPELLANT, 2 LB)
MOTOR CHAMBER TEMP., O F ELECTRONS PER CM
A B C
4,860 5,225 6,270
< 10 10 3 x 1011 2 x 10 11
^* °°° m ^°. ° ^^ ^^ ^ + o ~^° ^= * " e^ , v~ * n v * ^ ^ ° ^= n ^ *' ^^^ ^ * ° ^ n °° ^ ^ ° o ^ ^^ ~~^ ~°
SURVEYOR VERNIER ENGINE (L|UU|D)
1. MAIN RETROROCKET IGNITION
MAIN RETROROCKET (S0L/0
—C^)— 280,000 R 27,000 FT
l MAIN RETROROCKET BURNOUT
I VERNIER DESCENT GU| DANCE PH 4. VERNIER SHUTOFF
^-----__^ ------' l]FT S
R A DAR
FUEL MON0METHYLHYURAZ|NE OXIDIZER N|TROSENTETROX|DE
SURVEYOR ADD IT] 0NAL PROBLEM AREAS
19. DYNAMIC PENETRATION AND EROSION OF DUST-LIKE MATERIALS IN A VACUUM ENVIRONMENT By Leonard V. Clark and Norman S. Land
Two programs are discussed which relate to possible problem areas for manned and unmanned landings on a dust-covered lunar surface. The first concerns the penetration characteristics of projectiles impacting into fine particles in a vacuum pressure environment, and the second an investigation of the interaction between jet exhausts and adjacent dust surfaces, also in a vacuum pressure environment. This interaction includes erosion and visibility considerations.
One requisite for a successful soft landing on the moon is the ability of the vehicle upon landing to remain on the surface in a position acceptable for surface operations and relaunch to earth. Certainly, a factor which governs this ability is the nature of the surface. Until that nature is adequately known, the design of vehicles such as the Apollo-LEM (lunar excursion module) must include the effects of a wide range of possible surfaces. The possibility of a dust-covered surface is supported by many astronomers and scientists; however, there is general disagreement as to its origin, quantity, distribution, and particle size. In view of the dust concept, attention must be given to a number of possible problem areas related to manned and unmanned landings on such dust-like surfaces. For example, the possibility of a dust-covered lunar surface of considerable depth introduces concern over the unknown depth to which bodies could possibly penetrate the surface upon landing. Still another area of concern is the possibility of dust entrainment by the retro flow field, which may result in the formation of a crater immediately beneath the vehicle and in visual obscurement of the surface by the accompanying erosion products. The purpose of this paper is to review two Langley Research Center programs in these problem areas. In general, an attempt is made to determine whether there are phenomena associated with the landing of vehicles on a dust-covered lunar surface that would pose particular problems for the Apollo system.
_ ............. n
SYMBOLS do D h M nozzle exit diameter particle diameter, microns height of nozzle above surface Mach number
P C combustion chamber pressure, lb/sq in. abs r R Te y y radial position gas constant, ft2 sect-oR exit temperature dust depth ratio of specific heats PENETRATION IN DUST-LIKE MEDIA The attitude and positioning requirements of the LEM vehicle very definitely promote interest regarding the characteristics of bodies landing or impacting on dust-like media in a vacuum environment. For this reason the results of reference 1 are of interest as attention is given to the penetration characteristics of projectiles impacting into fine particles in a vacuum environment. As an indication of the results of this study, a sample of the data is shown in figure 1. This figure presents the measured penetration depths for a spherical projectile as a function of the environmental pressure at the impact surface. The dust beds used in this study consisted of aluminum oxide particles. Penetration data are included for dust beds of average grain size from 600 to 3 microns over a range of pressures between atmospheric and approximately 10 -5 torr. The figure shows that the depth of penetration into relatively unpacked particles of size equal to or greater than 32 microns (particles of this size have an obvious granular consistency) is independent of the environmental pressure. However, for particles of less than 32 microns (particles of this size are rather powdery in texture), the depth of penetration decreases rapidly with decreasing pressure until the pressure is reduced to about 1 torr. For lower pressures the depth of penetration remains unchanged to 10-5 torr, the vacuum limit of the investigation. Figure 1 further shows that the depth of penetration in the granular particles is apparently unaffected by particle size, whereas the penetration in the powdery materials increases with decreasing particle size.
f ^f-0 t2
From these data it appears that particle grain size and degree of vacuum pressure must be given consideration when investigating the impact characteristics of dust-like media. These data provide a background for an experimental investigation of the erosion of similarly sized materials caused by an impinging jet in a vacuum environment.
EROSION IN DUST-LIKE MEDIA
Recently, considerable attention has been given to investigating the interaction of jet exhausts and adjacent dust surfaces; however, the necessary experimental work has yielded little useful quantitative data, primarily because of lack of scaling considerations. Accordingly, preliminary tests have been conducted in the Langley 60-foot vacuum sphere on the behavior of a horizontal layer of dust when subjected to a vertical jet blast. Data were taken which indicate erosion and visibility as a function of time, Scaling The parameters involved in experimental studies of jet erosion have to be scaled properly to yield results which have meaning. The results of analytical studies (see ref. 2) indicated that the significant considerations associated with the flow of the jet and its subsequent interaction with and erosion of the surface are the following: (1) Simulation of the flow-field density distribution (2) Ratio of gravitational force to shear force in the boundary layer (3) Surface Reynolds number (4) Ratio of cohesive force to shear force in the boundary layer For the erosion studies conducted in the 60-foot vacuum sphere, the scaling equations given from a consideration of the first three items were used to scale the model for a set of prototype values that were selected as being representative of current thinking. These values are given in table I. Quantitative data on interparticle forces are meager but indicate that these forces are negligible except for very small particles. The ratio of the cohesive force to shear force in the boundary layer was therefore not used in scaling considerations for these preliminary tests.
TABLE I.- TYPICAL PROTOTYPE CONDITIONS [Fuel: Unsymmetrical dimethyl hydrazine; N204]
y. . . . . . . . . . . . . . . . . . . . . . . . . . . 1.25
P C ) lb/sq in. abs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ........................... M.
4.5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1,980 . . . .
. . . . . .
a e r ;.}
The length scale for the tests was approximately 10 and the time scale was approximately hat is, Prototype length - 10 and Prototype time ; 23 . Model time Model length 23t The times given in the subsequent erosion and visibility figures are model times. APPARATUS AND TEST PROCEDURE Apparatus
X-ray technique for dust depth measurement.- Dust depth during the transient
erosion process was measured by an X-ray absorption technique. This technique is illustrated in figure 2. An X-ray machine mounted above the dust bed projects rays downward through the dust. The bottom of the dust bed (3/4 -inch steel plate) essentially blocked the X-rays except those which could pass through a radial slot in the bed plate. A film drum beneath the plate carried X-ray film across the"slot and provided a complete time history of the erosion process. Any point on the film then had coordinates of time and radial position, and the transparency of the film was a function of the film exposure (dust depth). A commercial densitometer was used to read the film to obtain the erosion time history. The typical film shown in figure 2 indicates how the erosion starts at a point away from the nozzle stagnation region and begins by digging an annular crater. The mechanics of the phenomena have been discussed in reference 2.
Description of visibility measurement technique.- The technique used to obtain an indication of the visibility during erosion of the surface by the jet is illustrated in figure 3. The visibility measurement consisted of recording the output of a photocell which looked at a collimated beam of light originating from across the dust bed. The light beam was 5/8 by 1316 inch in cross section. Visibility through the dust cloud was measured along a path parallel to the surface of the dust - 3 nozzle diameters above the surface and 3 nozzle diameters to one side of the nozzle. The sample record shown has coordinates of percent visibility and time. "Unaffected" denotes that all the light is being received, and "opaque" means that all the light is being blacked out by the erosion products.
Test Procedure In the experimental studies of erosion four types of dust were used: coarse and fine sand, ground pumice, and aluminum oxide particles. The average grain size of these materials ranged from 5 microns for the aluminum oxide particles to 260 microns for the coarse sand. The dust beds were prepared to a depth of 6 inches with no deliberate attempt at packing. From scaling considerations an isentropic expansion nozzle designed for a Mach number of 3.4 with atmospheric air was used for the model nozzle. The model
nozzle exit diameter was 1 inch and was chosen to keep the sphere from filling up too rapidly. The nozzle stagnation pressure ranged from 11.6 to 15 pounds per square inch absolute and the temperatures from 500 F to 600 F. A movable deflector plate was mounted a short distance below the nozzle exit in an attempt to eliminate starting transient shock waves from impinging on the dust bed® One second after the start of nozzle flow this deflector plate was moved aside, and the end of this movement was called time zero. At the start of a test run, the pressure of the 60 -foot vacuum sphere ranged from 2.3 X 10- 4 to 4®8 X 10-4 torr. At the end of a 10- to 20- second run, the sphere pressure ranged from 4.9 X 10-2 to 2®3 X 10-1 torr® RESULTS AND DISCUSSION A part of the test data is presented in figures 4 to 8. Figures 4 and 5 present the crater profile at various times for the jet impinging on a granular material (coarse sand) at nozzle heights of 5 and 2®5 nozzle exit diameters. The lower nozzle height (2.5 exit diameters) corresponds to the approximate point at which the LEM retro will shut off and the vehicle will free-fall onto the lunar surface® As anticipated, the results of the tests show that surface erosion is highly dependent on the height of the jet above the dust surface® An additional item of interest is shown by the data presented in figure 5, that is, that the sides of the craters in the granular materials reach angles on the order of twice the static angle of repose. (The standard static angle of repose for granular materials is approximately 35 0 .) The sides of the craters are held there by the nozzle-exhaust pressure field, and an inward slump occurs when the nozzle flow ceases. The data from figures 5 and 6, that is, the crater profiles for a nozzle height of 2.5 exit diameters for coarse sand and aluminum oxide particles, can be compared to show the dependence of erosion on interparticle cohesive force. This observation follows from a discussion presented in reference 1 of the action of particles to penetration under vacuum conditions (that is, the granular particles displace from the impact region whereas the powdery particles simply pack locally). It was observed that the powdery particles erode somewhat erratically and in clumps, whereas the granular particles erode by the action of individual grains. The visibility data for these crater profiles are presented in figures 7 and 8. These figures present the mean level of visibility as a function of test time® Data scatter varied from 1-percent visibility for the high nozzle heights to 7 percent for the low nozzle heights. Figure 7 gives the visibility profiles for coarse sand at nozzle heights of 5 and 2.5 exit diameters. Apparently, visibility also depends to a large extent on nozzle height above the surface. Both of the curves shown in this figure and in the subsequent figure returned to 100-percent visibility when the jet was shut off. At the indicated point in the lower curve the jet had eroded to a sufficient depth to expose the bedplate, whereupon the visibility increased. Finally, an indication of the effect of material grain size on visibility can be obtained from a comparison of figures 7 and 8® The latter figure shows the visibility profiles for aluminum oxide particles at nozzle heights above the surface of 4®6 and 2 ®5 exit diameters. At
the high nozzle heights the granular material affects visibility more than the powdery material; whereas, at the low nozzle heights the fluctuation of visibility with time makes for a difficult comparison. CONCLUDING REMARKS In conclusion, the data from these dust studies indicate that the magnitude of any landing problem for the lunar excursion module (LEM) created by the existence of a layer of dust on the lunar surface is dependent to a great extent upon jet height above the surface, hover time (or, correspondingly, the time of descent), and dust-particle size. Additional experimental work is needed in order to define better the effects of the various parameters involved. Specifically, the results of dust studies appropriate to the Apollo mission indicate the following tentative conclusions: 1® From the penetration data it appears that a simulation of the lunar pressure environment by 1 torr is just as satisfactory as a simulation by 10 -5 torr. 2. Visibility and erosion problems increase with decreasing jet height above the dust surface 3. The cohesiveness of the particles greatly retards the erosion process. 4. Cohesive particles erode somewhat erratically and in clumps, whereas cohesioeless particles erode by the action of individual grains. 5. Visibility tends to increase when the bottom of the dust bed is reached.
6® The sides of the craters in the granular particles sometimes reach angles greater than the static angle of repose (being held there by the nozzle-exhaust pressure field), and an inward slump occurs when the nozzle flow ceases.
REFERENCES 1® Clark, Leonard V., and McCarty, John Locke: The Effect of Vacuum on the Penetration Characteristics of Projectiles Into Fine Particles. NASA TN D -1519,
2. Roberts, Leonard: The Action of a Hypersonic Jet on a Dust Layer® Paper No. 63-50, Inst. Aerospace Sci., Jan. 1963.
°^ r ^° ° ° °°° °°° ~ ^ °° - " ^~° ° ^
° po ~s ° °^* , *~^ ~ ° o s ^ v °° ~ °^ ~~
^° ^= ° ^ ~^
PENETRATION IN DUST-LIKE MEDIA
GRAIN SIZE, MICRONS
o 600 o 32
X-RAY SOURCE JET &^ .
DUST BED —' ^ fF|UMURUM
DEB RIS —
L DUST BED
00 V|8|O|L|7Y, PERCENT 0
EROSION OFGRANULAR MATERIALS h ~5 EXIT U|AM.; U~2hOMICRONS dn~^
| O -- l o 2 O
I' 4 dn
~° ^ xm . ^ °°^ ~~o °~ ~^^ °. ",~
* °* ^* " °°° ° ,^^ °" * m^ » °^ °° °" ^ ~= °°
EROSION OF GRANULAR MATERIALS h ~l5 EXIT U|AM.; D~26WMICRONS d =^ ^— NOZZLE
" U l 2 3 O I 2 9 r d_ n 4 5 t
EROSION OF POWDERY MATERIALS h ~l5 EXIT D|AM.; 0~jMICRONS
0 l 2 ^ 'O
3 r ^d n
VISIBILITY PROFILES IN COARSE SAND
D = 260 MICRONS 100 80 60 VISIBILITY, PERCENT 40 20
8 6 TIME, SEC
VISIBILITY PROFILES IN ALUMINUM OXIDE
D = 5 MICRONS h, EXIT DIAM. 4.6
100 80 60 40 20
!.5 0 2 4 6 8 10 12
20. VISIBILITY AND DUST EROSION DURING THE LUNAR LANDING By Leonard Roberts SUMMARY A theoretical study is made of the dust erosion and loss of visibility that results when the lunar excursion module (LEM) approaches the lunar surface. The theory is compared with available experimental data, and the results are then used to determine a least-visibility estimate for the LEM conditions. INTRODUCTION The lunar excursion module (LEM) of Apollo, in its descent toward the surface of the moon, will necessarily have its thrusting rocket directed downward in order to support the weight and check the descent of the vehicle. As it approaches the surface, the forces exerted by the exhausting rockets will be sufficiently large to cause a disturbance of the lunar surface material; when this disturbance occurs, entrainment of dust by the gas flow may result in the formation of a crater immediately beneath the vehicle and a dust cloud around the vehicle. Such phenomena (fig. 1) could present a hazard to the safe operation of the vehicle, and it is therefore important to determine the factors which could produce such a situation. In order to investigate on a laboratory scale some of the problems associated with landing on a dusty surface, it is necessary to have some understanding of the scaling laws so that the results may be properly interpreted and the correct conclusions drawn with respect to the full-scale vehicle. The present theoretical study was undertaken in order to provide some insight into the mechanism of erosion and dust entrainment and to establish the more important scaling parameters so that an experimental program could be undertaken. Some of the details of this study are to be found in reference 1. Such a program is presently being carried out at Langley Research Center in a 60-foot-diameter vacuum sphere ' and it is hoped that the results will show the extent to which erosion and reduction in visibility affect the lunar-landing operation. The nature of the lunar surface is as yet unknown, especially at a scale that is comparable to the dimensions of the vehicle, but it seems likely that the surface will consist of a deposit of dust of varying depth over a substructure of harder rock ' although there is little real evidence to allow any conclusive quantitative estimate of this depth. In the present study, therefore, the depth of the dust layer and the size of the particles remain arbitrary.
A quantitative description of the erosion and the associated dust cloud can be obtained only by considering the details of the mechanism of dust entrainment by the aerodynamic flow of the rocket exhaust gases over the surface. A limited experimental study (ref. 2) has been made of the effects of an exhaust jet impinging on a dusty surface and this work provides some information on the flow field and a qualitative picture of the erosion. Theoretical work in the subject has been carried out in the last 2 years at the Langley Research Center (refs. 1 and 3), and an experimental program has been partly completed in the 60-foot-diameter vacuum sphere as discussed in paper number 19. The present paper will summarize briefly the theoretical results, compare them with the available experimental information, and apply the results to the full-scale vehicle. SYMBOLS a
dimensionless momentum of entrained dust particles dust packing factor
CE attenuation coefficient Cf friction coefficient D g h k k' M m r R T p t
dust particle diameter, ft gravitational constant vehicle height, ft hypersonic similarity parameter, roughness height, ft Mach number mass of dust eroded radial distance, ft gas constant temperature pressure, lb/sq ft time, sec radial velocity
light-intensity ratio depth of erosion angle of repose of cohesionless particles bulk density of dust coefficient of viscosity shear stress, lb/sq ft stress ratio azimuthal angle from jet center line gas density ratio of specific heats
r 8 P y
Subscripts; c g coh
combustion-chamber conditions gravitational condition cohesive location of maximum shear nozzle exit THEORY General Flow Description
The exhaust flow considered is assumed to be an inviscid ideal gas issuing steadily from a nearly parallel nozzle at hypersonic velocity into a vacuum. Most of the mass and momentum of the jet are contained in a central core in which the density decreases both along and normal to the axis. Since the flow is hypersonic, it is undisturbed by the presence of the surface beneath the vehicle until it impinges on the surface, where a shock wave, approximately parallel to the surface, is formed. On passing through the shock, the gas is diverted to flow primarily in a radial direction over the surface until at large distances the density becomes negligibly small. The flow over the surface produces pressure and shear forces, which act on the dust particles at the surface, causing them to be entrained by the flow and 157
The extent to which this momentum transfer takes place depends on the imposed shear stress and on the maximum shear stress that the surface can withstand; when the maximum permissible shear stress is small compared with that imposed by the flow, the dust entrainment will take place at a high rate. The erosion rate, therefore, depends on such dust properties as the particle size and density, the dust cohesive strength, the depth of the dust layer, and, of course, the gravitational field. In addition, the erosion rate depends on the vehicle thrust and height above the surface, the flow Mach number and Reynolds number, and the surface roughness, since these quantities determine the imposed shear stress distribution. The dust cloud formed flows primarily outward away from the vehicle but may also be directed upward by impact with the edge of the eroded region. The extent of this cloud and the concentration of dust within it increase as the vehicle approaches the landing point, so that visibility is reduced. A quantitative description of the interaction described above entails a study of the various contributory regions as shown in figure 2. In the following sections the exhaust flow field, the surface pressure distribution, and the associated distribution of shear stress (over the ground) are described. Attention is then turned to the conditions at the surface which determine the dust entrainment and surface erosion. Finally, consideration is given to the dust cloud and its effect on visibility. Exhaust Flow and Impingement It is supposed that the exhaust flow issues isentropically from a nozzle of exit radius rn at a high Mach number Mn. Most of the mass within the jet is concentrated near the center of the jet so that the density, while decreasing radially, also decreases in the azimuthal direction until at some angle the density is negligibly small. The essential character of this flow is described quantitatively by considering the balance of mass and momentum flux in the jet using an assumed density profile which varies as (cos 9) k where 8 is the azimuthal angle measured from the center line of the jet. The resulting expression for the density on a spherical cap at a distance h from the nozzle exit is written (see ref® 1 for details of analysis) r(h)-2 (cos 9) k (1) Pn = 2 where k = y(y - 1)Mn2 . Thus, the density distribution depends on a single hypersonic parameter k defined in terms of the Mach number of the exhaust and 158
the ratio of specific heats y. It is seen that the density decays according to the inverse square law, as is expected. For the LEM vehicle k is approximately 6.3, and this value was duplicated in the experiments (paper no. 19) by adjusting the Mach number of the nozzle.. The pressure distribution over the surface is found to be p = Thrust AJh 2/(k + 2)J and this expression is valid for h 2/(k + 2) > rn. Above the surface a saucer-shaped shock wave (concave upward) forms with a standoff distance of about 0.06h. Viscous Effects The state of the gasdynamic flow over the surface beneath the vehicle depends on the local Reynolds number, which may be expressed in terms of a basic Reynolds number parameter Thrust nrnµc Tc and in terms of h 2^(k + 2) and 8. rn 4 cos 8 ) k+4
The condition of the flow at the point of maximum dynamic pressure on the surface (1) = tan7 1 2/(k + 2)) is shown in figure 3 as a function of the Reynolds number and the dimensionless height h 2/(k + 2). For the LEM vehicle the
Reynolds number as defined above is about 10 5 , and it is seen that the flow over the surface is noncontinuum for vehicle heights above about 300 feet but continuum and laminar below this height. If the surface is rough (i.e., roughness scale k' greater than the boundary-layer thickness), the flow is considered to be in a rough dynamic condition, and the shear forces acting on the surface will be of the same order of magnitude as the dynamic pressure and independent of the Reynolds number. For the LEM vehicle the flow ceases to be laminar when the vehicle is below about 100 feet in height if the surface roughness is of the order of 1 foot. Although the surface forces are independent of Reynolds number under these conditions, the viscous forces on individual particles that have been lifted into the flow are important, and for this reason the Reynolds number remains an important parameter. The Reynolds number for the experiments in the 60-footvacuum sphere was therefore made to agree approximately with the full-scale value.
Surface Erosion The shear forces on the surface are approximated by
T = C fpu2
where Cf is a constant of order unity. The maximum value of T occurs at A = Am = tan- 2/(k + 2) and is written in terms of the vehicle thrust as
TM Thrust 7rh 2^( 2 k + 2)]
2 e Cf
and the shear force distribution by
_ sin 812 cos 8^k+2 m
The imposed gasdynamic shear force T tends to move the particles from the surface into the flow, whereas gravitational and cohesive forces oppose this tendency. These latter forces may be written (per unit surface area) T* = agDc tan a + Tcoh where c tan m ~ 0.4 for most particles and Tcoh is the cohesive shear stress. It is known experimentally that 'Tcoh is small compared with a9Dc tan a for particles of diameter larger than about 10- 4 ft. For smaller particles, however, Tcoh becomes much larger than the gravitational stress and becomes the dominant resisting stress for particle diameters less than about 3 x 10-5 ft. On the assumption that the cohesive force between two particles is independent of the size, the stress 'Tcoh varies inversely as the square of the diameter, and if it is assumed that Tcoh = cgDc tan m when
D=3x 1075 ft
ag3x10- 5 xo.4 2 D2 (3 x l0-5)
or Tcoh - 1. 7 X 10-2 (105 D)-
where the diameter is in feet and the stress in lb/sq ft. This assumed variation Of Tcoh with D is conservative in that Tcoh probably varies more rapidly with D and is larger than the value given by the inverse square law. If the applied gasdynamic shear stress is considered to represent a certain rate of transfer of momentum to the surface and is balanced by the resisting surface shear stress plus the rate of transfer of momentum to particles entrained by the flow, then
where u is the flow velocity over the surface, au (where a < 1) is the mean particle velocity, and ^ is the rate of introduction of the mass of particles into the flow. Clearly when Tm < T*, that is, when (('' 2 < 2C f (CrgDc tan a + Tcoh) ,[ [h 2^(k + 2fl there is no erosion. When the particles are large, T coh $:^ 0 and a "gravity cutoff" to the erosion occurs when
Thrust n[+/(k + 2)^
— (2Cf c tan a)crgD
or with Cf = 1 and c tan m = 0.4, when h
On the other hand when the particles are cohesive (and therefore very small), then crgDc tan m ~ 0 and a "cohesive cutoff" occurs when
Thrust 2 - 2C f 1.7 X lO' 2 (105 D) - 2 - 0 ® 05 (105 D) _ 2 n [h 2/(k + 2 J
that is, when h + 2 = Thrust (105 D)
Between these limiting conditions, erosion will occur according to the equation
M = T- T*
oc dt = 1--auT* The quantity a is itself a function of the dust particle size and of the viscous and pressure forces that act on the particle. An analysis of the motion of particles induced into the flow shows that a is given, approximately, by
36h1 a2 =
k+ 2 1 + 1 300 crD2 2RTc
Thrust X D 2 2 nhk+2 Il e fRTC
Thus, the erosion rate, evaluated at 8 = 8m (where u = is written
2RT c approximately),
For the experimental conditions in the 60-foot vacuum sphere, the denominator of this latter expression was approximately unity, in which case the erosion rate is expected to vary as (Dust diameter) X (Height)-5/2
Figure 4 shows this variation in a dimensionless logarithmic plot, together with the limited data that are presently available. Good agreement exists except for 4 the 5-micron dust (D - 0®2 X 10- feet), which is cohesive and erodes differently from the granular dust. Visibility In order to calculate the visibility it is necessary to determine the quantity of dust entrained by the flow. In the present case, the dust is supposed to be entrained over a disk of radius 242/(k + 2) at a mean rate equal to half the rate at the location of maximum shear. The total rate of removal over this area is then 2 2 Tm (1 - rg) 2n ( k + 2 ) au8--O m where the factor -rm (1 - r g) represents the effective applied shear stress and
QgD cot a 2(('' Thrust _ T* - TM
e n h 2/(k + 2) L The fractional projected area of these particles is then determined and is written: Fractional projected area = CE
+ 2 o^c k+ 2
Now the amplitude of light is attenuated as e-CE, and the light intensity, therefore, which varies as the square of the amplitude, is given by
^ _ e-2CE
Io For the conditions of the experiments in the 60-foot-diameter vacuum sphere, r9 -= 0; and the intensity varies approximately as e
a log [Ihq2/(k + 2) -2D
This variation is shown in figure 5 where log e I I
is plotted against
for two sizes of particle and two values of the jet height. rn k + 2 Ca n Although data are insufficient to substantiate completely the theoretical expression, the agreement between theory and the limited experimental results appears good. APPLICATION TO LEM CONDITIONS In applying the theory to the landing problem of the Apollo lunar excursion module (LEM), account has been taken of the effects of gravitation and cohesion which, for most of the experimental data, were negligible. The effect of dust size on visibility is rather unusual in that, in the absence of gravitational and cohesive forces, larger particles tend to have the greatest effect on visibility, as may be inferred from figure 5. However, as the particles increase in size they become affected by gravity; that is, they become more resistant to erosion, and affect visibility less. Figure 6 is a logarithmic plot of LEM height against dust particle size and illustrates the effect of dust size on visibility. The solid curve shows the height at which visibility is 70 percent of that in the absence of dust particles. From this figure it is seen that visibility is affected most by particles of diameter about 10- 2 feet if the vehicle is at a height of 30 feet. For lower vehicle heights visibility is further reduced and the particles which cause the greatest loss in visibility are larger than 10- 2 feet. This type of information can be used to determine a "least visibility" envelope. This least visibility would occur if the lunar surface were such as to provide the worst kind of particles with regard to visibility throughout the descent. The least visibility envelope is shown in figure 7. Theoretically, the least visibility may be shown to be a function of a single parameter
ith2= µc RTc n
h k + 2 36g
and is given by
_ _ exp
4+ 1 + ^+
Y0_ _ o
4 4 1/2 +l 1+ 2 +
Figure 7 shows that visibility in the direction shown is definitely not impaired until the vehicle descends to a height of about 40 feet. Below this height visibility may be reduced appreciably and, below a height of 30 feet, may result in an unacceptable condition. Lateral visibility is generally better than that indicated in figure 7. The height of the cloud or spray may be calculated, and this, too, depends only on the single parameter g. Visibility in the vertical direction will not be impaired by the disturbance of dust; however, radiation from the hot exhaust behind the standoff shock may be troublesome. CONCLUSIONS The purpose of this paper is to provide a theoretical framework within which experimental results on dust erosion and loss of visibility may be placed and properly interpreted. A fairly complete theoretical model has been presented which can describe the erosion and associated loss in visibility for a vehicle landing on a dusty surface. These theoretical-model results agree with the available data, but further experiments are required before the results can be applied to the LEM vehicle with complete confidence. The theory supplies a complete set of scaling laws so that simulation of vehicle maneuvers near the surface may be carried out and properly interpreted in the 60-foot-diameter vacuum sphere. Results that show the least visibility as a function of vehicle height have been obtained. These results are particularly useful, inasmuch as they are virtually independent of lunar surface conditions.
1. Roberts, Leonard: The Action of a Hypersonic Jet on a Dust Layer. Paper No. 63-50, Inst. Aerospace Sci., Jan. 1963. 2. Stitt, L. E.: Interaction of Highly Underexpanaed Jets With Simulated Lunar Surfaces. NASA TN D-1095 1 1961. 3. Roberts, Leonard: Exhaust Jet - Dust Layer Interaction During a Lunar Landing. NASA paper presented at the Thirteenth International Astronomical Congress (Varna, Bulgaria), Sept. 23-29, 1962.
LUNAR LANDING HAZARDS
• REDUCED VISIBILITY • VEHICLE DAMAGE BY DUST PARTICLE IMPACT
• UNEVEN LANDING SURFACE
SURFACE FLOW CONDITIONS 100NON-C ONT.
LA MI N A R F LO W
W - I
7n 'NF k+2 h /—
ROUGH DYNAMIC FLOW
1 1 11.1111...'..' 1 ...............
VEHICLE REYNOLDS NUMBER, 7rrn µ c V/R Tc ——
3XI0-2 DUST DIAM.
h THEORY HEI LOCATION OF MAX. SHEAR (THEORY) t, sec 0—
10-2260,a EROS ION RATE, — 1 dy dt 3XIO-3- 125/,--'o ,/-— R Tc (M
2 60/.L !\
Y 2 .5 -
Fn 3xI0-4 1 1 HEIGHT, + 2 ' rh ^/ ^+2 k n' 5.0 1
h-= 10 rn EXP
L) DUST DIAM. LOG (10 260g e
h = 5 4- _ Fn
I .1 4_1
BASE PLATE EXPOSED
10 h (D^ 2 (T F n n 2 k+2 loo
EFFECT OF DUST SIZE ON VISIBILITY
LEAST VISIBILITY ENVELOPE
VISIBILITY, I All" \I .5
45 15 30 LEM HEIGHT, FT 5 h k+2 10 15
1 20 ^12— 2 (THRUST)2 4 ^r2 fc6 RT, 0-9 Jl
LUNAR RESEARCH FACILITIES
21. LOLA, THE LUNAR ORBIT AND LANDING-APPROACH SIMULATOR By William T. Suit and Ralph W. Stone, Jr. SUMMARY The design considerations for a lunar orbit and landing-approach simulator (LOLA) are briefly discussed. This device simulates the visual cues that man would encounter while operating a vehicle during the orbiting and landing-approach phases and other phases of the lunar-landing mission. The current status of this equipment is indicated. INTRODUCTION Until the advent of space travel, the standard practice has been to determine man's capability in handling flight vehicles by investigations under the actual conditions to be encountered. In tests of man's ability in this new field, standard practice must be abandoned, as the test would constitute the actual mission. Simulated conditions must, therefore, be provided for the required investigations. A lunar orbit and landing approach simulator (LOLA) is being developed at the Langley Research Center to investigate man's ability to perform control tasks related to the orbiting and landing-approach phases and other phases of the lunarlanding mission. Experience in Project Mercury flights has indicated that man is highly dependent on the visual sense when in a weightless condition. Therefore, LOLA will provide a realistic display of the visual cues that man would encounter while controlling a vehicle during the orbiting and landing-approach phase of the lunar-landing mission. The dynamics of the problem will be simulated by means of an analog computer. The purpose of this paper is to discuss the design considerations for LOLA and to describe briefly the equipment presently envisioned for this simulator. The present status of the simulator is indicated. DISCUSSION General Considerations The lunar orbit and landing-approach simulator LOLA was conceived to provide • visual environment adequate for testing man's ability to navigate and control • spacecraft in the vicinity of the moon. An artist's conception of LOLA (fig. 1) shows the equipment that will be necessary to provide this environment. This equipment consists of four models of the lunar surface, viewing systems to transmit views of these models to a display area, and a four-porthole display system where the views are presented to the pilot. The viewing systems shown have six ---- 171
degrees of freedom and are driven by computer signals. The pilot will observe the presentation and will take the control action that he considers necessary to accomplish the desired mission. The control signals are transmitted to the computer which, in turn., drives the camera system so that the pilot in effect is flying the camera system over the simulated lunar surface. With this device a pilot can perform, within the design limits provided, all the tasks related to lunar approach, orbiting, descent, landing approach, and the portion of take-off to orbit and escape consistent with the minimum altitude possible with the simulator. Thus the presentation to the pilot is the same as would be seen from the portholes of a vehicle operating in the vicinity of the moon. Lunar Surface Models The primary factors influencing the size, scales, and number of models for the simulator are; (1) A desired simulated altitude range from 322 km to 46 meters of the lunar surface. (2) The fact that the viewing optics could not practically come any closer to the models than 3/4 inch. (3) A requirement that the simulator be designed to study a wide range of possible orbits and descent trajectories. (4) The physical size of the models had to be limited to the extent that the models not only would be practical. to make but also that they could be housed in an existing structure. These requirements dictated the use of four models. The first model was necessarily a sphere so that complete orbits could be simulated. Twenty feet was chosen as a convenient and practical diameter for this model. The scale of this model was 1 inch = 14.5 km. Since the maximum altitude to be simulated was 322 km, a maximum travel for the viewing system of 24 inches was chosen. This simulates an altitude of about 322 km. The minimum altitude attainable with this model, which was 11.3 km, was set by the 3/4 -inch closest approach distance. This altitude range is indicated in figure 2. The model sizes and scales required for the other three models were determined by the different trajectories to be simulated. Several feasible descent trajectories are shown in figure 2. The model scales were set so that the extreme trajectories shown could be simulated without making the models too large to be practical. Once the scale for a model was chosen, the limits of travel of the viewing system and its closest possible approach to the model determined the altitude range possible for the particular model. The second model, map 1, has a scale of 1 inch = 3.2 km and covers the altitude range of 11.3 km to 2.4 km. Map 1 simulates a distance of 1,5+5 km and is 40 feet long by 15 feet wide. The third model, map 2, has a scale of 1 inch = 0.8 km and covers the altitude range of 2.4 km to 0.6 km. Map 2 simulates
a distance of 319 km and is 33 feet long by 25 feet wide. The fourth model, map 3, has a scale of 1 inch = 61 meters and covers the altitude range 0.6 km to 46 meters. It simulates a distance of about 26 km and is 34.9 -feet long and 22-feet wide. Map 3 is shown as just a dot in this figure. Before the actual map models of the lunar surface could be made, a landing area,had to be selected. Figure 3 shows the areas on the lunar surface covered by the models. The crater Alphonsus was selected as the landing site. Although there is a definite scientific interest in Alphonsus, Alphonsus also is a region surrounded by some of the most rugged mountains on the moon and confronts the pilot with an exacting navigational task. The first model, the sphere, rotates and can tilt to simulate orbital inclinations up to 15 0 . The sphere represents a band of the lunar surface from 60 0 north latitude to 600 south latitude. The second model, map 1, is a spherical section with a molded relief surface and is a scaled-up portion of the sphere. This model represents the area shown in figure 3. The third model, map 2, is also a spherical section with a molded relief surface and again is a scaled-up portion of the preceding model. This model represents the area indicated. The final model is a flat model and represents the immediate landing site. Model Lighting All four models that have been discussed are back-lighted. This lighting eliminates the difficult problems of shadows and uneven lighting caused by the viewing systems and support structures when other types of lighting are used. Also, no lighting equipment is placed around the viewing system. The backlighting approach becomes very desirable in consideration of the approach to within 3/4 inch of the surface with the viewing system. To give the molded relief models the proper appearance when back-lighted required that shadow patterns be painted on them. Viewing and Display System Considerations The models are to be viewed by two camera clusters mounted on the transport mechanisms illustrated in figure 1. The transport mechanisms as well as the sphere are actuated by computer signals. The viewing system transport mechanisms move to give the three translational motions of a spacecraft; namely, longitudinal, lateral, and vertical motions. The cameras themselves will have three degrees of angular freedom, which give the viewing systems six degrees of freedom relative to the simulated lunar surface. The primary viewing and display system will be a closed-circuit television system. The television system will consist of two camera clusters of four cameras each. One of the four cameras in each cluster furnishes the display information for one of the portholes in the display area. This display information will be presented to the pilot so that the view seen through the display system portholes will have the same appearance as the actual lunar surface under similar cockpit configuration and flight conditions. The television viewing system or simulated
six-degree-of-freedom spacecraft can then be controlled by the pilot through the computer so that the pilot will have control over his display and will actually be flying the viewing system over the simulated lunar surface. To give a continuous presentation with only two viewing systems and four models required that the two viewing systems view the models alternately. One camera group will view the sphere and then map 2, whereas the other will view map 1 and then map 3. When a descent is simulated, the system viewing the sphere will furnish display information to the pilot until the viewing system reaches its lower limit of travel. Before the limit is reached the second viewing system will begin scanning map 1. When the first viewing system reaches its limit of travel, the switch will be automatically made to the second viewing system. A similar switch will be made between map 1 and map 2 and then between map 2 and map 3 to provide a continuous display over the complete altitude range of the simulator. The display cockpit for the closed-circuit television display will have four portholes, each with a 45 0 field of view. Each individual television camera will correspond to a porthole in the display system, and the display for each porthole will have a 65 0 field of view so that the pilot may move his head slightly without seeing the edge of the display.
Present Status of the Simulator Now that the proposed simulator has been described, the progress to date and the existing and proposed equipment are discussed. Figure 4 shows the internally lit 20-foot-diameter spherical model shell, the 15 -foot by 40 -foot back-lit spherical segment shell for map 1, and one transport track and transport mechanism. This equipment is now undergoing acceptance tests. When these tests are complete, painted gores of the lunar surface will be attached to the sphere to provide a lunar model with painted features. The 15 -foot by 40 -foot segment will have a molded relief attached in front of it. The molds for this relief are presently being constructed. When the molded relief is complete, it will be painted so that its shadow pattern matches the shadow pattern of the sphere. The equipment required to complete the simulator is being acquired on contract or will be acquired in the coming fiscal year (196+). The complete simulator is scheduled for completion in a little over a year. Since part of the maps and a transport mechanism will be available before a television viewing system can be obtained, a 180 0 motion-picture camera-projector display system has been developed. The camera can be mounted on the transport mechanism and movies taken as this transport moves to simulate a programed trajectory. The motion pictures then can be projected onto a spherical screen giving a 2n solid-angle display. The pilot will have no control over his display but will be an observer. However, this presentation can be used to test man's ability to perform the observational tasks which would precede any control action and to determine his orbital ephemeris. Short experimental films with this 1800 system will be made in the near future.
17 4 ..
LOLA'S RELATION TO THE LUNAR MISSION LOLA, its configuration, and current status have been discussed® There is as yet one question to answer; that is, what information will be provided by this device and how does it affect the lunar mission? LOLA is a device that permits the study of the man-machine relation. LOLA should be able to provide information on those control tasks best performed by man and those best allocated to automatons. Thus, information will be obtained so that the most effective integration of man and machine can be achieved so that an effective and reliable control system for the lunar mission may be produced.
LUNAR ORBIT AND LANDING APPROACH SIMULATOR
APPROACH AND LANDING AREA ON
Page intentionally left blank
slow 22. DESCRIPTION OF A LUNAR-LANDING RESEARCH FACILITY
By Thomas C. O'Bryan
A research facility designed to study the piloting problems in the final phase of a lunar landing is presently under construction at the Langley Research Center. This facility consists of a gantry structure 250 feet high and 400 feet long from which a vehicle is suspended. A crane system supports 5/6 of the weight of the vehicle through servocontrol vertical cables while the remaining 1/6 of the vehicle weight pulls downward and simulates the lunar gravitational force. During research flights, the overhead crane system is slaved to move with the vehicle linear motions in order to keep the cables vertical. The establishment of requirements for a lunar landing will be accomplished by measuring the performance of test pilots conducting landing tasks.
A research facility designed to study the piloting problems of the final phase of a lunar landing is presently under construction at the Langley Research Center. Simulation with this facility starts at about the altitude where the Langley lunar orbit landing approach simulator stops. Because gravity on the moon is only 1/6 of that on earth, thrust levels for lunar operations are very low compared with those required for VTOL flight on earth. In order to produce horizontal accelerations for braking and maneuvering during lunar landing, large attitude angles (up to 30 0 or more) will be required. These angles are much larger than those experienced in earth landings and may impose serious visibility and attitude- and thrust-control problems. This facility is designed to study these problems with a vehicle similar to the Apollo lunar excursion module (LEM).
DISCUSSION OF FACILITY
The force vectors for a rocket-propelled landing vehicle for earth, lunar, and simulated-lunar gravitational conditions are shown in figure 1. The illustration of the simulated lunar gravity condition (g16) indicates that if a vertical force equal to 5/6 of the weight is provided, a vehicle will hover with a thrust equal to 116 of its weight and the same balance of forces will exist as depicted for the lunar hovering condition (g/6). In the case of translation, the application of the proper thrust again results in the same pitch angle and balance of forces on the vehicle.
The facility chosen to simulate the motions of a vehicle in a lunar gravitational field is depicted in figure 2. The gantry structure supports a traveling crane from which a vehicle is suspended. The crane system supports 5/6 of the vehicle weight through servocontrol vertical cables while the remaining 116 of the vehicle weight pulls downward and simulates the lunar gravitational force. During research flights, the overhead crane system is slaved to move with the vehicle linear motions to keep the cables vertical. A gimbal system on the vehicle permits the angular freedom required for pitch, roll, and yaw. The building shown at the edge of the gantry houses a control room from which the flight tests are monitored. The facility will be capable of testing vehicles weighing up to 20,000 pounds including a full-scale LEM model using its actual fuels. The vehicle will have six degrees of freedom in a volume 400 feet long, 165 feet high, and 50 feet wide. A catapult system in the first 100 feet of the structure length will permit horizontal test velocities up to 50 ft/sec and vertical test velocities up to 40 ft/sec at the initiation of a test. The system supporting 5/6 of the weight has been explored with the piloted scale model shown in figure 3. The suspension system followed the vertical motions of the model and allowed limited horizontal movement plus freedom in pitch, roll, and yaw. The system in this model performed satisfactorily and gave valuable data for the design of the servosystem for the full-scale facility. A vehicle designed as a research tool, not as an actual lunar lander, is being constructed for use on the Langley lunar-landing research facility, and, as such,.-it has been provided with a large degree of flexibility in instrumentation and control modes. A photograph of a 110-scale model of the vehicle is shown in figures 4 and 5. The gross weight of the vehicle is 10,000 pounds and includes a two-man crew and 3,300 pounds of fuel. The tubular steel framework of the vehicle, which houses the propulsion system, has conventional oleo struts attached to the four corners. The landing gear is designed for a sinking speed of 10 ft/sec with an additional increment of 6 ft/sec absorbed by aluminum honeycomb elements that crush after the oleos are fully compressed. The pilots' compartment is centrally located on top of the frame but is separated from it by a firewall. Since the vehicle is suspended from fixed points on the frame, some provision must be made to control the center of gravity as fuel is used. This is accomplished with a pair of vertically moving weights on the front and rear of the vehicle that are programed as a function of fuel consumed. The propulsion system, including the main motor and attitude motors, utilizes 90-percent hydrogen peroxide fuel. The main motors used for deceleration produce 6 1 000 pounds of thrust with a 10-to-1 throttling range. The motors are divided into two clusters of five each; the cant angle of each cluster is ground adjustable from 150 to 45 0 to allow study of the jet-blast problem. The attitude-motor thrust is ground adjustable to produce accelerations from 0.1 to 0.5 radian/sec 2 about all axes. In a typical flight test, including deceleration, maneuvering, and hovering, there is sufficient fuel for about 3 minutes of operation.
Since this is a manned system, its control and operation will be described in relation to the pilot. He flies from the right side of the cab with all control functions available to him. The left side of the cab has no controls or displays and is available for any revised control system that may be required to duplicate another vehicle. The pilot's field of view may be considerably greater than he will have in an actual lunar Lander, but the plastic bubble covering the pilot's compartment can be masked off to reproduce the field of view that an actual vehicle might allow. The pilot's instrument panel presents necessary controls and displays for visual flight. It is anticipated that the requirements for display will be developed as the simulation program proceeds. The pilot's thrust or throttle control, located on his left, the collective pitch handle of a helicopter. The throttle system adjustable in order to provide variations in thrust as a function tion. Attitude is controlled with a two-axis side-arm controller roll at the pilot's right and with conventional rudder pedals for is similar to is ground of stick posifor pitch and yaw.
Three modes of attitude control are available for the pilot's selection during a flight: attitude command, rate command, and acceleration command. Since the attitude-motor thrust is not adjustable in flight, thrust modulation as required in the various modes is obtained electronically with a pulse-widthmodulation system. Without going into great detail, it should be mentioned that safety has been given considerable attention in the design of the entire facility. For example, a flight test can be aborted automatically, or manually, by the pilot or by ground personnel in the control room. An abort will be initiated anytime that the vehicle has a velocity that exceeds its deceleration capability. An abort signal stops all systems and brings the crane and its lift system to a halt before the vehicle strikes the ground or any part of the gantry. A comparison of the research vehicle and the lunar excursion module is presented in figure 6. The research vehicle is not significantly smaller than the LEM, and the variations available in its systems will enable a simulation of the actual stability and control of the LEM. The establishments of requirements for performing a lunar landing will be accomplished by measuring the performance of the test pilots conducting landing tasks. In the first phase of the research program, which is planned to last 6 months, velocities will be generated by the vehicle without the aid of the catapult. During this first phase, thrust requirements for the main motor and the attitude motors will be investigated along with the attitude-control augmentation with and without failure modes. The second phase of the research program will involve catapult flights up to maximum velocity and is planned for a maximum duration of 6 months. Piloting techniques, visibility, and abort conditions will be the major items of investigation during the second phase.
CONCLUDING REMARKS This report describes the Langley lunar-landing research facility designed to study the piloting problems of a lunar landing with a vehicle similar to the Apollo lunar excursion module.
FORCE VECTORS FOR ROCKET-PROPELLED LANDING VEHICLE EARTH (g) HOVERIN6
SIMULATED LUNAR (gt6)
5 31*Z /.1 7 /1F/6
TRANSLATION-TO MOVE 100 FEET IN ABOUT 12 SECS
31*, A.17 F/6
g= 32.2 FT/SEC2
L- 2055 -1
LANGLEY LUNAR-LANDING RESEARCH FACILITY
COMPARISON OF RESEARCH VEHICLE AND A PROPOSED LEM
L -15 89 -A-
Page intentionally left blank
RENDEZVOUS DOCKING SIMULATOR By Howard G. Hatch, Jr.
A six-degree-of-freedom simulator designed for research on the piloted control of rendezvous docking has been constructed at the Langley Research Center. This system is fully operational and has been man rated. The first research program, which concerns the Gemini-Agena docking maneuver, is now in progress and the objectives are to demonstrate normal operation, investigate failure modes, and study target-lighting effects. Other docking programs have been scheduled and include studies for the Apollo lunar excursion module (LEM) and resupply docking problems for the manned orbital research laboratory (MORL). Although the simulator system was designed to study docking, it may also be used for other piloted maneuvers such as the touchdown phase of lunar landing.
Many studies concerning rendezvous in space have shown the feasibility of such an operation (refs. 1 to 3); however, the final docking phase of rendezvous has been of concern because it involves the actual contact of two vehicles in space. The facility described in this paper provides a realistic simulation of the docking phase of rendezvous allowing motion in six degrees of freedom. In addition to the docking maneuver, the facility may be used for many other studies in which it would be desirable to have six degrees of freedom.
DISCUSSION OF FACILITY
Description of Facility The six-degree-of-freedom rendezvous docking simulator at the Langley Research Center as shown by the model in figure 1 consists of an overhead carriage, a dolly, and a cable-reel system which provide three degrees of translation for the suspended payload, which in turn provides three degrees of rotation. Presently the rotation is obtained by a gimbal system that is driven by an electrohydraulic servosystem. The whole simulator can be driven in a closed-loop fashion through an analog computer to provide complete freedom of movement at the command of a pilot who can be located in the payload or at a remote-control location. Although the payload is suspended on eight cables, the number of cables and their relative attachment angles minimize disturbing swinging motions.
Performance Capabilities The system performance capabilities as shown in table I are as follows: In translation, longitudinal motion is obtained by the carriage which moves along a 210-foot track at velocities up to 20 fps with accelerations up to 8 ft/sec t , and lateral motion is provided by the dolly which can move 16 feet along a track attached to the carriage with rates up to 4 fps and accelerations up to 4 ft/sec t . The vertical motion is obtained by reeling and unreeling the cables through 45 feet at speeds up to 10 fps and accelerations up to 8 ft/sect. In rotation, the pitch and yaw axes are capable of angular accelerations of 2 radians/sec t and can achieve rates of 1 radian/sec. The roll axis is capable of angular accelerations of 8 radians/sec t and can achieve a rate of 4 radians/sec. With these rotational characteristics, the gimbal system is capable of simulating most of the manned spacecraft presently conceived.
Translational Drive For the translational system, four 30- horsepower mechanically synchronized d-c motors drive the carriage; one 10-horsepower motor drives the dolly; and one 30- horsepower motor drives the cable reel. Although the payload weighs 5,000 pounds, the vertical horsepower required is low because a nitrogen-hydraulic counterbalance system is used to support the payload at all vertical positions at essentially zero g as far as the motor is concerned. The motor, then, has only to accelerate the free-floating payload inertia through a 100:1 gear ratio. The carriage and dolly use a pinion and rack arrangement for traction while the vertical system uses an 8 -foot cable reel. The motors receive power commands from groundbased motor-generator sets which are in turn driven from the control center. The control center converts low-power analog output signals to the high-current signals needed for the drive motions. The translation system is position seeking, is accurate to one-half of 1 percent in each axis, has a time constant of 0.8 second for a 1-foot step command, and can respond to 1/2 cps for small incremental oscillations and 1/5 cps for 1-foot amplitude oscillations.
Rotational Drive The rotational gimbal system utilizes hydraulic motors to drive each of the gimbals. These motors are commanded by electrically driven valves which respond to the pilot's commands by way of the analog computer through an electronic network mounted on the support frame. The drive motors receive hydraulic fluid from a single pump unit that is mounted on the support frame behind the vehicle. The gimbal system can be detached from the supporting cables, placed on a ground stand, and operated as an independent unit for simulations that require only three degrees of freedom.
Safety Numerous safety features are incorporated in the system to prevent or to check uncontrolled motions. These safety features allow the system to be man
i C8 . %Isow
rated. At the control center, such features as velocity and acceleration limiters are used to prevent uncontrolled motion and, on the overhead system, overspeed switches are used as a backup to switch in regenerative braking if the controlcenter velocity limiter fails. In the event of a power failure, uncontrolled motions are checked by hydraulic arresting gears mounted at the limits of travel of the carriage, dolly, and cable-reel system. On the gimbal system, electrical stops are used to limit the angular motion, and microswitches are also installed on the pitch and yaw axes to cut off the hydraulic pressure if the angular limit is reached. The analog computer is limited so that output signals greater than the system limits will not be transmitted.
A number of projects have been planned for this facility. The first is a study of the Gemini docking maneuver. The objectives of this study are: to demonstrate the pilot's ability to perform the maneuver in normal operation, to investigate pilot control during simulated jet failures in order to determine operational techniques that would allow completion of the maneuver, and to study the effects of target lighting on pilot performance. Some of the other piloted projects which have been planned are: Apollo-LEM docking, MORL resupply docking, and lunar take-off and landing hovering studies.
A six-degree-of-freedom simulator that provides complete freedom of movement at the command of a pilot has been constructed at the Langley Research Center. The first research project scheduled for the simulator, the Gemini-Agena docking maneuver, is now underway. Future projects scheduled for the simulator include the Apollo-LEM docking and resupply docking of the manned orbital research laboratory. In addition, studies of the touchdown phase of lunar landing are under consideration.
REFERENCES 1. Brissenden, Roy F., Burton, Bert B., Foudriat, Edwin C., and Whitten, James B.: Analog Simulation of a Pilot-Controlled Rendezvous. NASA TN D -747, 1961. 2. Lineberry, Edgar C., Jr., Brissenden, Roy F., and Kurbjun, Max C.: Analytical and Preliminary Simulation Study of a Pilot's Ability to Control the Terminal Phase of a Rendezvous With Simple Optical Devices and a Timer. NASA TN D -965, 1961. 3. Brissenden, Roy F., and Lineberry, Edgar C., Jr.: Visual Control of Rendezvous. Aerospace Engineering, vol. 21, no. 6, June 1962, pp. 64-65, 74-78.
TRANSLATIONAL DIRECTION OF TRAVEL LONGITUDINAL LATERAL VERTICAL LENGTH FT 210 16 45 RATE FPS 20 4 10 ACCELERATION, FT/SEC2 8 4 8
ROTATIONAL ATTITUDE ANGLES PITCH ROLL YAW RATE, RAD/SEC 1 4 I ACCELERATION, RAD/SEC2 2 8 2
24. NEW DYNAMIC RESEARCH FACILITIES SUITABLE FOR SUPPORT OF APOLLO--LUNAR-EXCURSION-MODULE MISSION By D. William Conner SUMMARY Design features and areas of research are given for several dynamic research facilities now under construction which could be used in support of the Apollo-lunar-excursion-module mission. These facilities include the Langley lowfrequency noise facility and the vibration backstop, combined environment chamber, 55 - foot vacuum cylinder, and 60 -foot free-body dynamics facility of the Langley dynamics research laboratory. INTRODUCTION This paper presents a description of the following research facilities: the Langley low-frequency noise facility, and the Langley dynamics research laboratory including the 55 -foot vacuum cylinder and the free-body dynamics facility. LANGLEY LOW-FREQUENCY NOISE FACILITY Figure 1 shows the general configuration of the environmental research facility for large-scale testing in intense noise at subaudible or near subaudible frequencies below 50 cycles per second. It consists of a cylindrical test chamber with a loudspeaker located in one end wall. At the opposite end of the chamber is a movable wall which maybe positioned for tuning purposes or may be removed completely for installing test specimens or for open-chamber testing. Behind the loudspeaker is a chamber in which the speaker driver is located. Adjacent to the test chamber is a control room with observation windows, driver controls, and instrument recording equipment. Pertinent features of this facility are given in the following table: Chamber: Test diameter, ft Test length, ft . Type. . . . . . Noise level, db . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 . . . . . . . . . . . . . . . . . . . . 20 . . . . . . . . . . . . . . . Reverberation . . . . . . . . . . . . . . . . . . . . 165
Exciter: . . . . . . . . . . . 1 to 50 Frequency, cps . . . . . . . . . . . . . . . . Spectrum . . . . . . . . . . . . . . . . . . . . . . . . . Discrete or random
Force, lb ® ® . . . . ® . . . . ® . . . . . . . a . . ® . . . . . . ® .
Stroke, ft . . . . . . . ® . . . . . > . . . . . . . ® . . . ® . ® .
14 Speaker diameter, ft .I . . . . . . . . . . . . . . ® . ® . . ® . ® . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Hydraulic Type As shown in figure 2, the test chamber is of heavy steel construction and is mounted on a large pile-supported concrete base. The loudspeaker is of aluminum honeycomb construction. Both test chamber and loudspeaker are designed to keep structural response frequencies out of the facility operating range. The low-frequency noise facility (fig. 2) is equipped for intense noise environmental tests of a spacecraft during boost such as an Apollo or the undeployed lunar excursion module (LEM). The tracks for the movable wall will be useful in positioning and securing the vehicle. Included in tests will be straingage and accelerometer measurements of vehicle structure and internal noise measurements. The onboard systems can be operated and onboard astronauts can perform assigned functions. Random noise environment simulating noise levels and spectra shapes of concern in the flight mission can be generated during the tests. In actual practice the Apollo or lunar excursion module would fit completely within the chamber. Vehicle structures too large to enter the chamber would be placed just outside the mouth. The various areas of research expected to be carried out with this facility are: manned spacecraft, launch vehicles, psychological and physiological effects, and transmission and propagation. This facility has the capability for experiments concerning the effects of low-frequency noise on the psychological and physiological behavior of animals and humans. Low-frequency-noise transmission data can be obtained by using test techniques described for manned spacecraft for various structural materials and structural shapes. In studying the propagation of low-frequency noise, the tuning wall would be completely removed and the facility would be used to generate noise for a range of discrete and random frequencies. Acoustic measurements would be made in the surrounding area and propagation data obtained for a range of conditions provided by prevailing weather. The facility is scheduled for completion by November 1963 and should be ready for research use about January 1964. LANGLEY DYNAMICS RESEARCH LABORATORY The dynamics research laboratory is shown in figure 3. Its features are as follows: Backstop: Vibration exciters to 30,000 pounds Combined environment chamber: 6 -foot diameter by 6 -foot length 10-8 torr LN2 cryowall 2 19 000-pound shaker
vacuum cylinder: 10-1 torn 50,000 g-pound whirl table with exciter
free-body dynamics facility: Air-bearing support Solar simulator Infrared planetary target
The central portion of the laboratory houses a large 36- foot-high backstop of about 1,000 tons total weight. It will be used to investigate the dynamic response of large models and components. The building doors roll down into the ground and roof hatches roll back to permit the mounting of structures which extend above the roof level. Shown in a cable harness support is a 110-scale dynamic model of the Saturn V which will be investigated in 1964. A number of vibration exciters are available, the largest having a force output of 30,000 pounds. The backstop has been completed and all the shaker equipment should be operational within a few months. The combined environment chamber shown in one corner of the building has a working volume about 6 feet in diameter by 6 feet in length and is mounted on a backstop. It can be evacuated to 10- 8 torr and has a LN 2 cold wall and quartz-lamp radiant heaters. A 2,000-poundoutput electromagnetic shaker projects through the floor of the chamber. Combined environments of vacuum, extreme temperature, and vibration can thus be imposed on test specimens or equipment. Acceptance tests of this chamber are currently under way and it should be ready for research use by September 1963. The 55 -foot vacuum cylinder located behind the central laboratory building is normally separated into two areas. Above the floor grating is an open area which is to be used for dynamic studies of large space structures, such as solar collectors or full-size prototype of the lunar excursion module shown here. It can accommodate test objects up to a 55- foot-diameter sphere. A 20- by 20-foot sliding door at grade elevation provides easy access. Below the floor grating is located a 50,000 g-pound whirl table. This whirl table, limited to a maximum of 1008 or 2,000-pound payload, is equipped with a hydraulic shaker so that test objects can be subjected to combined vibration and acceleration in a vacuum. For tests of large objects on the whirl table, the flooring can be completely removed. An airlock and quick inbleed valves have been provided for future man-rating but no life-support equipment has been procured. No temperature control has been provided. The chamber will initially be equipped with mechanical pumps which will provide a vacuum capability of 10- 1 torr. Provisions have been made for adding diffusion pumps when funds permit to obtain a vacuum of 10- 4 torr. Contractual work should be completed by October 1963 and the chamber should be ready for research use by January 1964. The free-body dynamics facility is designed for checkout of space-guidance systems which utilize solar sensors and horizon scanners. It consists of a 60- foot-diameter sphere equipped with a frictionless,air-bearing support for the guidance-systems package. A high-quality solar simulator projects a 24-inchdiameter light beam by means of lenses and mirrors toward the center of the sphere. This optical system is mechanized to rotate once per day. Rotating
about the same axis is a framework immediately inside the shell of the sphere upon which is mounted a circular infrared planetary target. The diameter of this heated target can be varied to simulate varying distances to a planet. The rotation of both the solar beam and the planetary target can be independently controlled at very precise rates of rotation. Thus a guidance system using a solar sensor and a horizon scanner can be thoroughly tested. With the air bearing operating, the sphere can be evacuated to a pressure of 1 millimeter of mercury. The scheduled completion date for this facility is February 1964, and it should be ready for research operation by May 1964. The areas of research covered by this dynamics research laboratory may be summarized as follows: dynamic response of large models and components, effects of combined environment such as vibration, vacuum, acceleration, and temperature, and performance of guidance and control components.
A description of several research facilities which could prove useful to the lunar-excursion-module mission has been presented. Features of these facilities together with the areas of research intended for coverage have been pointed out.
LORI-FREQUENCY NOISE FACILITY
TEST CHAMBER S DRI SPACECRAFT SPECIMEN TUNING CHAMBEI II CONTROL (J \ROOM —MOVABLE TUNING WALL DRAM TRACKS
LOW-FREQUENCY NOISE FACILITY
DYNAMICS RESEARCH LABORATORY
T VACUUM .,YLI NDER
IRL BLE COMBINED IV I RONMENT CHAMBER
LUNAR-GRAVITY SIMULATOR FOR FULL-SCALE IMPACT TESTING By Robert W. Herr
Two subjects relating to the lunar-landing problem are discussed. The first part of the paper deals with a combined dynamic-model and analytical program which is currently underway to determine the effects of pertinent variables on the stability and structural integrity of lunar-landing sygtems. A one-sixth scale model of the lunar excursion module (LEM) is being constructed for the experimental portion of the program. The analytical studies will utilize both digital and analog computer programs. The second part of the paper describes a proposed facility suggested by George W. Brooks (ref. 1) for impact testing of full-scale lunar landers in a simulated lunar gravitational field. In this scheme, the lunar lander is dropped onto a relatively massive impact platform which is accelerating downward at
earth g, thereby giving a relative acceleration between the test vehicle and
g. The advantage of this system is that the test vehicle 6 is completely free of encumbering cables and is thus not subjected to varying gravity forces associated with other techniques. impact platform of
This paper discusses two subjects relating to the lunar-landing problem. The first of these constitutes a generalization of the landing dynamics problem discussed by Ulysse J. Blanchard in paper no. 5. The second part of the paper deals with an impact simulator which can be used for conducting large-scale and full-scale impact tests in simulated lunar and planetary gravitational fields.
LANDING DYNAMICS STUDIES
Experimental Program In connection with the landing dynamics studies, the Langley Research Center has currently underway a combined dynamic-model and analytical program to determine the effects of pertinent variables on the stability and structural integrity of lunar-landing systems. Gravity enters into the scaling of replica models for simulation of landing dynamics by virtue of the fact that the product of a characteristic length and the acceleration due to gravity must remain a constant. Hence, dynamic similarity
during free-fall drop tests on the earth's surface is achieved only if the replica model is one-sixth the size of the lunar vehicle. The general configuration of the models being constructed is shown in figure 1. This is a generalpurpose model which is designed to facilitate the investigation of a wide range of variables. The mass, mass center of gravity, and moments of inertia may be readily adjusted over a wide range, including the values currently anticipated for the lunar excursion module (LEM). The landing gear illustrated in figure 1 is a four-point inverted-tripod configuration, but provisions are being made for the testing of any number of legs from three to six. The initial landing gear will contain crushable shock-absorbing material in each of the lower V-struts, as well as in the upper strut, to enable the legs to absorb radial and side loads. Compression springs may be added in series with the crushable absorber to investigate the effect of elasticity on the stability boundaries. For a given model configuration, the primary emphasis in the drop tests will be the determination of the most critical combination of linear and angular touchdown velocities, attitude, and landing-surface characteristics. One of the important objectives in this respect is the determination of whether the most critical stability conditions involve essentially a two-dimensional pivoting about the forward gear or whether considerable three-dimensional motion is involved, such as would be encountered on a sidehill landing. The results of these tests will have a substantial effect on the complexity of the mathematical model required to duplicate the experimental stability boundaries with a reasonable degree of accuracy. Another goal of the experimental program is to determine the importance of including fuel-sloshing effects on the touchdown stability boundaries. Analytical Programs Two separate analytical programs dealing with landing dynamics are now underway. One of these utilizes an analog computer, while the other is being programed for a digital computer. The analog-computer study concerns a two-dimensional mathematical model of a lunar lander impacting on two struts with either linear or nonlinear elastic and damping properties. This parametric study is being pursued to establish the relationship between the peak decelerations, rebound, and stability for various strut damping and spring forces and for various initial velocities and touchdown attitudes. The computer produces time histories of the acceleration, velocities, and displacements of the lander body and struts. These outputs are then combined on an oscilloscope (fig. 2) in such a way as to display pictorially the stability of the configuration and permit rapid construction of stability boundaries. 200
The objective of the digital program is to derive and solve equations sufficiently descriptive of the LEM landing dynamics to facilitate evaluation of the design. A three-dimensional analysis would be an order of magnitude more complex and time consuming than a two-dimensional analysis® Intuitively, it would seem that a two-dimensional analysis would be sufficient to encompass the most critical landing conditions, and the study is proceeding on this assumption until such time as the model experiments prove otherwise. The stability of rigid bodies can be expressed in terms of general geometric parameters in a fairly simple manner. A part of the digital program is to determine to what extent the stability criteria developed for rigid bodies are applicable to the landing of elastic deformable bodies.
Although model testing and mathematical analysis are important tools for indicating the worth of a landing-gear configuration, final proof of the design can best be established by drop tests of the full-scale vehicle. In order to maintain dynamic similarity during impact testing of a fullscale vehicle, it is necessary that the lunar gravitational field be simulated. This simulation may be accomplished by various techniques, most of which utilize a cable attached at the center of gravity of the test vehicle to support a part of the mass of the vehicle. George W. Brooks (ref. 1) has suggested a scheme, depicted in figure 3, in which the test vehicle is completely free of encumbering cables and is thus not subjected to the varying gravity forces associated with other techniques. If it is desired to simulate the lunar gravitational field, the relative acceleration between the test vehicle and the impact platform is adjusted to the value of the lunar gravity. Since the test vehicle is under the influence of 1 g, the impact platform must be constrained to drop at earth g. 5 This can be accomplished by unwinding the platform support cable from a shaft
6 g downward acceleration of the platform and thus gives the desired relative accelg. Another method of achieving this effect is to replace the inertia 6 wheel with a counterweight. For full-scale tests of lunar landers, the mass of the counterweight would be about one-eleventh of the primary mass, which consists of the impact surface and the supporting structure. Other reduced-gravity fields may be simulated by merely changing the mass of the counterweight, or for the case illustrated in figure 3, by changing the inertia of the flywheel. The impact surface is enclosed within a cage and may be tilted to simulate uphill or downhill landings. The test vehicle is given the desired horizontal impact velocity by a trapeze arrangement located at the top of the cage. The vertical impact velocity is controlled by the height of the trapeze above the impact platform. When the test vehicle is released from the trapeze, it falls at 1 g until it impacts the platform® Just prior to impact, a brake is released 201 eration of
affixed to a flywheel. Proper choice of the flywheel inertia results in a
on the flywheel shaft, allowing the cage to accelerate downward at
g. At the 6 end of the test time, the flywheel brake is reapplied to bring the cage to a halt.
To establish the feasibility of this scheme, a small-scale model of the reduced-gravity simulator has been built and tested. The combined weight of the g is platform and cage is approximately 700 pounds. The drop distance at 5 16 feet, giving a test time of a little over 1 second. The models used in the study were not scale models but served well to indicate the markedly reduced impact stability to be expected in a lunar gravity field. The landing-gear spread on these models was about 1 foot. It was found that the drop time of 1 second was ample for determination of the stability of these small models. If a larger facility were to be considered for reduced-gravity stability testing of a full-scale vehicle the size of LEM, a considerably longer test time would be required. Calculations indicate that a minimum of 4 to 6 seconds would be required to determine the stability of a vehicle of this size. .Test times of this order suggest that it might be desirable to incorporate a pilot in the test program and determine his ability to extend the stability boundary by applying corrective control torques immediately after touchdown. The total drop height required, including a 2g braking distance, is approximately 20 times the square of the test time. Thus, for a 4- second test period, 320 feet of drop height is indicated. An additional 2 seconds of test time raises this to 720 feet, a height for which the side of a cliff might serve as a test site. For these drop heights, it would become necessary to program the flywheel inertia or the mass of the counterweight in order to counteract the increasing aerodynamic drag during the drop. Considerable simplification of design with little loss of usefulness could be achieved by limiting the test time to 1 or 2 seconds. This should be ample to encompass the initial impact and energy-absorption phase. Proof-testing of the gear under realistic maximum load conditions would thereby be achieved with a total drop distance of 20 to 80 feet. With appropriate instrumentation of the test vehicle, the kinetic energy remaining after energy absorption could be computed and used in a simple rigid-body analysis to determine the stability of the vehicle.
1. Brooks, George W.: Techniques for Simulation and Analysis of Shock and Vibration Environments of Space Flight Systems: Experimental Techniques in Shock and Vibration, Will J. Worley, ed., ASME, c.1962, pp. 93 -105.
LUNAR-LANDING RESEARCH MODEL
„L)-1%, Figure 1
PICTORIAL DISPLAY ON OSCILLOSCOPE
SCREEN LANDER BODY
TRAPEZE TEST VEHICLE IMPACT PLATFORM
This action might not be possible to undo. Are you sure you want to continue?