B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

Bernd-Robert Hoehn, Peter Oster, Thomas Tobie, Klaus Michaelis

ISSN 0350-350X
GOMABN 47, 2,129-152
Struni rad/Professional paper
UDK 620.22.05 : 621.892.094 : 620.1.05

TEST METHODS FOR GEAR LUBRICANTS

Abstract
Base oil type, oil viscosity, additive type and content have a strong influence
on typical gear failures. In general it is not possible to quantify the lubricant
influence on load carrying capacity simply from the knowledge of physical or
chemical oil data. Therefore many test methods were developed for the
evaluation of mechanical-technological lubricant properties. Simple low-cost
bench test methods often show poor correlation with practice. From
experience and systematic investigations it can be shown that testing of gear
lubricants can adequately be performed only in gear test rigs using specified
test gear geometry [1].
In continuous work over many years the standard FZG back-to-back gear
test rig was developed and improved for different types of gear failure
simulation. The standard FZG oil test A/8,3/90 is widely used for the
evaluation of the scuffing properties of industrial gear oils. Automotive gear
oils of GL4 level can be tested in the step test A10/16.6R/90, axle oils of
GL5 level in the shock test S-A10/16.6R/90. For the slow speed regime the
wear test C/0.05/90:120/12 can be applied. The influence of lubricants on
the micropitting performance of gears can be evaluated in a screening short
test GFKT-C/8,3/90 or in the full micropitting test GF-C/8,3/90. Different
pitting tests are available as single stage test PT-C/9:10/90 or application
test PTX-C/SNC/90.
The aim of the paper is to describe the influence of the lubricant on the
different failure modes in gears, how to quantify this influence in adequate
test methods and how to introduce the results of such tests as strength
values for the lubricant into load carrying capacity rating methods.

1. Introduction
For better efficiency in transmissions, lower viscosity grades of gear oils and less oil
volumes are used. With increasing power density and extended oil drain intervals
the demands on the lubricant as a design element increase accordingly. The
introduction of lubricant properties in load carrying capacity rating requires not only
knowledge of their physical properties as e.g. viscosity, viscosity-temperature- or

goriva i maziva, 47, 2 : 129-152, 2008. 141

Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

viscosity-pressure-behaviour but also the quantitative influence of an EP oil on

scuffing, wear, micropitting and pitting of gears. Numerous test methods were
developed to describe lubricant properties. Simple and low-cost bench tests on the
one end as well as full scale field testing on the other end can be taken into
consideration [1]. The aim would be a test method as simple, cheap and quick as
possible to produce comparable results to practical applications.
Figure 1 shows some of the test arrangements developed for the prediction of
scuffing and wear properties of gear oils. Closest to practice are methods using test
gears. Methods on disk type rigs often simulate one point on the path of contact of a
gear mesh with its local rolling and sliding velocity. Most of the rigs use simple
specimen under pure sliding conditions.
Figure 2 shows scuffing results on some of these rigs for different lubricants
including milk and beer [2]. In the often specified four ball and Timken tests milk and
beer are rated with higher scuffing load capacity as a non-EP mineral oil ISO VG
220 or a hydraulic oil with ZDTP additives ISO VG 46. Only the FZG test using test
gears gives a correct relative rating of these lubricants. Even results from twin disk
machines with a close simulation of the kinematics of one point on the path of
contact can be quite misleading. Two standard hypoid gear oils and one with an
additional MoS2 content were evaluated in a hypoid gear, a spur gear and a twin
disk test (Figure 3). While the two gear tests showed rather poor performance for the
MoS2 containing lubricant of less than 50 % of the standard oils the twin disk result
was at 300 %.

Figure 1: Lubricant test arrangements (schematic)

142 goriva i maziva, 47, 2 : 129-152, 2008.

B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

2. Test rig and test gears

With this experience a number of test methods for gear lubricants was developed
using the FZG back-to-back gear test rig where over 500 machines are available
world-wide [3].
2.1 Test Rig
The FZG gear test rig is a back-to-back rig with centre distance a=91.5 mm [4]. The
Figure 2: Scuffing results for different lubricants [2] test gears and the slave
gears are connected by two
shafts. One shaft is divided in
two parts with the load clutch
in between. One half of the
load clutch can be fixed to the
foundation with the locking
pin while the other half is
twisted e.g. by means of a
lever and weights. After
bolting the clutch together the
load can be removed and the
shaft be unlocked. Thus a
static torque is applied to the
system which can be checked

Arrows indicate results outside test range

Figure 3: Scuffing results in disk and gear tests Figure 4: FZG back-to-back gear test rig
goriva i maziva, 47, 2 : 129-152, 2008. 143
Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

in the torque measuring clutch as the twist of a calibrated torsion shaft. The rig can
be operated either with a two speed AC motor at 1500/3000 rpm or a variable speed
DC motor between n = 100 - 3000 rpm. The gears are normally dip lubricated with a
heater and a cooling coil in the test head for temperature control. With an additional
oil supply device also spray lubrication can be applied.

2.2 Test Gears

For the different standard tests there are two standard test gear geometries in use
[3]. For scuffing tests gear type A with high sliding at the pinion tip is used while gear
type C with balanced sliding is used for wear, micropitting and pitting tests. The main
data of the test gears are summarized in Table 1. Test gears are case carburised
with a surface hardness of HRC = 60 + 2 and a case depth of Eht = 0.6 - 0.9 mm.
The surface roughness is controlled to the required values of the respective test
procedure. Test gears are manufactured in batches of 100 pairs by one
manufacturer and closely controlled to metallurgical properties, gear accuracy and
surface finish.

3. Test Methods
For the different gear failure Table 1: Main data of test gears
modes which can be
symbol unit A-type C-type
influenced by the lubricant
type, viscosity and additive centre distance a mm 91,5
system adequate test number of teeth pinion z1 - 16
methods were developed in gear z2 - 24
close co-operation with module m mm 4,5
industry. In some cases pressure angle 20
different modifications of the helix angle 0
test methods are available face width b mm 20 14
adjusted to the specific
profile shift factor pinion x1 - 0,8532 0,1817
application. In the following gear x2 - -0,5 0,1715
often used methods are pitch diameter pinion dw1 mm 73,2
described. gear dw2 mm 109,8
tip diameter pinion da1 mm 88,8 82,5
3.1 Scuffing Load gear da2 mm 112,5 118,4
Capacity Tests material MAT - 20MnCr5 16MnCr5
A widely used test method heat treatment - - case carburized
for the evaluation of the
scuffing properties of
industrial gear lubricants is
the FZG gear oil test A/8.3/90 according to DIN ISO 14635-1 [4], which is equivalent
to other standards as IP 334 [5], ASTM D-5182 [6], CEC L-07-A-95 [7] and identical
to ISO DIS 14635-1 [8].

144 goriva i maziva, 47, 2 : 129-152, 2008.

B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

A-type gears are loaded stepwise in 12 load stages between a Hertzian stress of pC
= 150 to 1800 N/mm. They are operated for 15 min at a pitch line velocity of 8.3 m/s
and a starting oil
temperature of 90 0C in typical typical scuffing load stage
each load stage, under lubricant type
ISO VG non mild EP
EP
conditions of dip lubrication EP oil
without cooling. In the visual
test the gear flanks are ATF 32-46 2-4 ATF 32-46
inspected after each load
stage for scuffing marks. turbine oil 32-68 3-5 turbine oil 32-68
Failure load stage is
indicated when the faces of industrial gear oil industrial 100-
100-320 5-7
gear oil 320
all pinion teeth show a
summed total width of
transmission 100-
damaged areas which is transmission oil 100-220 -
oil 220
equal or exceeds one tooth
width. In the gravimetric test
the gears are dismounted axle oil 150-
150-220 - axle oil
and weighed to determine 220
their weight loss. From the
Table 2: Typical results of the standard scuffing test
curve of the weight loss the specific wear parameter can be evaluated as well as the
scuffing load stage which is indicated by a steep increase of the wear curve.
Table 2 shows typical results of gear lubricants in this test. It is obvious that for
industrial applications with high scuffing demands of minimum API GL3 as e.g. in
street cars or wind turbines as well as in automotive applications as e.g. in manual
transmissions of commercial vehicles with API GL4 and in hypoid rear axles with
API GL5 requirements the standard test has not sufficient discriminating power.
For automotive gear oils of API GL4 level no accepted test method is available, for
API GL 5 oils the CRC or the FZG L-42 test [9] can be used. In systematic
investigations on the standard FZG gear test rig the scuffing risk of the standard
procedure was increased by varying speed and specific load, load application and
sense of rotation. A step test A10/16.6R/90 for lubricants up to the level of GL 4 and
a shock test S-A10/16.6R/90 for discrimination between GL4 and GL5 were
developed. A-type test gears with reduced pinion face width to b = 10 mm (A10) are
used at increased speed of 16.6 m/s (16.6) and reversed sense of rotation (R), at oil
temperature of 90 C (90). In the step test load is stepwise increased until scuffing
occurs. In the shock test (S) the gears are directly loaded in the expected load stage
and PASS or FAIL is stated. A detailed description of the test procedures can be
taken from the FVA Information Sheet [10].

goriva i maziva, 47, 2 : 129-152, 2008. 145

 Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

A comparison of results from different scuffing test methods is shown in Figure 5.

With these tests it is possible to cover the whole range of scuffing load capacity
between API GL1 to GL5. Besides a relative ranking of lubricants with respect to
A/8.3/90 A10/16.6R/90

S-A10/16.6R/90
800

700 14
13
600
12
500
11
400
10
300 9
8
200 7
6
100
4
0 2
C1 C2 C3 G1A G3 G4 G6 G7
GL3 GL4 GL4+ GL5

Figure 5: Scuffing results for different lubricants Figure 6: Comparison of scuffing test methods

their scuffing performance the results of these tests can also be recalculated to
critical scuffing temperatures and be introduced into the DIN [11] and ISO [12]
standards of scuffing load capacity rating. On the basis of the critical scuffing
temperature, different test methods can be compared as shown in Figure 6.

3.2 Slow Speed Wear Test

The Chevron Test
[13] for tractor test conditions C/0,05/90/12 C/0,05/120/12 C/0,57/90/12
hydraulic fluids of low pitch line velocity 0,05 m/s 0,05 m/s 0,57 m/s
viscosity grade can pinion speed 13 rpm 13 rpm 150 rpm
normally not wheel speed 8,7 rpm 8,7 rpm 100 rpm
discriminate different oil temperature 90C 120C 90C
industrial or pinion torque 378,2 Nm 378,2 Nm 378,2 Nm
automotive gear oils running time 2 x 20 h 2 x 20 h 1 x 40 h
of higher viscosity.
revolutions of shaft 2 2 x 10400 2 x 10400 1x240000
Therefore a new
universal wear test test result weight loss of pinion and gear
C/0.05/90:120/12 was Table 3: Test conditions of the wear test C/0,05/90:120/12
developed (see Table
3). Pitting test gears type C-PT (C) are used at a pinion speed of 0.05 m/s (0.05) -
equivalent to 13 rpm at the pinion - at oil temperatures of 90 and 120 C (90:120) at
highest load stage 12 (12) - equivalent to a pinion torque of T1 = 378,2 Nm. As a
minimum the method consists of two test parts. A detailed description of the test
procedure can be taken from the DGMK Information Sheet [14]. For running the test

146 goriva i maziva, 47, 2 : 129-152, 2008.

B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

the standard FZG rig must be modified with a speed reducer between driving motor
and slave gear box.
Part 1: C/0.05/90/12 Part 1 runs two times 20 h with intermediate weighing of
pinion and gear. The pitch line velocity is v = 0.05 m/s, and the oil sump temperature
is maintained at oil = 90C. This condition gave the highest wear rat es for all tested
lubricants.
Part 2: C/0.05/120/12 Part 2 runs also two times 20 h with intermediate weighing
of pinion and gear. The oil sump temperature is increased to oil = 120C with the
other parameters kept constant, to check the additive reaction at elevated
temperature.
The two parts have always to be carried out for testing a lubricant. Part 3 can be
added for more detailed information in one operating condition.
Part 3: C/0.05/90/12 or C/0.05/120/12 or C/0.57/90/12 Part 3 runs 40 h without
intermediate weighing. C/0.05/90/12: Repeating the test conditions of part 1 can
show how mechanical and chemical changes on the flank surface are relevant under
changing operating conditions. C/0.05/120/12: Repeating the test conditions of part
2 can be appropriate when the operating conditions in practice are predominantly at
a higher temperature level and the results of part 2 are not yet sufficient or have not
yet arrived at a steady state level. C/0.57/90/12: Changing the pitch line velocity to v
= 0.57 m/s, corresponding to a pinion speed of n1 = 100rpm, can show the influence
of higher speed and thus higher film thickness and better lubricating conditions on
the wear behaviour. The oil temperature of oil = 90C was chosen, because in
most cases at the lower temperature higher wear rates were found.

Figure 7: Results from the FZG wear test

Different gear lubricants -

hydraulic oils, ATFs, turbine
oils, industrial gear oils and
automotive gear oils of GL4
and GL5 performance - were
tested under these
conditions. Typical test
results are shown in Figure 7.
It can be seen from this
Figure that different
categories of wear behaviour
can occur:

goriva i maziva, 47, 2 : 129-152, 2008. 147

Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

- high or medium wear rate in both parts 1 and 2 (e.g. Dexron 32 D)

- high or medium wear rate in part 1 and low wear rate in part 2 (e.g. GL5-SP
80W)
- low wear rates in both parts 1 and 2 (e.g. UTTO-Z 46).
Predictions of wear results from viscosity parameters or additive content are not
possible. The experimental results of the wear test can be introduced as specific
wear rate values clt into the wear calculation method according to Plewe [15].

3.3 Micropitting Test

For low churning losses
and high efficiency in
torque converters low
viscosity grade lubricants
are often used. To
compensate for the wear
and scuffing performance
loss due to a lack of
viscosity, various anti
wear and EP additives
are used in higher
concentrations. The type
of the base oil and the
additive type show a
Figure 8: Typical micropitting test results
large influence on
micropitting failure of
gears [16].
The standard test method GF-C/8,3/90 for the evaluation of the micropitting
performance of lubricants was established. The test requires an oil spray device for
constant oil temperature of oil = 90 C (90) at an oil flow rate of V = 2 l/min. Gear
type C with a specified high surface roughness of Ra = 0.5 0.1 m (GF-C) are run
at a pitch line velocity of v = 8.3 m/s (8,3). A one hour run-in in load stage 3 (pC =
500 N/mm2) is followed by the short term test in load stages 5 through 10 (pC = 800
- 1500 N/mm2). Running time is 16 h per load stage. After every load stage the
gears are dismounted and the profile deviation is measured. If the profile has
changed to a profile deviation of 7.5 m (corresponding to a change of DIN accuracy
from 5 to 6) the test is terminated, the failure load stage is reached.
In case of failure in load stages 8, 9, 10 or over 10 an endurance test with the same
gear pair is performed. 80 h in load stage 8 are followed by maximum 5 times 80 h
in load stage 10. The gears are inspected after every test sequence and the profile
deviation is again measured. When the profile error exceeds 20 m (corresponding
to a change of the DIN accuracy to 9) or large pitting occurs, the endurance test is

148 goriva i maziva, 47, 2 : 129-152, 2008.

B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

terminated. A detailed description of the test method can be found in [17]. Figure 8
shows examples of lubricants with different micropitting capacity.
Because of the very long running time of the standard micropitting test an additional
micropitting short test GFKT-C/8,3/90 for screening purposes was developed [18].
The test is run at dip
lubrication conditions and in
a shortened step test
procedure.
The one hour run-in in load
stage 3 is followed by 16 h
each in load stages 7 and
9. The test result is the
evaluation of the
micropitting class. When
the failure criterion of 7,5
m profile deviation is
exceeded after load stage 7
the lubricant is rated with
micropitting capacity low, if
7,5m is exceeded after
load stage 9 the rating is
micropitting capacity
medium and if the profile
deviation after load stage 9
is below 7,5 m the rating is
micropitting capacity high.
A comparison between
results of the standard test
and the short test showed
good correlation (Figure 9).
Predictions of micropitting
results from viscosity
Figure 9: Results of short and standard micropitting test
parameters or additive
content are not possible.
The experimental results of the tests can be introduced into a micropitting capacity
rating method according to Schrade [19].

3.4 Pitting Tests

Lubricant type, viscosity and additive system influence the pitting life of gears.
Especially for automotive gear oils a single stage test method PT-C/9:10/90 for the
evaluation of the relative pitting life was developed. Running an oil in the pitting test
requires a minimum scuffing load stage of at least load stage 9 (pC = 1350 N/mm).
Otherwise the gears will experience scuffing under the conditions of the pitting test.

goriva i maziva, 47, 2 : 129-152, 2008. 149

Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

Gear type C with a specified low surface roughness of Ra = 0.3 0.1 m (PT-C) are
run at a pitch line velocity of v = 8.3 m/s in torque stage 9 for low viscosity gear oils
below VG 100 and in torque stage 10 for medium and high viscosity gear oils of VG
100 or higher (9:10). The gears are dip lubricated with a constant oil temperature of
oil = 90 C (90). A cooling coil with water supply has to be mounted on the top cover
of the test gear box.
A run-in of 2 h in torque stage 6 (pC = 1100 N/mm) is followed by the test run in the
respective torque stage until the failure criterion is reached. Failure criterion is
normally 4 % pitted area on one tooth flank of the pinion. The number of load cycles
until failure is reported and compared to the life of a reference oil. Because of the
rather large scatter of pitting life, at least 3 test runs with one lubricant are required
for a statistically meaningful result. For the pitting failure life a Weibull distribution is
assumed and the load cycles for 50% (L50) failure probability are reported [20]. This
value is compared with the L50 values of reference lubricants. Figure 10 shows
results of different gear lubricants in the standard single stage test PT-C/9/90.

The result of the standard pitting test is often strongly influenced by the occurrence
of micropitting. Therefore
a practice relevant pitting
test PTX-C/10/90 was
developed using
modified test gears PTX-
C with tip and root relief
as well as lengthwise
crowning at the wheel
together with
superfinished surfaces of
pinion and wheel to
suppress micropitting.
The application test PTX-
C/SNC/90 [21] extends
N: number of test runs for each lubricant the test method to two
different load levels
Figure 10: Pitting results in load stage 9 for different lubricants dependent of the result
of the single stage test in
load stage 10. If the
mean pitting life of three runs in load stage 10 is equal or less than 15 million pinion
cycles at least two more test runs are added in load stage 9. If the result in load
stage 10 is over 15 million cycles the test is followed by additional test runs in load
stage 11.
Figure 11 shows the result of a lubricant in the application test.

150 goriva i maziva, 47, 2 : 129-152, 2008.

B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis Metode ispitivanja...

Test results from the pitting test can be introduced into the ISO calculation
procedure [22] by defining a new time strength branch of the SN-curve compared to
the standard SN-curve for the non-EP oil of
same viscosity. The approach is very
conservative because the endurance level is
kept constant. Improvements are therefore
only calculated for gear pairs with limited life.

4. Summary
The standard FZG back-to-back gear test rig
and required modifications for certain tests
are described. Standard and standardized
test methods for the evaluation of lubricant
influence on scuffing, wear, micropitting and
pitting failures of gears are described. Test
results of different gear oils from the market
place and their introduction in calculation
Figure 11: Result of the application test methods are shown.

References
1. Hhn, B.-R., Michaelis, K. and Doleschel, A. Limitations of Bench Testing for
Gear Lubricants, ASTM STP 1404: Bench Testing of Industrial Fluid Lubrication,
West Conshohocken, PA, 2001.
2. Wirtz, H. Schmierstoffe und anwendungsbezogene Schmierstoffprfung, Vortrag
Technische Akademie Wuppertal, 1980.
3. Hhn, B.-R., Oster, P. and Michaelis, K. Influence of Lubricant on Gear Failures -
Test Methods and Application to Gearboxes in Practice, Tribotest journal 11-1,
September (2004).
4. DIN ISO 14635-1 Zahnrder - FZG-Prfverfahren - Teil 1: FZG-Prfverfahren
A/8,3/90 zur Bestimmung der relativen Fresstragfhigkeit von Schmierlen.
5. IP 334 Determination of Load Carrying Capacity of Lubricants, FZG Gear
Machine Method.
6. ASTM D-5182 Standard Test Method for Evaluating the Scuffing (Scoring) Load
Capacity of Oils.
7. CEC L-07-A-95 FZG Gear Machine: Load Carrying Capacity Test for
Transmission Lubricants.
8. ISO 14635-1 Gears FZG test procedures - Part 1: FZG test method A/8,3/90 for
relative scuffing load-carrying capacity of oils.
9. Winter, H. and Michaelis, K. Scoring Load Capacity of EP-Oils in the FZG L-42
Test, Fuels and Lubricants Meeting, Toronto, Oct. 18-21, 1982, SAE Technical
Paper Series 821183, S. 1 - 7.

goriva i maziva, 47, 2 : 129-152, 2008. 151

Metode ispitivanja... B.-R. Hoehn, P. Oster, T. Tobie, K. Michaelis

10. FVA Information Sheet No. 243 Scuffing Test EP-Oils, Method to Assess the
Scuffing Load Capacity of Lubricants with High EP Performance Using an FZG
Gear Test Rig, November 1995.
11. DIN 3990, Teil 4 Tragfhigkeitsberechnung von Stirnrdern: Berechnung der
Fretragfhigkeit.
12. ISO DTR 13989 Calculation of Scuffing Load Capacity of Cylindrical, Bevel and
Hypoid Gears.
13. ASTM Standard D 4998-89 Standard Test Method for Evaluating Wear
Characteristics of Tractor Hydraulic Fluids.
14. DGMK Information Sheet No. 377 Method to Assess the Wear Characteristics of
Lubricants in the FZG Gear Test Rig, Deutsche Wissenschaftliche Gesellschaft
fr Erdl, Erdgas und Kohle e.V., 1996.
15. Plewe, H.-J. Untersuchungen ber den Abriebverschlei von geschmierten,
langsam laufenden Zahnrdern. Diss. TU Mnchen, 1980.
16. Winter, H. and Oster, P. Influence of the Lubricant on Pitting and Micro Pitting
(Grey Staining, Frosted Areas) Resistance of Case Carburized Gears - Test
Procedures, AGMA Technical Paper 87FTM10.
17. FVA Information Sheet No. 54/I-IV Test procedure for the investigation of the
micro-pitting capacity of gear lubricants, July 1993.
18. DGMK Information Sheet: Short Test Procedure for the micropitting load capacity
of gear lubricants. 2005.
19. Schrade, U. Einflu von Verzahnungsgeometrie und Betriebsbedingungen auf
die Graufleckentragfhigkeit von Zahnradgetrieben, Diss. TU Mnchen, 1999.
20. FVA Information Sheet No. 2/IV Influence of Lubricant on the Pitting Capacity of
Case Carburized Gears in Load-Spectra and Single-Stage-Investigations, July
1997.
21. FVA Information Sheet No. 371 Practice Relevant Pitting Test, October 2003.
22. ISO DIS 6336 Calculation of Load Capacity of Spur and Helical Gears.

UDK kljune rijei key words

620.22.05 metode ispitivanje maziva lubricant rig testing methods
ispitnim ureajem
621.892.094 maziva za zupaste prijenosnike gear lubricants
620.1.05 FZG ispitni ureaj FZG gear lubricants testing rig

Authors
Bernd-Robert Hoehn, Peter Oster, Thomas Tobie, Klaus Michaelis
[email protected]
Gear Research Centre FZG, Technische Universitt Mnchen, Germany

Received
19.9.2007.

152 goriva i maziva, 47, 2 : 129-152, 2008.