Gear Inspection: Troubleshooting Tips

You inspect your gears to make sure you’re producing the best product you possibly can. Should you detect a problem, this checklist will
help you quickly determine the source.
In the June 2005 issue of Gear Solutions we presented an article titled “An Elementary Guide to Gear Inspection,” which was followed by
“A New Standard in Gear Inspection” in October. Then, in the February 2006 issue, “Guarding Against Gearing Deviations” was
published. These articles can be downloaded from the magazine’s web site www.gearsolutions.com. In this article we will focus on the
practical side of gear inspection; in particular, what you should do when you measure certain errors during the hobbing process.
Size Variation
During gear hobbing the size of a gear is measured throughout the production run as a process control by the operator. As shown in
Figure 1, this measurement is normally done using balls or pins of a specified diameter to contact the gear tooth profile at the pitch
diameter. For larger gears a span measurement over several gear teeth will be done.

Figure 1: Measurement over pins

If excessive size change occurs during production, check the following items:
1) Size changes from a “cold” morning startup to normal operating temperature. The machine system may be thermally unstable. Record
the size change over time versus temperature of the machine and/or cutting fluid. A “warm up” period may be required for your particular
machine. Also, certain gear machines have thermal sensor(s) for automatic adjustment of the machine size, and this system may have
failed.
2) Size changes during hob shifting. Check that the cutter has been properly sharpened without gash lead error. Due to the cam relief in a
hob, any gash lead error will create a conical shaped hob. This will cause a size change in the part as the hob is shifted from one end to
the other. The gash lead error will be approximately 2.5 times the taper error. Measuring the outside diameter of the hob at several points
across the face is a quick way to determine if gash lead error is present. (Figure 2)

Figure 2: Hob gash lead error

3) Size changes from part to part. On a mechanical hobbing machine this can be cause
d by an error in the positive stop for center distance. With a long travel indicator measure the actual center distance from piece to piece
and look for variation. Another cause of this problem could be a hob slide that is normally clamped after hob shifting that has failed to
clamp.

Composite Inspection Error

During gear hobbing a composite inspection is normally conducted throughout the production run as a process control by the operator.
This is accomplished on a simple rolling machine using a master gear with the product gear. (Figure 3)

Figure 3: Schematic of gear rolling process

Composite error will occur as two conditions. First as a tooth-to-tooth error, and second as a once per revolution error of the test gear.
The once per revolution error is commonly known as TCE, Total Composite Error, and is defined by the AGMA as the Radial Composite
Variation. (Figure 4)
 Figure 4: Graphical record from composite inspection
Composite Error condition #1: Large ƒi” and Uniform Tooth-to-Tooth Errors
1) The large ƒi” is typically a dirty condition or a nick on a gear tooth. Nicks are normally caused by material handling problems. Adding a
tip chamfer to the design of the gear tooth profile, and hob, will reduce this type of damage. (Figure 5)

Figure 5: Dirty or nicked tooth on gear

Figure 6: Total composite variation, Fi”
Composite Condition #4: Once per Revolution Non Uniform Error
1) The index change gears are damaged or installed improperly on a mechanical hobbing machine.
2) The hobbing machine work spindle or index drive system is worn or has runout.
Profile Inspection Error
The following examples show the different types of profile errors that can occur, with an explanation of the probable cause. (Figure 7)

Figure 7: Profile inspection with negative tip tooth

Figure 8: Hob sharpened with positive rake
1) Incorrect hob sharpening. The hob has been sharpened with positive rake error making the hob tooth larger toward the outside
diameter and the gear tooth smaller. (Figure 8)
2) Incorrect swivel angle setting on the hobbing machine.
3) Bad hob.

Figure 9: Profile inspection with positive tip

1) Incorrect hob sharpening. The hob has been sharpened with negative rake error, making the hob tooth smaller toward the outside
diameter with gear tooth larger. This is the opposite of the condition shown in Figure 8.
2) Incorrect swivel angle setting on the hobbing machine.
3) Bad hob.

Figure 10: Profile inspection with leaning teeth

1) Incorrect hob sharpening. A hob sharpened with gash lead error will cause both a leaning profile and size change as the hob is shifted.
See Figure 2.
2) Loose hob head swivel on the hobbing machine.

Figure 11: Profile inspection with uniform wave

1) Hob had runout when mounted. The hob runout can be caused by a bad cutter, damaged arbor, or dirty mounting conditions.
2) Incorrect hob sharpening. The hob was sharpened with runout during the mounting on the sharpening machine or sharpening arbor.
Extreme gash to gash spacing error on the hob.
3) A loose or worn hob arbor end support on the hobbing machine.
4) Excessive backlash in the hob spindle system on the hobbing machine.
5) Excessive backlash in the work spindle system on the hobbing machine.
 Figure 12: Profile inspection with non-uniform wave
1) A loose or worn hob arbor end support on the hobbing machine.
2) Excessive backlash in the hob spindle system on the hobbing machine.
3) Excessive backlash in the work spindle system on the hobbing machine.
4) Hobbing machine system in poor overall condition.
Helix Deviation Inspection Error
Helix error can be caused by the gear blank, the fixture, the machine setup, and machine damage or misalignment. The following
examples show the different types of helix errors that can occur, with an explanation of the probable cause.

Figure 13: Helix inspection with wobble

1) Part blank bore not perpendicular to the face of the blank.
2) Part blank faces not parallel.
3) Fixture not accurate, or misaligned.
4) Tailstock misaligned, or center loose or damaged on the hobbing machine.
5) Inspection arbor problem.
6) Worm spindle bearing problem on the hobbing machine.

Figure 14: Helix inspection with taper

1) Tailstock misaligned on the hobbing machine.
2) Hobbing machine misaligned.

Figure 15: Helix inspection with angular error

1) Tailstock misaligned on the hobbing machine.
2) Hobbing machine misaligned.
3) Incorrect differential change gears or helix value (CNC).
4) Loose or work hob arbor end support on the hobbing machine.
5) Incorrect helix checking machine settings.

Figure 16: Helix inspection with breakout error

1) Tailstock loose or worn.
2) Part slipped in the fixture.
3) Fixture not clamped properly.
4) Hob arbor end support improperly installed or worn on the hobbing machine.
5) Hob saddle gibs out of adjustment.
6) Backlash in the cutter spindle drive system on the hobbing machine.
7) Backlash in the work spindle drive system on the hobbing machine.

Figure 17: Helix inspection with periodic error

1) Runout of multiple thread hob.
2) Thread to thread spacing error on multiple thread hob.
3) Feed screw or bearings of the feed screw on the hobbing machine.

 Profile Deviations:

If you hear the word "profile" of a gear, you should relate it to the arc that goes from the bottom of the tooth to the top of it, basically up to
the tip or outside diameter of the gear.
In general, deviations in the profile, are only measured at the 92% of the active length, and it is called "profile evaluation range".
As it is defined by the ISO 1328, "Profile deviation is the amount by which an actual profile deviates from the design profile. It is in the
transverse plane and normal to the involute profile."

The image describe how the profile is obtained from the roll angle of the gear tooth.
Type of Profile Errors:

Three important types of errors are related to profile, Profile Form Deviation, Profile Slope deviation
and Total profile deviation.
Profile Form Deviation (ffa):
Distance between two involutes of the actual base circle, that enclose the actual involute profile within the
profile inspection range
Can also be defined as the divergence of the measured profile from a best fit line or curve.
Profile Slope Deviation (fHa):
Distance between two nominal profiles that intersect the average profile at start and end points of the
profile range or Profile slope deviation is the amount of deviation from a nominal involute profile over
the evaluation range.
Total Profile Deviation: (Fa):
Is the total amount of profile error, including slope and form errors or the distance between two nominal
profiles enclosed within the profile test range.
Helix deviations:

AGMA’s current inspection handbook defines “helix deviation” (formerly tooth alignment variation and lead
variation) as the difference between the measured helices to the design helices. In practice an appropriate
measuring machine aligns the measuring probe on the test gear at the pitch circle diameter and the “lead” is
traced and recorded graphically, with a correct unmodified helix being represented as a straight line on the
chart. Helix measurement is used to determine correct face contact between mating gears. Incorrect helix
will create uneven loading and noise.
The ISO 1328 explained in the following terms:
" Helix deviation is the amount, measured in the direction of the transverse base tangent, by which an actual
helix deviates from he design helix ".

Type of helix errors:

Helix Form Deviation (ffb):
Distance between two helical lines that enclose the actual lead within the lead inspection range. Helix form
deviation is the divergence from a best fit curve along the helix
Helix Slope Deviation (fHb):
Helix slope deviation is the amount of deviation from a nominal helix over the evaluation range. Distance (in
transverse plane) between two nominal leads that intersect the average lead (helix) at start and end points
of the lead inspection range.
Total Helix Deviation (Fb):
Distance between the two nominal leads enclosed within the lead inspection range.The total amount of
error including angle and form errors.
This picture is quite helpful because it gives you an idea bout the way the tooth is measured and what is the
meaning of the traces showed in the right side.

Pitch Deviations:
The notation pitch, should make you think about tooth location, and the effect that this type of deviations
would have into the gear motion and torque transferring. It is related to, backlash, contact ratio variation...
It is looking at the location of a gear tooth with respect to the others, as well as the global location between
all gear teeth. Measures, the difference between the actual position of the tooth and the "should be here"
position.
The application calculate the values of the pitch error (fu), the single pitch deviation (fp) and total pitch
deviation (Fp).

Types of pitch deviations:

Single pitch deviation (fp):

It is the difference between two adjacent teeth index values (+,-) or the algebraic difference between the
actual pitch and the corresponding theoretical pitch in the transverse plane, defined on a circle concentric
with the gear axis at approximately mid-depth of the tooth.

Pitch Error (fu):

It is the difference between adjacent pitches.The difference between actual dimensions of
two successive right or left flank transverse pitches.

Total cumulative pitch deviation (Fp):

Maximum cumulative pitch deviation of any sector of the corresponding flanks of a gear. It is represented by
the total amplitude of the cumulative pitch deviation curve.
Pitch deviations

Gear index and pitch definitions.

Transmission error deviations (tangential composite deviations):

In this case, those deviations are related to the gear motion. You need to look at the gear in touch with a
master gear and how the center of the gear changes when it revolves one complete turn. That will affect
load transferring, and bearing cycle load reactions.
The application gives two different values:
Two flank working deviation( f´´i):
Maximum difference between the effective and theoretical circumferential displacements at the reference
circle of the gear under inspection, when meshing with a master gear, testted product gear being turned
through one complete revolution.

Single flank working deviation (f´i): 

Value of the tangential composite deviation over a displacement of one pitch. It is a true tangential
measurement and is indicative of the functional characteristics of the gear.The non uniform motion is called
"transmission error"
As a reference, here you have also a table with the quality grades comparison between DIN and
AGMA standards:

Gears are one of the basic elements used to transmit power and position. As designers, we desire them to meet various demands :
1. Maximum power capability
2. Minimum size
3. Minimum noise (silent operation)
4. Accurate rotation/position
To meet various levels of these demands requires appropriate degrees of gear accuracy. This involves several gear features.
7.1 Accuracy of Spur Gears and Helical Gears
Gear accuracy of spur gears and helical gears, is described in accordance with the following JIS standards. JIS B 1702-1 : 1998
Cylindrical gears – ISO system of accuracy – Part 1:
Definitions and allowable values of deviations relevant to corresponding flanks of gear teeth. (This specification describes 13 grades of
gear accuracy grouped from 0 through 12, – 0, the highest grade and 12, the lowest grade).
JIS B 1702-2 : 1998 Cylindrical gears – ISO system of accuracy – Part 2: Definitions and allowable values of deviations relevant to radial
composite deviations and runout information (This specification consists of 9 grades of gear accuracy grouped from 4 through 12, – 4, the
highest grade and 12, the lowest grade).
These new standards for gear accuracy differ from the former standards of JIS B 1702-1976 in various ways. For example, the gear
accuracy used to be classified into nine grades (0 to 8) in the former standards. To distinguish new standards from old ones, each of the
grades under the new standards has the prefix “N”.
(1) Single Pitch Deviation (fpt)
The deviation between actual measured pitch value between any adjacent tooth surface and theoretical circular pitch.

Fig.7.1 Single pitch deviation fpt

(2) Total Cumulative Pitch Deviation (Fp)
Difference between theoretical summation over any number of teeth interval, and summation of actual pitch measurement over the same
interval.

Fig.7.2 Total cumulative pitch deviation

Fig.7.3 Examples of pitch deviation for a 15 tooth gear

(3) Total Profile Deviation (Fa)
Total profile deviation represents the distance (Fa) shown in Figure 7.4. Actual profile chart is lying in between upper design chart and
lower design chart.

Fig.7.4 Total profile deviation Fα

(4) Total Helix Deviation (Fβ)
Total helix deviation represents the distance (Fβ) shown in Figure 7.5. The actual helix chart is lying in between upper helix chart and
lower helix chart. Total helix deviation results in poor tooth contact, particularly concentrating contact to the tip area. Modifications, such
as tooth crowning and end relief can alleviate this deviation to some degree.
Shown in Figure 7.6 is an example of a chart measuring total profile deviation and total helix deviation using a Zeiss UMS 550 tester.
Fig.7.5 Total helix deviation Fβ

Fig.7.6 An example of a chart measuring total profile deviation and total helix deviation
(5) Total Radial Composite Deviation (Fi” )
Total radial composite deviation represents variation in center distance when product gear is rotated one revolution in tight mesh with a
master gear.

Fig.7.7 Chart of total radial composite deviation

(6) Runout Error of Gear Teeth (Fr )
Most often runout error is measured by indicating the position of a pin or ball inserted in each tooth space around the gear and taking the
largest difference. Runout causes a number of problems, one of which is noise. The source of this error is most often insufficient accuracy
and ruggedness of the cutting arbor and tooling system. And, therefore, it is very important to pay attention to these cutting arbor and
tooling system to reduce runout error. Shown in Fig. 7.8 is the chart of runout. The values of runout includes eccentricity.
Fig. 7.8 Runout error of a 16-tooth gear
Please see page 620 to 625 in referring to the selected standard values for each allowable error.
7.2 Accuracy of Bevel Gears
JIS B 1704：1978 regulates the specification of a bevel gear’s accuracy. It also groups bevel gears into 9 grades, from 0 to 8.
There are 4 types of allowable errors:
1. Single pitch error
2. Pitch variation error
3. Accumulative pitch error
4. Runout error of teeth (pitch circle)
These are similar to the spur gear errors.
1. Single pitch error
The deviation between actual measured pitch value between any adjacent teeth and the theoretical circular pitch at the mean
cone distance.
2. Pitch variation error
Absolute pitch variation between any two adjacent teeth at the mean cone distance.
3. Accumulative pitch error
Difference between theoretical pitch sum of any teeth interval, and the summation of actual measured pitches for the same
teeth interval at the mean cone distance.
4. Runout error of teeth
This is the maximum amount of tooth runout in the radial direction, measured by indicating a pin or ball placed between two
teeth at the central cone distance.
Table 7.1 presents equations for allowable values of these various errors.
Table 7.1 Equations for allowable single pitch error, Accumulative pitch error and pitch cone runout error (μm）

The allowable pitch variation error value is defined as; Single pitch error tolerance x k-value Table 7.2 shows the k-value. The k-value
varies depending on the tolerance value of a single pitch error.
Table 7.2 k-values

Besides the above errors, there are seven specifications for bevel gear blank dimensions and angles, plus an eighth that concerns the cut
gear set:
1. The tolerance of the blank tip diameter and the crown to back surface distance.
2. The tolerance of the outer cone angle of the gear blank.
3. The tolerance of the cone surface runout of the gear blank.
4. The tolerance of the side surface runout of the gear blank.
5. The feeler gauze size to check the flatness of blank back surface.
6. The tolerance of the shaft runout of the gear blank.
7. The tolerance of the shaft bore dimension deviation of the gear blank.
8. The tooth contac