R. & M. No. 2854

A.R.C. Technical Report


72 MARlss~
L I B R A . ~-~':~Y







~. . . .




L .............................. t.........................

The Critical Whirling Speeds and Natural
Vibrations of a Shaft Carrying a
Symmetrical Rotor


(Eng.), A.F.R.Ae.S.

Crown Copyright Reserved






The particular system investigated is essentially a rotor mounted on a cantilever shaft having an additional lateral support. COMMUNICATED BY THE PRINCIPAL DIRECTOR OF SCIENTIFIC RESEARCH (AIR).~ :oR . .. There is some discrepancy between calculated and experimental results for the vibration of the non-rotating system.. and preliminary experiments have been carried out on a model rig s. and these deflection forms were compared with that of the non-rotating system due to a static load at the point of attachment of the rotor.A.. Report Structures 97. and the moment of inertia of the rotor can be varied without changing its mass. was obtained experimentally. The dynamics of the system. A. . . The critical whMing speeds and natural vibrations of a single shaft ~flexibly supported and carrying a flexible rotor of appreciable moment of inertia have been investigated' and good agreement has been obtained between experimental and cMculated results for the rotating system.. . The lower natural frequency of vibration of the non-rotating system is shown to decrease as the inertia of tile rotor increases. 1.Sc.--The experiments described in this report are part of a programme of model experimenfs-'-d~signe---=dto~.5 (1919). 1 . . Flexibility of known value is introduced symmetrically at the lateral support. and for the non-rotating system vibrating naturally. It is shown that for this particular system there will be one critical whirling speed which increases as the inertia of the rotor increases.... These circular vibrations are exponentially stable and a critical whirling speed occurs only when the frequency of the forced vibration due to rotor unbalance coincides with a natural frequency of vibration of the rotating system. which has four flexible arms enabling i t t o distort out of the plane of rotation. !i ..The Critical Whirling Speeds and Natural Vibrations of a Shaft Carrying a Symmetrical Rotor By E. the whole having degrees of flexibility introduced additional to that of the shaft itself. the forms of vibration considered involving only shaft bending without torsion. 1951.--The problems associated with shaft whirling have already been discussed b y the author in a previous report 1. . The whole system has kinetic symmetry. DOWNHAM. . . . . -Vt:~-2 . were investigated theoretically using methods suggested by Morris "a and based on the Jeffcott theory of shaft whirling ~. confirming the fundamental theory of shaft whirling. when vibrating and whirling. The effect of tile inertia couples d u e t o the rotor.a. * R. For a particular rotational speed of the rotating system the natural vibrations will be circular. 2 854 December. B.:"'~'~'": Summary. The experiments described in the report are part of a programme of model experiments 1 designed to establish an accurate method for calculating the critical whirling speeds of complex systems. . establish an accurate method for calculating the critical whirling speeds of complex systems.~-'. MINISTRY OF SUPPLY Reports and Memoranda IVo. received 16th April..R.. which is thought to be due to the operational characteristics of the flexible bearing. Flexibility is also provided in the rotor. I 9 5© ..S. Introduction. .. on the deflection forms for the whirling system.F. there being four resonant frequencies which vary with rotational speed. "i .E. (Eng.). • " ~ n "1 ~":' ~: ". .Ae.

a and c respectively. .g. and another allowing for the flexibility of the arms.~ sin ~t -. The positions of the principle axes of the rotor in a general displacement of the system are represented b y GA. . 0)3 respectively and are obtained by summing the components of the angular velocities #. and the results compared with measured critical whirling speeds and natural frequencies. .1. . 1 the line OZ is drawn through 0 the point of attachment between the rotor and the shaft parallel to the centre-line of the bearings. Thus 0)1 ---. The positions of these axes is determined b y imagining first a small rotation # in the xz-plane about the y-axis. Analytical Treatment. c (c being the polar moment and a the moment about a diameter in lb in. then a small rotation 0 about the displaced x-axis and finally a rotation ~t about the axis GC. especially when the measured natural frequencies of the system were remote from the natural frequency of the rotor arms. 6 and f2 in the planes BCG. In Fig. .--2. and hy = a0)1 sin ~t + a0)2 cos ~t + c0)8 0 2 . . BG and CG are represented b y 0)1. 0)2.a~o~sin ~t + c0)3 • .The natural frequencies of the non-rotating system and the-critical whirling speeds were calculated for different amounts of rotor inertia. thus causing a change in the mode of vibration and an increase in the natural frequency. A comparison between calculated and measured frequencies gave good agreement. GB and GC. . .1. the inertias relative to these axes being a. sec ~ units) are considered.a0)1 cos ~t -.--2. This is considered to be due to the outrigger bearing imposing a greater constraint on the non-rotating system when vibrating transversely than on the rotating system when whirling. . These frequencies were then measured experimentally. The rotations ~ and 0 are assumed to be small which would be the case in practice and ~ is the angular speed of the shaft. O Y and OZ. The instantaneous angular velocities about the axes AG. ---. It is concluded therefore. . (2) . (1) if2 The angular momenta due to the angular velocities of equations (1) are respectively aol. Following these investigations the natural frequencies of the rotating system were calculated over a range of shaft speeds. .6 cos ~t 0). ACG and ABG respectively. a0)2 and c0)3 and therefore the component angular momenta in the directions y to z and z to x are given b y h. accurate calculations of the critical whirling speed and natural frequencies can be made using the theoretical treatment described. of the rotor assumed to be displaced an amount h from O' i n the plane X O Y . . . . . = # cos ~gt + $ sin ~t 0) 3 ~ ~ . The point O' represents the displaced position of 0 having co-ordinates X and Y relative to the major (fixed) axes OX and OY. Theoretical I~¢vestigations. and it was shown that the allowance for flexibility in the rotor arms increased the accuracy of the calculations. The whirling condition. The motion of the system is considered relative to the three axes Gx.1. . one assuming the rotor arms to be rigid. 2. The point G represents the c.In this case the dynamics of a shaft (Fig. Two sets of calculations were made. t h a t providing the constraints in a system are known. 1) supported in bearings of symmetrical stiffness carrying a load W of mass m and of appreciable moments of inertia a. Gy and Gz drawn parallel to the major axes OX. . The comparison gives good agreement between calculated and measured critical speeds but the calculated natural frequencies of the non-rotating system were approximately 10 per cent below those measured. . a.

therefore the equation m(c - a)~t? ~ - [(c - a)¢1.hy ---= -.. Also X0 a n d 4o become infinite when Q -. Y. (8) . . ..~]4~gllQ2X0 "~-. . .m [ Y + h~92 sin (t?t + ~)] in the direction of the Y axis . # and 0. . .]/Q and 4o = . . . (10).m[52 + hi? 2 sin ( g t + V)]Yn -. ..a ) z ~ s g ~ o = . (a) An inertia force -.(aO . . . . . . . t h e n only one value of ~92 given by e q u a t i o n (13) will be positive and hence there will be one critical speed only. .c 9 ¢ ) z ~ t . . . . .my~lht? ~ . (13) will give the critical whirling speeds [where A = (Y11¢11-- Zl19)] . (3) .11 qs0 ----.m[~ ~ + h 9 2 sin ( g t + ~)~zll . (4) The co-ordinates of G relative to the axes O X and O Y will be X h cos (~gt + ~) a n d Y -.. . . .. . .. . .-. .(a# + c~9~) about an axis t h r o u g h O' parallel to Gy in the direction z to x .ct?~b) about an axis t h r o u g h O' parallel to G x in t h e direction z to y .m[R + h~Q 2 COS ( ~ t .( a # O = . . . .(aO .m [ X + hi? ~ cos ( g t + ~)] in t h e direction of the X axis .. . .. hy = a # . a t • • • • • • • n • O • • • • • o • • . If c is greater t h a n a which is usually the case.A[_ ~])~Yll . . . ... (6) (c) A n inertia force -. . (10) (11) (12) where Yn is linear deflection due to unit load at the point of a t t a c h m e n t ¢11 Angular deflection due to unit couple at the point of a t t a c h m e n t z~1 Angular deflection due to unit load at the point of a t t a c h m e n t = linear deflection due to unit couple at the point of a t t a c h m e n t A particular solution of equations (9).m z n h D 2 from which Xo = - mhg~[(c - a)(y~l¢~l - z~2)~ ~ + y~.(a0 -.h ~ 2 COS ( g t + Cff20)Zll + ~)IZll . . . .. .h sin (zgt + ~) respectively where ~ is the angle between O'G and G A and (tgt + ~) is the angle between O'G and G X . (5) (b) A couple due to the change of angular m o m e n t u m -..a0 + c~gq~.. . in these equations gives • ( .(6~¢ + C90)~11 Y = -. . (11) and (12) will be of the form X = X0 cos (zgt + V) Y = Xo sin ( g t + v) = q~0 cos (zgt + V) 0 ---. . . . - myll]~ ~ - 1 = 0 . AI 3 . . . It follows therefore t h a t O' will describe a circle about the axis O Z with steady motion. .c9¢)¢11 . . . . .. .q~0 sin (tgt + V) and substituting for X./ c ~ 9 0 . = -. . (7) (d) A couple h. . . . .. . . .Substituting for ~1. (9) The equations of motion are therefore X = -.a ) ( ~ l l ~ 2 -7[.J'J~[X -7[. . 0 = -.0..m h ~ 2 z n / Q where Q = - m(c - a)(yl~¢11 - zl?)~ ~ + [(~ - my~3~ ~ + a)¢1~ - 1. Since h is small it m a y be assumed t h a t the following forces will act on t h e shaft at 0 ' . . . = -. .ncy~f2 ~ + 1)Xo + (c . ~o~ a n d ~3 from (1) in (2) gives h. .. . .[(C .

The effect of rotor inertia on the natural vibrations of the non-rotating system.1.(a4. (14) It may be shown that there are four real roots to this equation two of which are positive and two negative.2.aco2 -k C~(D)Zll40 : and - mz~o~xo + [ ( .1. .2. C611/2co -~.3 in conjunction with the flexibility coefficients obtained from Appendix I. .--Using equation (16) section 2.1. when the forced circular vibration due to out. 2. yi. .2.~ Inserting these Values in equations (9). . .. (11) and (12) with h put equal to zero gives (-. . :Eliminating Xo and 40 from these two equations gives t h e frequency equation maA~o' -.--Using equation (13) section 2.1). .1. i..--The natural vibrations of the rotating system are next considered.- 2 .1. . . For these there will be a solution of the form X = X0 cos (~ot + ~) 4 = 4o cos (~ot+ ~) Y = X0 sin (cot+ ~) . (15) The lower of the two frequencies obtained from equation (15) is considered to be the most i m p o r t a n t for the purpose of these investigations and is given approximately by 1 azll ~ oa~--my~ + Yl~ + a~A Zll 2 myll~ . The negative roots correspond to circular vibrations in the opposite direction to the rotation of the shaft and the positive roots to circular vibrations in the same direction. 2.--2. The natural vibrations of the rotating system.--The frequencies of vibration of the non-rotating system are obtained by substituting /2 = 0 in equation (14)which gives maA~o 4 (a¢1~ -k myra)co2 -k 1 = 0 from which 1 (my~ + a61~ ± ~/[(my~ + a¢~) ~ -. 1 per cent of the more accurate solution obtained from equation (15).. 4 .a ~ + 0 c/2~)¢~1 + 1]~0= 0.2.1 in conjunction with the flexibility coefficients obtained from Appendix I the critical whirling speeds were calculated as in 2.2. Calculated Results for the Experimental System.of-balance has the same frequency as a positive natural vibration of the rotating shaft.0 . . The whirling speed occurs when co = ~9. . . ..2.mcAgw a -. (i0). The effect of rotor inertia on the critical whirling speed of the system..3.1. 0 = 4osin (cot + ~) ~ which constitute circular vibrations of amplitude X0 and 40 an'd frequency ~/2~. This means that for any value of ~9 there will be four natural free vibrations which are circular. 2. and A being positive and c > a. . + my11)eo a -~. appropriate to the case where the forcing due to the out 0f balance h is absent...e. .4maA] co ~ . (16) The solution given by equation (16) for a particular case has been found to be within 0 . ¢i. • .2. .my11~o~ + 1)Xo + (-. the natural frequencies of the nonrotating system were calculated for the particular range of rotor inertias investigated in the first serms of experiments (section 3. .2.1...1 -=. The natural vibrations of the non-rotating system.

Whirl forms and amplitudes at the free end of the shaft were recorded photographically and shaft speeds were recorded electronically and synchronised with the recording of the whirl forms. The flexibility coefficients were deduced from this measurement by the methods of Appendix I and used in the theoretical calculations of section 2. In order to facilitate the computation this equation is written in the form m a A co~ t2 = (aq~il @ myii)co 2 @ 1 cco(mAco~ _ ¢~) - - . . The deflection form of the shaft itself under the various conditions tested was obtained by illuminating the shaft and using photographic methods of recording. .3 were made with the assumption that the rotor arms were rigid.-t. Dimensional details of the rig relevant to the calculations are given in Appendix II. . The correction for arm flexibility is then made by substituting [611 + #a~/2] for $~ in the frequency equation (17) w h e r e / ~ : y~/l~ ~. . long and was operated as a cantilever with a flexible outrigger bearing of symmetrical stiffness placed 10 in.2.inca f2~o8 -. . amplitude-frequency curves have been plotted (Fig. From these curves the critical whirling speeds for particular values of rotor inertia Were obtained and are compared in Fig. . The model rig described in a previous report 1 was used for the experimental investigations. each arm carrying a brass weight weighing 0. maAco 4 -. The calculations of 2. 5 . An approximate allowance for the flexibility of the arms was made to the calculations using a m e t h o d propounded by Morrlst The natural frequency of the arm was measured and from this measurement and a knowledge of the mass of the arm and its tip load the flexibility coefficient y~ was estimated.(a~l~ + myli)co 2 + c~[2co + 1 = O. (17) When the rotational speed ~9 : 0 the frequencies of vibration of the rotor are given by ± ~ and 4. Details of Experiments. 3. dial gauge.~/~--A and o~ : c -. a The :family of curves given by plotting co against D was obtained by substituting appropriate values of co in equation (17) above and solving for ~2. . . of the non-rotating system. . The position of the brass weights could be varied along the arms so that the inertias of the rotor could be changed without affecting its mass. deflections under load being measured by means of a 0. 3 with the calculated critical speeds of section 2. The shaft used was L-in.1. The arrangement of shaft and rotor is shown in Fig. 7. 3.1.1 lb. .2. from the fixed end of the shaft. Details of Model Rig and Experimental Techniques.2.--The natural vibrations of the rotating system are given by the four real roots of equation (14) section 2.~ where ~ and r2 are the values of co2 which satisfy the equation maAco 2 ~ (a$x~ + my~)co ~ + 1 = 0. A foura r m rotor was fitted to the free end of the shaft to simulate a four-blade propeller..1. Critical whirling speeds of the system and the natural vibrations.--3. Experimental Investigations. which were later compared with the experimental results. Before carrying out whirling experiments the rotor inertias were measured using the unifiler suspension method and the spring constants of the svstem were obtained by direct loading.--From measurements of whirl amplitudes and shaft rotational speeds.2.4. viz. 2. .2.2. If co is plotted against ~9 there will be four asymptotes given b y co = 0.2.2. The arms were made from screwed rod.31 The natural vibrations of the rotating system. 2) showing the effect of varying the inertias of the rotor. l~ being the length of the rotor arm. The effect of flexibility of the rotor arms on the natural vibrations of the rotating s y s t e m .--3.0005-in. co ---.2. diameter and 12-in.2..

The Critical Whirling Speeds of the System and the Natural Vibrations of the Non-rotating System. Although the excitation was in a lateral direction. In Fig.--From Fig.e. when vibrating transversely (non-rotating system) and when whirling were recorded and are compared in Fig. but there is some discrepancy between measured and calculated natural frequencies of the non-rotating system. I t was not practicable to record the three deflection forms at the same tip amplitudes and for the purpose of comparison the measured forms were adjusted accordingly. The vibrating and whirling deflection forms were recorded with maximum possible rotor inertias.p. to a value well above the critical whirling speed of the shaft. The frequencies of the natural vibrations of the rotating shaft were obtained at various speeds of shaft rotation by running the shaft at a particular speed and exciting it over the frequency range of the exciter . 4 by the increased curvature of the transverse vibration form compared with the static deflection form. four frequencies are obtained in divergent pairs from the original two. When the shaft whirls the inertia couple theoretically acts in opposite phase to the displacement of the shaft thereby increasing the critical whirling speed : again this is shown in Fig. An inductancetype proximity pick-up was used to investigate the resonant frequencies of t h e shaft. The deflection forms of Fig. 4. a = 0. 3. 4 show clearly the effect of the inertia couples arising from the appreciab'le inertia of the rotor. The rotating shaft was excited b y m e a n s of a motordriven crank elastically coupled to the shaft.e. The Natural Vibrations of the Rotating System.2. The range of shaft speeds over which investigations were carried out was chosen to cover speeds from zero r.1. 3 it is seen that the calculated values of critical whirling speeds agree closely with those measured. When ~ is zero there are four roots 6 .1. the difference being approximately 10 per cent over the range of rotor inertias investigated.9. either in a forward or reverse sense with respect to the direction of rotation of the shaft.m. 4 by the decrease in curvature of the whirl deflection form.2. 3 compared with the calculated frequencies of section 2.--For these investigations the rotor inertias were set at the maximum possible values (c = 0. as definite frequencies at particular shaft rotational speeds. The results of these experiments are shown plotted in Figs. compared with the static form. T h e theoretical explanation of this phenomenon may be obtained from equation (14) (section 2. 4. At ~ ----.1.--4. i. reducing the natural frequency.0526 lb in.2. This confirms the results of previous experiments on this rig 2. 5 and 6. 5 the measured frequencies have been plotted as they actually appeared. 4. sec ~) and the rotor was accurately balanced. Discussion of the Results.2. i. The natural vibrations of the rotating system. When the system is rotating.0 there are the two natural frequencies appropriate to the non-rotating system (see section 2.. sec 2.3). 5 and 6 show the measured natural frequencies of the rotating shaft. The deflection forms of the shaft when statically loaded at the rotor end.--Figs. the signal from the pick-ups being recorded from a cathode-ray tube by means of a continuous-film camera and the frequencies measured against a fifty-cycle timing trace.1. the resonant frequencies were observed visually on a monitor cathode-ray tube and then recorded to give an exact frequency reading. Theoretically when the shaft is vibrating transversely the inertia couple will act in phase with the displacement of the shaft and have the same effect on the system as an increase in the flexibility coefficient Y11. This is shown in Fig. It is considered that the outrigger bearing imposes an extra constraint on the non-rotating system when vibrating transversely which is not present when the shaft is whirling thus changing the mode of vibration and increasing the natural frequency. because of the inertia coupling it was effective in exciting the circular vibrations.The lower natural frequency of the non-rotating system was measured for the same values of rotor inertias used in the whirling experiments and is shown in Fig..) which gives the frequencies of the free vibrations (~o) for any rotational speed (~).026 lb in.

6 the natural frequencies of Fig. R. . . The self-aligning outrigger bearing used in the experiments is shown to operate satisfactorily when the shaft is whirling. AUthor 1 E.s. . When the shaft is rotating the roots of equation (14) are no longer symmetrical. . . A programme of tests o n a large model rig is at present being carried out at Royal Aircraft Establishment to investigate the constraints in plain bearings of different lengths and clearance. 5. . . Mag. and the measured value of y was 60 c. . The Escalator Method in Engineering Vibration Problems. The Strength of Shafts in Vibration. . 1950.~. whereas on correcting for this flexibility the frequencies obtained are lower than those measured although the discrepancy between theory and experiment is less than in the latter case. two calculated (one assuming rigid rotor arms and one allowing for flexibility in the rotor arms) and a set of measured results obtained by exciting the rotating system. Lockwood. having regard to the constraints imposed on the system. which is to be expected as the natural frequency of the r o t o r arms. 4 J. the effect on the critical whirling speed of introducing arm flexibility into the calculations was found to be of the order of 1 per cent. In the case of the root originating in + y again the correction for rotor arm flexibility is too-great for the lower rotational shaft speeds but as the shaft speed increases the two curves of calculated and measured results quickly converge as the frequency of the circular vibration becomes appreciably greater than the natural frequencies of the rotor arms. . . the negative roots are asymptotic to o~ = 0 and co = -. . . 5 H. the higher frequencies are affected to a greater degree.. Chapman and Hall. 2768. 1929. . . : the negative roots of ~o correspond to the reverse whirls and the positive roots correspond to forward whirls. . . providing the constraints in a system are known precisely. measurements are being taken of similar constraints in fullscale test rigs and aircraft power plant installations. . Conclusions. 6. 3 J. especially when the frequencies are appreciably different from the natural non-rotary frequencies.--The experiments described show that. In Fig. . The Experimental Approach to the Problems of Shaft Whirling. . . It is seen that the lower frequencies are not appreciably affected by the flexibility of the rotor arms.~/(611/mA) respectively and the positive roots are asymptotic to ~o = + ~/(¢~/mA) and co = c(~?/a) respectively. Where possible.55.).s. it is possible to predict accurately the critical whirling speeds and the natural frequencies of the rotating system for a rotor of appreciable moment of inertia. forms an important aspect in design calculations. x x x v l I . 5 are plotted algebraically. The m e t h o d has the additional advantage that it is easily applied to the basic theory since only a change in the flexibility coefficient ¢~1 is required. . 1947.--The effect of bearings on tile critical whirling speeds of shaft systems. and also the nature of the support given to a system when plain bearings are used. 1950. Vol. The allowance for flexibility of the rotor arms increases the accuracy of the calculations. Downham . The Lateral Vibration of Loaded Shafts in the Neighbourhoodof a Whirling Speed. REFERENCES No. . are higher than those measured when the flexibility of the rotor arms is ignored. . Crosby. March. Morris . the negative roots decrease in frequency with increase of shaft speed and the positive roots increase in frequency. 304 to 314. C. . Downham . . 2 E. June. . . Further Developments. . . . The Effect of Want of Balance. .of equation (14) given by ~o = :t: c¢ and co = 4. . . Some Preliminary Model Experiments on the Whirling of Shafts. positive and negative appropriate to forward and reverse motions respectively. Morris . . Jeffcott . Title.P. & M. pp..p. etc. Three sets of results have been plotted (co against s~). Phil.H. . was 58 c. 1919. The calculated roots of the negative frequency originating in -. . though the restraint on bending increases the natural frequency of vibration when the shaft is not rotating. However. June. .p. .

x) + C -.. 1003C.x) dx 2 2} E 1 dy dx Ely At point 2..x) + C .M o 2 = W ( l . >- o 2. Let S be stiffness (lb/in./in. For the actual system a = 10 in.P ( a . E 1 = 5446 lb in. + ~-- Sy~ I + -~-j y= = W + ~-.x) M12 -~ W ( 1 . Then and Mo2 = W(l -. A couple C is assumed to act about point 1. 1 = 12 in. w(600- 167) + 50c Therefore Y~ ----- 5446 + 10952 : 0-0264W + 0-00305C.. S = 82. 8 6 8 3 W + 0.APPENDIX I The Calculation of the Flexibility Coefficients Y n ¢ n and zn The flexibility coefficients are best calculated by means of the energy method. P w SPRING ~ STIFFNESS S' Ib.P(a -. The installation to be considered consists of a cantilever shaft 01 carrying a load W at 1 and supported by a flexible prop at point 2. s..89 lb/in. Therefore P = Sy2 = 0 . "~ Z. Between 0 and 2 E l d~Y -.x) + C. 8 .. Let M02 be the bending moment at any section of the shaft between 0 and 2 distance x from 0.) of the prop. P = Sy~. Let M12 be the bending m o m e n t at any section of the shaft between 1 and 2 distance x from 0. Ely2 = W x = a. Let P be the force exerted by the flexible prop. Therefore Therefore y ----y~.

Equating the expressions on the right-hand sides of equations (1) and (2) gives (2) ½[0. 888 lb O. . and the work done b y couple C and load W ./la~ = 0 ' 0 0 0 3 3 . ./lb zn = O" 00525 in. 9 12 in.0013C ~] whence = ~1 [ W2y n + 2WCz~ + C~¢n] Yn = 0. Then due to the load W and couple C. calculated in Appendix II.. are : ./lb and ¢n = 0"0013 radn/lb in. the linear deflection at point 1 = Wy l + Cz11. 0443 lb O./lb 411 = 0"0013 radn/lb in. ./lb in.. z11 = O.89 lb/in..OI050WC. . . = -}[W~yn'+ 2WCzi-~ + C~¢n].47 lb/in. 00525 in. A P P E N D I X II Dimensional Details of Model Rig Effective length of shaft Distance of outrigger bearing from the' fixed end of the shaft Equivalent weight of shaft Weight of rotor ~' Hence equivalent end load Measured spring rate of outrigger bearing support Measured stiffness of shaft alone Measured stiffness of shaft and outrigger bearing support combined The deflection coefficients.~ . .3 lb/in.Yll = 0" 0367 in..0013C ~]. . .+ 0.. 10 in. Angular deflection at point 1 : WZ~ + CCn .::.' . 9323 lb 32. (1) Let Yn be the linear deflection at 1 due to unit load at 1 zn be the linear deflection at 1 due to unit couple at 1 = angular deflection at 1 due to unit load a t 1 ¢n be the angular deflection at 1 due to unit couple at 1. .The work done in bending the shaft is given b y V -- 1 2EI o Mo2~dx-J- ~ M122dx + P 2Y~ = ½[0-0367W ~ + O. . . From measurements taken on the rotor Ua. 27. 9.. O.0367 in.0367W ~ + O:OI050WC + 0.

4 /___× 9( 2V. ~ _ _ x _'. ' ~ ( 9 ¢.ao / Lt} ins ~ CRITLCAL SPEED 17... 10 30 .. l 1 OI 0 0 I0 15 ~HAFT S P E E O - FIG.~0 LB. A ( ~ (C-A). i~S~ • •.)" ~.:.c FIG.3 LB IN~~ t-A)- The whirling condition. 20 ~5 ~P5 The effect of rotor inertia on the critical whirling speed of a rotating shaft.27.'~.1 .45 . / °. ~ *X¢.~o ~..7 RRS.515 LB. 1. 15'85 i I 0"~. 5.:~ .E~ iNS~ 17-6 .' 0 0 .~--9. ~ ® 18.' / B~ I (c-A): t. IB.I0. 2.7-E..(~ (:-A). IN5 ~ ..

~G FI~EQUENCY .kT[CDEFL£CT[ONFORM e ® WHI~LIN5 c3 TRANSVERSE VlS~ATIOt~ • / / RO C MAiMDRIVIN~ 5H~FT <> 0 { ~ ~ 4 5 ~ 7 8 -~ IO II I~ DIST. 5"r.IZ IO NATURAL F~EQUENCY CRITICAL SPEED / (CALCULATE V E z 5PEE~ NkTU~AL~~~ E ~ ~ ~3 N g~ N" 0 J4 15 FIG.tVING5HAFT(INS) FIG. t . 3. Comparison between deflection forms of the non-rotating system when loaded statically at the rotor and vibrating naturally and of the rotating system when whirling.C.~NCE FI~OMPOINT OF kTTACHMENTTO MAtN D~. 4.P5 17 16 The effect of rotor inertia on the natural vibrations of the non-rotating system and on the critical whirling speed. 11 .

i I i / Iilt .P.~ O3 _t -o =: > \ b~O ~ o • o o . . - // g l CRITICAL WIIIRLIN~ SPEED' / I01 / / t • '//" xl / 0 -80 / /'] ~301 g i / i" h:. .~ / 501 //' ! i II/ II ~ [ / 401 [ i • t i i FIG. . .~o.~.I-~0_ 0 +. 12 . . ~ll i i .~ ~0 +j 40 + ~ ~0 F~EQUENCY OF NATUJ~.~ |~ / .y I -40 /- //.L VIBRATION5 '~'" C. I/!'. -GO_.B~ . 80 I00 I~0 14'Q The natural vibrations of a rotating shaft carrying a rotor of appreciable moments of inertia (measured and calculated).~ o '£d3 13N3Qb3~ 70~ ! _~'! ! 60t I CALCUL~'~ED~RV~ ASSUMING " ~. t:3 II # ' I !l! jl ' o2 - . 6. .

Details of the m o d e l shaft rotor s y s t e m .":~: ~ ' ° FIG.17/680 K9 9/54 D&Co.Adjustable Weights Flexible Bearing Main Driving Shaft Four Arm Rotor Whirling Shaft -. 34/263 13 B . P R I N T E D IN G R E A T BRITAIN J4300 Wt. 7.

) 1941 Aero and Hydrodynamics. Engines. Performance. 1946. Id). Id. 39 King Street. 6d.) April I. 9os. . 4½d. Wind Tunnels. (8is. (5IS. Engines. 8os. 5os. ANNUAL REPORTS OF THE AERONAUTICAL RESEARCH COUNCIL-1933-34 is. Parachutes. Aero and Hydrodynamics. I. 2050. . Airscrews. Structures. Test Equipment.R. I½d.o. & M.. (IS. Seaplanes. R. (4IS. Wind Tunnels. 31. 1951 . ISS. n . 3d. Engines.) Prices in brackets include 2hostage. 6d. Stability and Control. Vol. (4IS. 4os. . No. Io½d. Structures. 8d.O. I936. . 2570. Aerofoils. . No. 3os.) 1939-48 3 s. Stability and Control. (4s. & M. (64 s. Airscrews. & M . 2d. 1939 Vol. R. London W.) Vol. No. id.June 30. Structures.1 (Post Orders : P.) 1944 Vol. Structures.) is. Flutter and Spinning. Wind Tunnels. 5os.Andrew's Crescent. Airscrews. Flutter. Id. 6d. Parachutes.) INDEXES TO THE TECHNICAL REPORTS OF THE AERONAUTICAL RESEARCH COUNCIL--December I.) 1936 Vol. Belfast OR T H R O U G H A N Y BOOKSELLER s.) Vol. & M . is. 1935 to Dec. (is.December 31. I. 2600. Seaplanes. Aerodynamics. 6d. Noise. I & ) Aerodynamics General. 215o. Vol. 63 s. 4os. (5IS. II.C. Airscrews. Stability and Control. 3d.) Aerodynamics General. (2s. Performance. Code NO. etc. AND SEPARATELY-April. 6d. 423 Oxford Street.June 30. Bristol I . l~lanchester 2 . I. I.. 7d. Box No. 75s.E.) 1937 2s.) is. 47 s. and Panels. Structures. 8d. and a miscellaneous section. Vol. id. 1937 Vol. Controls. R . Instruments.June 30. (2s. Kingsway. I. 3d. Structures. 195o. (is. etc. II. (IS. Parachutes. Aerofoils. 195o . 1946. Airscrews. 6os.) 1943 Vol. etc. Stability and Control. 63 s. (is.) 1938 IS. & M.) IS.. Engines. 2s. 2d.) Vol~ II. Seaplanes. Materials. 5os. 2854 Publications of the Aeronautical Research Council ANNUAL TECHNICAL REPORTS OF THE AERONAUTICAL RESEARCH COUNCIL (BOUND VOLUMES) Aerodynamics General. (3 is. 4½d. (76s. 1938 Vol. Plates. January i. Materials. 84 s.) 194o Aero and Hydrodynamics.. Airscrews. Engines. . 4 d. London S. Airscrews. .) Stability and Control. Structures. R. 1936 . Flutter and Vibration. (6is. Flutter and Spinning. Aircraft. R. I. Engines. Flutter and Vibration. Materials. . 8d. July 1. 185o. 1947 . (9IS. (is. (IS. Stability and Control. Flutter. Cardiff .) 1934-35 IS. Performance. Airscrews. 13A Castle Street. Structures. 2854 . July 1. No. & M. No. & M. 1939. R. id. Fhltter.) INDEX TO ALL REPORTS AND MEMORANDA PUBLISHED I N THE ANNUAL TECHNICAL REPORTS. 1945 .1939 . (is. Vibration. (48s. Stability. London W.) Aerodynamics General. BJrmJngham 3 : 1 St. Icing. 80 Chichester Street. (85s. Engines. .June 30. II. 569.) IS. R. & M. 5os. July i. (85s. Performance. July. . Tower Lane. Structures. 2 Edmund Street. 9 d. 2350. Aerofoils. (64 s. 1945. No. (Sis. id. 2250. 2d. Navigation. 4s. No. 8d. Airscrews.) 1942 Vol.R . Edinburgh 2 . 2d. & M. Aerofoils..) AUTHOR INDEX TO ALL REPORTS AND MEMORANDA OF THE AERONAUTICAL RESEARCH COUNCIL-19o9-1949 . Aero and Hydrodynamics. 7½d. I. & M. 1947.2 . 8d. No. 6d. . 4½d. Performance. Performance. Miscellaneous . 4d. (5 is. 1946 . 23-2854 R. No. (3s. 84s. 3d. Flutter. 3d.) Stability and Control.) is. II. Aerofoils.1) .. Obtainable from HER MAJESTY'S STATIONERY OFFICE York House. (is. No. Vol. II. Seaplanes. (ISS.) Stability and Control. i½d.