100% found this document useful (1 vote)
144 views46 pagesIso TR 13989 1 2000 PDF
Uploaded by
冯俊锋Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
100% found this document useful (1 vote)
144 views46 pagesIso TR 13989 1 2000 PDF
Uploaded by
冯俊锋Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 46
TECHNICAL ISO/TR
REPORT 13989-1
First edition
2000-03-15
Calculation of scuffing load capacity of
cylindrical, bevel and hypoid gears —
Part 1:
Flash temperature method
Calcul de la capacité de charge au grippage des engrenages cylindriques,
coniques et hypoides —
Pattie 1: Méthode de la température-éclair
Roforonce number
ISO/TR 13989-1:2000(E)
© 180 2000 ISO/TR 13989-1:2000(E)
POF disclaimer
This POF tle may conta ombacted typataces, In aecordance with Adobe's leanaing poly, this lo may be printed er viowed but ehall not
be ediied unless the typefaces which are embedded are lcensed lo and insialled on the computer perfoming the ediirg. In downioading
ths te, partes accep! herein the responsBalty of nt ringing Adobe's censing poly. The ISO Certval Sectetaret accepts no list n
this area
‘Adobo is a tradomatk of Adobo Systome Inoorporate.
Details of the software products used to create this POF fle can be found in the General Info reatve to the fla; the POF-crection
paramotore were optimized for printing. Every caro has boon takon fo oreure tha tho filo ¢ suiteblo fer uso by ISO mombor bodies. Inthe
Unley event that a problem relaing 1 its feund, please inform the Centrel Secretariat at he address given betow.
@ 1$0 2000
Allnghisresorvod. Unis otherwise spectod, no par of tis publation may be reproduced or utzed in any form or by ary means, elotronic
‘or mechanical, including photocopying and microti, without permission in witing from either ISO atthe address below or ISO's member body
In he county of he requestor.
180 copyright ottce
Case postale 56 » CH-1211 Geneve 20
Tol +41 2274901 11
Fax + 41 227941079
Esrail copright@so.ch
Web wen. co.c0
Printed in Switzerlane
i @ SO 2000 ~ Allrighes reserved ISO/TR 13989-
Contents Page
Foreword.
Introduction
1 Scope.
2 Normative references
3 Terms, definitions, symbols and units..
3.1 Terms and definitions
3.2 Symbols and units
4 Scuffing and wear.
5.1 Contact temperature.
5.2 Flash temperature formula.
5.3 Transverse unit load.
5.4 Distribution of overall bulk temperatures ..
5.5 Rough approximation of a bulk temperature.
6 Coefficient of triction..
6.1 Mean coefficient of friction, method A.
62 — Mean coefficient of friction, method B
63 Mean coefficient of friction, method C
7 Parameter on the line of action
8 Approach factor
9 Load sharing factor
9.1 Buttressing factor
9.2 Spur gears with unmodified profiles
a3 ‘ ‘
94
95
96
97
98
99
10 Scutfing temperature and safety.
10.1 Scutfing temperature.
10.2 Structural factor
10.3 Contact exposure tim
10.4 — Scutfing temperature in gear tests.
10.5 Satety range
Annexe A (informative) Flash temperature formula presentation.
Annexe B (informative) Optimal profile modification
Bibliography.
@ 150 2000 ~ At rights reserved il ISO/TR 13989-1:2000(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards bodies (ISO
member bodies). The work of preparing International Standards is normally carried out through ISO technical
committees. Each member body interested in a subject for which a technical committee has been established has
‘the right to be represented on that committee. Intemational organizations, governmental and non-governmental, in
liaison with ISO, also take part in the work. ISO collaborates closely with the International Electrotechnical
‘Commission (IEC) on all matters of electrotechnical standardization,
The main task of technical committees is to prepare International Standards, but in exceptional circumstances a
technical committee may propose the publication of a Technical Report of one of the following types:
— type 1, when the required support cannot be obtained for the publication of an International Standard,
Gespite repeated efforts;
— type 2, when the subject is still under technical development or where for any other reason there is the
future but not immodiate possibilty of an agreement on an International Standard;
type 3, when a technical committee has collected data of a different kind from that which is normally
published as an International Standard "state of the art", for example).
Technical Reports of types 1 and 2 are subject to review within three years of publication, to decide whether they
can be transformed into International Standards. Technical Rogorts of type 3 do not necessarily have to be
reviewed until the data they provide are considered to be no longer valid or useful.
‘Technical Reports are drafted in accordance with the rules given in the ISO/IEC Directives, Part 3.
Attention is drawn to the possibilty that some of the elements of this part of ISO/TR 13989 may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights.
ISOITR 13989-1, which is a Technical Report of type 2, was prepared by Technical Committee ISO/TC 60, Geers,
Subcommittee SC 2, Gear capacity calculation
This document is being issued_in the Technical Report (type 2) series of publications (according to
subclause G.3.2.2 of Part 1 of the ISO/IEC Directives, 1995) as a “prospective standard for provisional application”
inthe field of scutfing load capacity of gears because there is an urgent need for guidance on how standards in this
field should be used to meet an identified need. In 1975, two methods to evaluate the risk of scuffing were
documented to be studied by ISO/TC 60. It was agreed that after a period of experience one method shall be
selected. Since the subject is still under technical development and there is a future possibilty of an agreement on
an International Standard, the publication of a type 2 Technical Report was proposed.
‘This document is not to be regarded as an “International Standard’. It is proposed for provisional application so that
information and experience of its use in practice may be gathered. Comments on the content of this document
should be sent to the ISO Central Secretariat
A review of this Technical Report (type 2) will be carried out not later than three years after its publication with the
options of: extension for anather three years; conversion into an International Standard: or withdrawal.
ISO/TR 13989 consists of the following parts, under the general tile Calculation of scuffing load capacity of
cylindrical, bevel and hypoid gears:
— Part 1: Flash temperature method
Part 2: Integral temperature method
Annexes A and B of this part of ISO 19989 are for information only.
v @ SO 2000 ~ Allrighes reserved ISO/TR 13989-
Introduction
‘Since 1990 the flash temperature method, presented in this part of ISO/TR 13989, was enriched with research for
short exposure times, consideration of transition diagrams, new approximations for the coefficient of friction, and.
‘completely renewed load sharing factors. In 1991 Prof. Blok contributed an extension of the flash temperature
formula which made it directly applicablo to hypoid gears,
‘Tho integral temperature, presented in ISO/TR 13889-2, averages the flash temperature and supplements empirical
influence factors to the hidden load sharing factor. The resulting value approximates the maximum contact
temperature, thus yielding about the same assessment of scuffing risk as the flash temperature method of this part
of ISO/TR 13989. The integral temperature method is less sensitive for those cases where there are local
temperature peaks, usually in gearsets thal have low cantact ratio or contact near the base citcle or other sensitive
geometries,
The risk of scuffing damage varies with the properties of gear materials, the lubricant used, the surface roughness
of tooth flanks, the sliding velocities and the load. In contrast to the relatively long time of development of fatigue
damage, one single momentary overload can initiate scuffing damage of such severity that affected gears may no
longer be used. According to Blok (12][13][14][15][16][17], high contact temperatures of lubricant and tooth surfaces
at the instantaneous contact position may effect a break-down of the lubricant film at the contact interface.
‘The interfacial contact temperature is conceived as the sum of two components:
— the interfacial bulk temperature of the moving interiace, which, if varying, does so only comparatively slowiy
For evaluating this component, it may be suitably averaged ‘rom the two overall bulk temperatures of the two
rubbing teeth. The latter two bulk temperatures follow trom the thermal network theory [18].
— the rapidly fluctuating flash temperature of the moving faces in contact. Special attention has to be paid to the
coefficient of friction. A common practice is the use of a coelficient of friction valid for regular working
conditions, although it may be stated that at inciaiont scuffing the coefficient of friction has significantly higher
values.
‘The complex relationship between mechanical, hydrodynamical, thermodynamical and chemical phenomena was
the objective of extensive research and experiments, which may induce various empirical influence factors. A direct
suppletion of empirical influence factors may enforce the related functional factors in the main formula to be fixated
to average values. However, correct treatment of functional factors (0.g. coefficient of friction, load sharing factor,
thermal contact coefficient) keeps the main formula intact, in confirmation with the experiments and practice.
Next to the maximum contact temperature, the progress of the contact temperature along the path of contact
provides necessary information to the gear design.
@ 150 2000 ~ At rights reserved v TECHNICAL REPORT ISO/TR 13989-1:2000(E)
ig load capacity of cylindrical, bevel and
Calculation of scuf
hypoid gears —
Part 1:
Flash temperature method
1 Scope
This part of ISO/TR 13989 specifies methods and formulae for evaluating the risk of scuffing, based on Biok’s
contact temperature concept.
The fundamental concept according to Blok is appicabio to ali machine elements with moving contact zones. The
flash temperature formulaa are valid for a band-shapod or approximately band-shaped Hertzian contact zone and
working congitions characterized by sutficienty high Péclet numbers.
2 Normative references
The following normative documents contain provisions which, through reference in this text, constitute provisions of
this part of ISO/TR 13989. For dated references, subsequent amendments to, or revisions of, any of these
publications do not apply. However, parties to agreements based on this part of ISO/TR 13989 are encouraged to
investigate the possibiliy of applying the most recent editions of the normative documents indicated below. For
undated references, the latest edition of the normative document referred to applies. Members of ISO and IEC
‘maintain registers of currently valid International Standards
ISO 1122-1:1998, Vocabulary of gear terms — Part t: Definitions related to geometry.
ISO 6336-1:1996, Calculation of load capacity of spur and helical gears — Part 1: Basic principles, introduction and
general influence factors.
ISO 10300-1-—), Calculation of fozd capacity of bevel gears — Part 1: Introduction and general influence faciors.
ISO 10825-1995, Gears — Wear and damage io gear teeth — Terminology.
3 Terms, definitions, symbols and units
3.1. Terms and definitions
For the purposes of this part of ISO/TR 13989, the terms and definitions given in [SO 1122-1 and ISO 10825 apply
3.2 Symbols and units
The symbols used in this part of ISO/TR 13989 are given in Table 1. The units of length metre, millimetre and
‘micrometre are chosen in accordance with common practice. To achieve a "coherent" system, the units for By, c»
Xu are adapted to the mixed application of metre and milimetre or millimetre and micrometre.
1), Tobe published.
© 150 2000 ~ Al righs esewved 1 ISO/TR 13989-1:2000(E)
Table 1— Symbols and units
‘Symbol Description Unit Rotorence
a centre distance mm Eo. (AS)
b facewidth, smaller value for pinion or wheel @ mm Eq. (11)
bent effective tacewidth mm Eq. (12)
by semi-width of Hertzian contact band mm Ea. (3)
By__| thermal contact coefficient Ni(am’*.m*s!*K) Eq (A.13)
Byn__| thermal contact coetticient of pinion Ni(am’*sm“*s!*K) Eq. (3)
Buz__| thermal contact coetticient of wheel Ni(enm’*-m“*s'*K) Eq. (3)
Cat tip reliet of pinion um Eq. (48)
Ca2_| tip roliet of whee! um Eq, (46)
Cot optimal tip relief xm Eq. (46)
Coqi__| equivalent tip relief of pinion xm Eq, (B.2)
Coge__| equivalent tip relief of wheel xm, Eq, (B.3)
Cn toot relief of pinion xm Eq. (B.3)
Ca toot relief of wheel um Eq. (B.2)
cm__| specific heat per unit mass of pinion Jika-k) Ea. (9)
cuz__| specific heat per unit mass of whee! JAka-6) Eq. (10)
cy mesh stifiness Ni(mm-um) Eo. (B.1)
dy reference diameter of pinion mm Eq. (34)
dp reference diameter of wheel mm Eq. (35)
dgy__| tip diameter of pinion mm Eq. (4)
dzg__| tip diameter of wheel mm Eq. (35)
Ey modulus of elasticity of pinion Nimm2 Eq. (A.10)
Ez modulus of elasticity of wheel Nimm? Eq. (A.10)
E teduced modulus of elasticity Nimm? Eq. (A.9)
Fox _| extemal axial force N Eq. (18)
Fa__| normal oad in wear test N Fig. 1
Fr nominal tangential force N Eq. (11)
Hy auxiliary dimension mm Eq. (B.3)
Hz auxiliary dimension mm Eq. (B.2)
Jama _| tip height in mean cone of pinion mm Eq. (43)
Jima | tip height in mean cone of wheel mm Eq. (44)
2 6 180 2000 ~All rights reserved ISO/TR 13989-
‘Table 1 — Symbols and units (continued)
Symbol Description Unit Roference
Ka __| application factor = Eq. (11)
Keq__| transverse load factor (scuffing) = Eq. (11)
Key _| face load factor (scuffing) = Eq. (11)
Kiia__| transverse load factor (contact stress) - Ea. (18)
Kyp__| face load factor (contact stress) = Ea. (14)
King _| muitiple path factor = Eq. (11)
Ky __| dynamic factor = Eq. (11)
‘niq__| normal module mm Eq. (8.2)
m___| revolutions per minute of pinion rimin Eq. (6)
‘np | number of mesh contacts = Ea. (16)
Pey__| Péciet number of pinion material = Eq. (9)
Pea __ | Péclet number of whee! material = Ea, (10)
2 | quality grade = Ea. (57)
Ray _| tooth flank surface roughness of pinion xm Ea, (28)
Raz __| tooth flank surface roughness of wheel xm Ea, (28)
Rm __| cone distance of mean cone mm Ep, (A.16)
mi __| reference radius in mean cone of pinion mm Ea, (43)
'ma__| reference radius in mean cone of wheel mm Eq, (44)
Se _| safety factor for soutting = Eq, (100)
Sez¢_| load stage (in FZG test) = Eq. (99)
t___| contact exposure time of pinion us Eq. (95)
12___| contact exposure time of whee! us Eq. (96)
fe | contact exposure time at bend of curve us Ea. (97)
‘tmac _| longest contact exposure time us Ea, (95)
«| gear ratio = Eq. (A8)
wy _| virtual ratio = Eq. (8.6)
ry _| sliding velocity mis Fig. 1
Ygt__| tangential velocity of pinion mis Eq. (3)
Yge__| tangential velocity of whee! mis Eq. (3)
vgsc_| Sum of tangential velocities in pitch point mis Ea. (25)
v__| pitch tine velocity mis Eq, (26)
wen__| normal unit load Nimm Eq. (3)
we__[ transverse unit load Nimm Eq. (5)
© 150 2000 ~ Al righs esewved 3 ISO/TR 13989-1:2000(E)
‘Table 1 — Symbols and units (continued)
‘Symbol Description Unit Reference
Xout__| buttressing factor = Eq. (54)
ous _ | buttressing value = Eq. (51)
Nowe | buttressing value = Eq. (51)
X¢__ | geomevyy factor = Eq.(A5)
Xy__| approach factor = Eq. (3)
XL__| lubricant factor = Eq. (25)
Xu thermo-elastic factor KN 6% Eq. (5)
Xmp__ | multiple mating pinion factor = Eq. (22)
Ya__| roughness factor = Eq. (25)
X%3__| lubrication system factor = Eq. (22)
Xw__| structural factor = Eq. (94)
Xap __| angle factor = Eq. (A6)
X¢__| load sharing factor = Eq. (3)
X= __| gradient of the scuffing temperature = Eq. (97)
21__ | number of teeth of pinion’ = Eq. (30)
<2__ | number of teath of wheel ~ Eq. (30)
iq_| transverse tip pressure angle of pinion : Eq. (31)
@a2__| transverse tip pressure angle of wheel a Eq. (20)
| ansverse pressure angle : Eq, (34)
Gan _| normal working pressure angle * Eq. (A2)
ce_| transverse working pressure angle : Eq. (7)
ej1__| pinion pressure angle at arbitrary point . Eq. (29)
B__| helix angle 3 Eq. (18)
o__| base helix angie a Eq. (49)
Pam _| base helix angie in midcone é Eq. (50)
Bu__ | working helix angle . Eq. (A2)
Tx__| parameter on the line of action at point A = Eq. (24)
Taa__ | parameter on the line of action at point AA = Eq. (68)
Tas __ | parameter on the line of action at point AB = Eq. (66)
Tau | parameter on the line of action at point AU = Eq. (49)
Ta__| parameter on the line of action at point B = Eq. (31)
Tee __ | parameter on the line of action at point BB = Eq. (70)
Tp _| parameter on the line of action at point D = Eq. (32)
Too _| parameter on the line of action at point DD = Eq. (72)
Tor _ | parameter on the line of action at point DE = Eq. (67)
SO 2000 - Allsights reserved ISO/TR 13989-
‘Table 1 — Symbols and units (continued)
Symbol Description Unit Reference
Te__| parameter on the Ine of action at point E = Eq. (24)
Tee _| parameter on the line of action at point EE = Eq. (74)
Tey__| parameter on the line of action at point EU = Eq. (49)
Tin__ | parameter on the line of action at point M = Ea, (86)
iy parameter on the line of action at arbitrary point = Eq. (7)
4__ | angle of direction of tangential velocity of pinion = Eq. (3)
2 ___| angle of direction of tangential velocity of whee! = Eq. (3)
1 __| pitch cone angle of pinion 7 Eq. (37)
| pitch cone angle of wheel : Eq. (38)
f_| transverse contact ratio = Eq. (76)
4 __| overlap ratio = Ex, (52)
‘noi | absolute (dynamic) visoosity at oll temperature mPas Eq. (27)
25 __| contact temperature c Eq. (1)
max _| maximum contact temperature c Eq. (2)
‘i __| flash temperature K Eq. (1)
im _ | average flash temperature K Eq, (22)
@hinax | maximum flash temperature K Eq. (2)
@imast_| maximum flash temperature at test K Eq. (94)
Ox__ | bulk temperature © Eq, (22)
nm __| interfacial buk temperature © Eq. (1)
kn [ bulk temperature of pinion teeth c Ex. (20)
@\2_| bulk temperature of whee! teeth c Ex. (20)
xt _| bulk temperature at test c Eq. (94)
i | oil temperature betore reaching the mesh © Eq, (22)
© _| scuffing temperature c Eq, (94)
%¢__| scuffing temperature at long contac time c Eq. (97)
’ui__| heat conductivity of pinion NisK} Eq. (9)
Zaz | heat conductivity of whee! Nis) Ea. (10)
1 _| coeticient of friction in pin-and-ring test =
tm | mean coefficient of friction = Eq. (3)
v4__| Poisson's ratio of pinion material = Ex, (A.10)
¥2__ | Poisson's ratio of wheel material = Eg, (A.10)
© 150 2000 ~ Al righs esewved 5 ISO/TR 13989-1:2000(E)
Table 1 — Symbols and units (concluded)
‘Symbol Description Unit Reference
Aut | density of pinion material kgime Eg. (9)
xe_| density of whee! material kgm? Eq. (10)
Pic __| relative radius of curvature at pitch point mm Eq. (25)
y1__| radius of curvature at arbitrary point of pinion mm Eo. (5)
ay2_| radius of curvature at arbitrary point of whee! mm Eq. (5)
Pret __| telative radius of curvature at arbitrary point mm Ea. (5)
= __| shatt angie : Eq. (A.15)
© _| quill shatt twist : Eq. (17)
2 _The term wheel is used for the mating gear of a pinion,
4 Scuffing and wear
4. Occurrence of scuffing and wear
When gear teeth are completely separated by a full fluid film of lubricant, there is no contact between the asperities
of the tooth surtaces, and usually there is no scutfing or wear. Here, the costticient of friction is rather low. In
‘exceptional cases a damage similar to scuffing may be caused by a sudden thermal instability [19] in a thick oil film,
which phenomenon is not treated here.
For thinner elastohydrodynamic films, incidental asperity contact takes place. As the mean film thickness
decreases, the number of contacts increases acccordingly. Abrasive wear, adhesive wear or scuffing becomes
possible, Abrasive wear may occur due to the rolling action of the gear teeth or the presence of abrasive particles in
the lubricant. Adhesive wear occurs by localized welding and subsequent detachment and transfer of particles from
cone or both of the meshing teeth, Abrasive or adhesive wear may not be harmful if itis mild and if it subsides with
time, as in a normal run-in process.
In contrast to mild wear, scuffing is a severe form of adhesive wear that can result in progressive damage to the
gear teeth. In contrast to pitting and fatigue breakage which shaw a distinct incubation period, @ short transient
overloading can result in scuffing failure
Excessive aeration or the presence in the lubricant of contaminants such as metal particles in suspension, or water,
also increases the risk of scuffing damage. After scufting, high speed gears tend to sutfer high levels of dynamic
loasing due to vibration which usually cause further damage by scuffing, piting or tooth breakage.
In most cases, the resistance of gears to scuffing can be improved by using a lubricant with enhanced anti-scuft?’
aciditves. It is important, however, to be aware that some disadvantages altend the use of anti-scuff additives:
corrosion of copper, emritlement of elastomers, lack of world-wide availability, etc
The methods described are not suitable for “cold scuffing’ which is in general associated with low speed. under
approx. 4 ms, through-hardened heavily loaded gears of rather poor quality.
4.2 Transition diagram
The lubrication condition of sliding concentrated steel contacts, which operate in a liquid lubricant, can be described
[20](21](22](23) in terms of transition diagrams. A transition diagram according to Figure 1 is considered to be
applicable to contacts functioning at constant oil bath temperature.
2} Tho less correct designation Extreme Pressure, EP, is replaced by ant-scutt
6 6 180 2000 ~All rights reserved ISO/TR 13989-1:2000(E)
‘At combinations of normal force #, and relative sliding velocity vq which fall below the line A1-S, in region |, see
Figure 1, the lubrication condition is characterized by a coetficient of friction of about 0,1 and a specific wear rate of
10-2 mm3i(N-m) to 10-© mm&/(N m) (i.e. volume wear per unit of normal force, per unit of sliding distance)
If, with 1g not above a value according to point &, the load is increased into region Il, a transition into a second
condition of lubrication occurs, This mild wear lubrication condition is characterized by a coefficient of friction of
about 0,3 to 0,4 and a specific wear rate of 1 mm3/(N-m) to 5 mm/(N-m).
a2
f
(itd wear)
501 °C
u
502 °C
= 0,25..0,35 16s 080
(scuffing - severe wear)
497°C
96
at
40°C
102 °C
225
Normal force F,,
B «0,10
(no wear" or extremely
nild wear)
498 °C
a3
6001 001 Od 1 10
Relative sliding velocity vg, m/s
Figure 1 — Transition diagram for contraform contacts with example of calculated contact temperatures
If load is increased still further, a transition into a third condition of lubrication, region Ill, occurs at intersection of the
line A2-S. This region is characterized by a coefficient of friction equal to 0,4 to 0,5. The wear rate, however, is
considerably higher, i.e. 100 mmN-m) to 1 000 mm®/(N.m), than in regions | and Il and the wom surfaces show
evidence of severe wear in the form of scuffing. If load increases at relative sliding velocities beyond point S, a
direct transition from region | to region Il takes place.
There is strong evidence that the position of the line A1-S-A3 depends upon lubricant viscosity [24] as well as upon
Hertzian contact pressure [20][21]. At combinations of Fy and vg that fall below this line, it is believed that the
surfaces are kept apart by a thin lubricant film which is, however, penetrated by roughness asperities. In this
Context, the term “partial elastohydrodynamic lubrication” has been used [21],
In region lil liquid film effects are completely absent. This region is identical to the region of "incipient scuffing” [25].
‘There is evidence that the transition which occurs at intersecting the line A2-5 is associated with reaching a cntical
value of the contact temperature. This is the fundamental concept according to Blok.
‘The transition diagram shown is applicable to newly assembled, i.2. unoxidized steel contacts, as occur in gears,
‘cams and followers, etc. It has been found that the diagram is applicable to tour-ball as woll as to pin-and-ring test.
results.
‘Along curve A1-S-A3 temperature ranges from an oil bath, respectively overall bulk, respectively interfacial bulk
temperature, of 28 °C at vy =0,001 m/s to a contact temperature of 498 °C at 1-10 mis, This temperature
behaviour strongly suggests thal the collapse of (partial) elasto-hydrodymamical lubrication does mot occur al a
constant contact or interfacial bulk temperature, for instance being associated with melting of chemisorbed material.
Instead, the pronounced decrease of load carrying capacity with increasing sliding velocity is supposed to be cue to
decreasing viscosity [24](26][27][28][29]
Contrary fo the above, calculated contact temperatures along curve A2-S-A3 tend to attain a constant value, e.g. in
the case of AISI 52100 steel specimens approximately 500 °C; see Figure 1. This suggests that the III transition is
associated with a transtormation in the steel, causing the wear mechanism of surfaces to change from mildly
adhesive to severely adhesive, perhaps involving a mechanism of thermo-elastic instability [30][31].
© 150 2000 ~ Al righs esewved 7 ISO/TR 13989-1:2000(E)
Therefore, the results indicate scutting is associated with a critical magnitude of the contact temperature. For steel,
lubricated with minoral ols, the critical magnitude does not depend on load, volocity and geometry, and equals near
500 *C.
4.3 Friction at incipient scuffing
‘As shown in the transition diagram, Figure 1, in the case of scuffing the coefficient of friction leaps from about 0,25
to about 0.5. The corresponding contact temperature proves to be about 500 °C. This contact temperature is the
sum of a measured interfacial bulk temperature of 28 °C and a calculated flash temperature of 470 °C. During the
flash temperature caiculation use is made of the coefficient of friction just before transition, = 0,35. If this method
has to be applied not only for pin-and-ring tests but also (during the design stage) for gear transmissions. one shall
agree upon the choice of the value of the critical magnitude of the contact temperature on one hand and the value
ofthe coetficient of friction to be used in the calculations on the other.
‘A gear load capacity can be predicted
50
— onthe safe side, with the coefficient of fretion r=
— accurately, with the coefficient of friction between s:= 0.25 and 1 = 0,35, dependent on the lubricant;
— according to previous practice, with a low coefficient of friction of regular working conditions, provided that the
limiting contact temperature is correspondingly low.
In torms of previous practice, for non-additive and low-additive mineral oils, each combination of oil and rolling
matorials has a critical scuffing temperature which, in general, is constant regardless of the operating conditions,
load, velocity and geometry.
For high-adgitive and certain kinds of synthetic lubricants the critical scuffing temperature may well vary trom one
set of operating conditions to another. So, this critical temperature must then be determined for each such set
separately {rom tests which closely simulate the operating condition of the gearset
5 Basic formulae
5.1 Contact temperature
As already mentioned in the introduction, the contact temperature is the sum of the interfacial bulk temperature,
yn. See 5.3, and the flash temperature, A}, see 5.2,
Og = Oni + On a)
%
Poe
A c EP
& Position in the path of contact.
Figure 2— Contact temperature along the path of contact
8 6 180 2000 ~All rights reserved ISO/TR 13989-
‘Only the flash temperature varies along the path of contact; see Figure 2.
‘The maximum contact temperature is
Pomac = Oui + Pimax (2)
where @imax is the maximum value of 4), being located either at the approach path or at the recess path.
Prediction of the probability of sculting is possible by comparing the calculated maximum contact temperature with
a atitcal magnitude. This critical magnitude of the contact temperature can be evaluated from any gear scuffing
tesi, or can be provided by field investigations.
For a reliable evaluation of the scuffing risk, it is important that an accurate value of the gear bulk temperature be
used for the analysis.
5.2. Flash temperature formula
The flash temperature formula of Blok [12][14][16][32] in @ most general representation, for (approximately)
band-shaped contact and tangential velocities differently directed (as for hypoid gears). see annex A, reads.
Li, Yow abs(v91 ~
ona tan XXL Mon = ga)
VP) Bar ylvat sings) + Bua flvge
3
inye)
For cylindrical or bevel gears, with band-shaped contact and parallel tangential velocities, the general
representation, see annex A, reads
Pan Xr °X 5B abs(vgt— vga)
ae Tem) Bar yen + Pe ooo a
or, n an equivalent representation,
04 = 252 ty HX Kr ved [25] 208 ( oni — fara) 6
where
dm is the mean coefficient of friction (see clause 6);
Xy_ is the thermo-elastic factor (see annex A);
Xu = 50 KN-M%5m-”2mm for commonly applied steel;
Xy__ is the approach factor (see clause 8);
X;_ is the load sharing factor (see clause 9);
wpr_ is the transverse unit load (see 5.3), in Nimm;
{she rotational speed of pinion, in rimin
‘yet is the local relative radius of curvature, in ma:
Pyrat tee ©
© 150 2000 ~ Al righs esewved 9 ISO/TR 13989-1:2000(E)
2, is the local radius of curvature of pinion flank, in mm:
vt
14h
‘a-sinayy (cylindrical gears) ~
Pye
Pyq_'8 the local radius of curvature of wheel tlank, in mm
if
Pye =, Fa sina, (eylindrical gears) @
For bevel gears, see equations (37) and (38).
For an adapted representation, see annex A
‘Two Péclet numbers have to be sufficiently high, which is satisfied in almost all cases where scuffing may occur.
For lower Péclet numbers the heat flow from the contact band into the gear teeth causes a different temperature
distribution for which formulae (8) to (6) are not valid
25 @
Yea ta Poa e , 5 (10)
Ane sine
Pep =
’\i is the density of pinion material, in kg/m;
‘x2 Is the density of wheel material, in keys;
cys is the specitic heat per unit mass of pinion, in Ji(kg:K):;
heat per unit mass of wheel,
eve is the spec Akg K);
Zany, Is the heat conductivity of pinion, in Nis K);
2x2 _ is the heat conductivity of wheel, in Ni(S-K)..
For cylindrical and bevel gears, sin:
5.3 Transverse unit load
The transverse unit load for cylindrical gears and bevel gears is
= Ra Kap ax Bg 7 litical gears) “
fron = Rep Rage Bi Eiap (bevel gears) (12)
7
whore
is the nominal tangential force on piteh circle, in Ni;
bis the facewidth, in mm;
batt = 0,85 (13)
10 150 2000 ~ Ai rights reserved ISO/TR 13989-
‘Ka is the application tactor (see ISO 6336-1 for cylindrical gears, ISO 10300-1 for bevel gears)
Ky _is the dynamic factor (see ISO 6336-1 for cylindrical gears, ISO 10300-1 for bevel gears):
Kap. is the face load factor;
Kap = Kup (see ISO 6336-1 for cylindrical gears, ISO 1030-1 for bevel gears); (14)
Kou. 1s the transverse load factor:
Kou. = Kia (886 ISO 6336-1 for cylindrical gears, ISO 1030-1 for bevel gears); (15)
Krmp is the muttiple-path factor:
‘The multiole-path factor Kp accounts for the maidistribution in muttiple-path transmissions depending on accuracy
and flexibility of the branches. If no relevant analysis is available, the following may apply.
— for epicyclical gear trains with np planets (11g > 3)
King
+025 )ny —3 (18)
— for dual tandem gears with quill shaft twist « degrees under full load
Kimp 14 (02/0) a7
— for double helical gears with an external axial force Fey
k,
Kp = 1+ 18)
me (18)
— tor other cases
Kmp=1 (19)
5.4 Distribution of overall bulk temperatures
The friction loss most typical of gear transmissions is the one caused by the meshing zone. In this source the heat
is generated mainly by tooth friction. The mechanical "pumping" energy expended for sideways expulsion of
supor'iuous oil may sometimes be far from negligible. The other unavoidable friction loss is trom the bearings,
either of the rolling or the sliding type. In high speed gear transmissions, sliding bearings may well generate much
more frictional heat than gears. Other heat sources are oil churning and friction from seals. All the above heat
sources have the following features in common
in each of these sources the fluid friction depends on some oil viscosity representative of the operating
condition;
— allof the heat sources are thermally interconnected through transmission elements to the sinks, such as the
ambient air or the cooling system.
The thermal interconnection allows calculation concepts such as:
— finite element methods for discrete components;
bondgrah methods,
— thermal network analogue methods [18]
© 1502000 Airighisresewed 1" ISO/TR 13989-1:2000(E)
The interfacial bulk temperature Gy, may be suitably averaged from the two overall bulk temperatures of the teeth in
contact, Qj and @\y. The following formula is valid to a good approximation (at high values of the Péclet
numbers):
_ Pun \at Cus * Be Jue Ove
Ou (20)
Bhar Vion * Bae roe
aur yh
In a fairly wide range of the ratio aoe a simple arithmetic average is valid o a reasonable approximation
2 yz
1
Omi = 5° (Ons * Ome) (21)
Bulk temperatures in excess of 150 °C for long periods may have an adverse etfect on the surface durability
5.5 Rough approximation of a bulk temperature
For very rough inquiry the bulk temperature may be estimated by the sum of the oil temperature, taxing into account
some impediment in heat transfer for spray lubrication, and a part which depends mainly on the flash temperature,
of which the maximum value is taken
Oy = Ogi + 0.47 Xs Xp “Om (22)
where
Ng =1,2 for spray lubrication:
.0 for dip lubrication;
Xge
,0. for meshes with additional spray for cooling purpose;
Xg=0,2 for gears submerged in oll, provided sufficient cooling;
t+,
3 for a pinion with n, mating gears; (23)
im is the average of flash temperature along path of contact, in °C:
[fever
an
te TeTn
(24)
However, for a reliable evaluation of the scuffing risk, it is important that instead of a rough approximation, an
accurate vaiue of the gear bulk temperature be used for the analysis.
6 Coefficient of friction
Several factors intluencing the friction between gear teeth vary throughout a meshing oycle. On one of the two
mating tooth tacos the relative motion is uniformly accelerating, on the other itis uniformly decolorating. Only at
pitch point position pure rolling occurs. In any other meshing position combined roling and sliding will occur. Also
the load acting on two mating tooth faces will vary from one meshing position to another. These conditions cause a
continuous variation of the film thickness, the lubrication regime and the coefficient of friction. Even in a similar
meshing position the coefficient of friction may vary for different teeth and different time.
12 150 2000 ~ Ai rights reserved ‘The local costticient of friction is considered to be representative quantity valid tor the local point concerned,
‘smoothing various influences. The geomotrically determined variation of the local cootficiont of friction is difficult to
calculate or to measure, hence instead of a local valve, a representative mean value of the coefficient of friction will
bo applied.
‘A mean value (along the path of contact) of the coefficient of friction has commonly been applied, and even that
value is uncertain. Too often, in test reports on friction, important influential quantities were neglected, for instance
the bulk temperature which determines the inlet viscosity and therefore the lubrication regime.
‘The mean coetticient of friction®) s4n, depends on the geometry of the path of contact, the tangential velocities, the
normal load, the inlet viscosity (which is identical with viscosity at teeth bulk temperature), the pressure-viscosity
coefficient, the reduced modulus of elasticity, the surface roughness, the normal relative radius of curvature.
Depending on further investigations, other quantities and influences may have to be accounted for, either in the
formula or in the description of the field of application. The number of quantities may be reduced by dimension
analysis [33], and a possible neglect of some minor influential quantities.
The coefficient of friction may be measured or estimated according to various methods. The limiting contact
temperature shall be chosen correspondingly to the coefficient of friction.
6.1 Mean coefficient of friction, method A
‘The coefficient of friction at the onset of scuffing may be measured in gear tests or pin-and-ring tests. The limiting
contact temperature is correspondingly high.
6.2 Mean coefficient of friction, method B
‘According to previous practice, whereby low coefficients of friction of regular working conditions are used, the final
caloulation of the coefficient of friction may be made with some appropriate formula, i.e. one containing a value of
absolute (dynamic) viscosity m_ that corresponds to the gear bulk temperature. The limiting contact temperature is
correspondingly low, see clause 10.
6.3 Mean coefficient of friction, method C
It at the start of a calculation the bulk temperature is not yet known, the mean coefficient of friction of common
working conditions may be estimated by
\o2
Hm = 0,060. seal Kiel (25)
Yex0 "Prac
where
wet is the transverse unit load, see equation (11) or (12), in Nimm;
gso_ is the sum of tangential velocities in pitch point, in mvs:
vena = 2: sin yg (28)
11 Is the pitch line velocity, in mis (yy > 50 m/s, substitute the value 60 in equation (26), instead of »);
ric is the transverse relative radius of curvature, in mm (see equation (6) tor Fy
Oo;
3). The mean coefficient of fiction is defined as the mean value of the local coefficients of tiction alang the path of contact.
‘though the actual local coefficient of friction at the pitch point vill difer from the mean coetficient of friction defined for the
\whole path of contact, that moan costfcient of tition may be exprassed In terms related to the pitch point.
© 1502000 Airighisresewed 13 ISO/TR 13989-1:2000(E)
Xi is the lubricant factor:
XL=1,0- (noi) for mineral oils;
8 (nei}-° for water soluble polyglycols
.7 (Moi for non water soluble polyglycols; (27)
1.8 (moi) for polyaifaoletins;
3 (7oi)°5 for phosphate esters;
XL= 1,5. (noil-8%5 for traction tids:
‘ig is the dynamic viscosity at oil temperature 65, in mPa's
Xq__ is the roughness factor:
Ray + Rag
2
y* (28)
ma-(
where
Rays the tooth flank surface roughness, Ru, of pinion, for newly manufactured gears, in jm (for
adequately run-in gears Ray may reduce to about 60 % of its intial value);
Rag is the tooth flank surface roughness, Ra, of wheel, for newly manufactured gears, in um (for
adequately run-in gears Rag may reduce to about 60 % of its initial value).
7 Parameter on the line of action
‘The points on the line of action are indicated by a dimensionless linear parameter /5, with the value —1 in the
‘tangent point on the pinion base circle and the value 0 in the pitch point [93]: see Figure 3.
® Base pitch
Figure 3— Parameter on the line of action
14 150 2000 ~ Ai rights reserved ‘At an arbitrary point on the path of contact,
tanays
tandem
Atthe lower end point of the path of contact,
2 {tanags _,)
ry = 2.{ 04025
1 (anew J
‘At the lower point of single pair tooth contact,
ry = fanaa Qn
Tandy
2 taney,
‘At the upper point of single pair tooth contact,
se tana, 2
244 2
tanya lanegg
At the upper end point of the path of contact,
re ta
tanaw
whore the tip pressure angles are defined by
tana) f tet
\\a cose,
z
tanargg = | —“22—] -1
087,
ISO/TR 13989-
(29)
(30)
(31)
2)
(33)
(34)
5)
‘The parameters of bevel gears either may be caloulated with the geometry of virtual quantities, see ISO 10300-1:—,
annex A, of with the following formulae (valid also if the shatt angle 5
At an arbitrary point on the path of contact,
lanays
tana,
py1= Rn tandy sinay -(1+ Fy)
Py2 = Rm “tand;-sinay -(u- Fy)
‘At marked points of the path of contact,
tand, (‘soa ‘|
oy =~ fara, (tana _
tandy (tana
_tandgy _,_ 2:1-c0sd,
Bana tana
@1S0 2000 ~ At rights reserved
5, + dis not equal to 90°)
(36)
(37)
(38)
(39)
(40)
15 ISO/TR 13989-1:2000(E)
_ tand (tence 4), 22085
tand, ( tanay tana
tana
es
Fane,
where the tip pressure angles are defined by
7
asa)
Tans 0086, in
—s_—F-
may
restive]
taney f
5 isthe pitch cone angle of pinion;
isthe pitch cone angle of wheel,
Rm is the cone distance of mean cone (midface of teeth), in mm;
Jigen is the tip height in mean cone of pinion, in mm:
dnan2_ 1S the tip height in mean cone of wheel, in mm;
rm iS the pitch radius in mean cone of pinion, in mm:
rm@__ is the pitch radius in mean cone of wheel, in mm,
8 Approach factor
(at)
(42)
(43)
(44)
‘The approach factor takes empirically into account an increased scutting risk in the beginning of the approach path,
‘due to mesh starting without any previously built up oil film. Its influence is relatively strong for large gears.
‘The approach factar is,
if the pinion drives the wheel (speed reducing),
Xy=1 for 5 >0
a
J - provides ny» 1, fF <0
— if the whee! drives the pinion (speed increasing),
Xy=1 for <0
Got Cas
Ny=t
a 50
provided Xy 2 1, for 7 >0
(49)
(48)
(47)
(48)
16 150 2000 ~ Ai rights reserved ISO/TR 13989-
where
Cett_is the optimal tip relief (see annex B), in jim;
Cet is the tip relief of pinion, um;
Ca2_ 18 the tip relist of whoel, um;
1, isthe parameter of arbitrary point (see clause 7);
This the parameter of point A (see clause 7);
Te is the parameter of point B (see clause 7);
9 Load sharing factor
The load sharing factor X- accounts for the load sharing of succeeding pairs of meshing teeth. By convention, the
load sharing facior is presented as a function of the linear parameter Fy on the path of contact, increasing ai the
approach path of transverse double contact, when the preceding pair of meshing teeth ends its action, and
decreasing al the recess path of transverse double contact, when the succeeding pair of meshing teeth comes into
action.
Due to inaccuracies a preceding pair of meshing teeth may cause an instantaneous increase or decrease of the
theoretical load sharing factor, independent of the instantaneous increase or decrease caused by inaccuracies of a
succeeding pair of meshing teeth at a later time.
The value of Xi: does not exceed 1,00 (for cylindtical gears), which means full transverse single tooth contact. The
region of transverse single tooth contact may be extended by an regularly varying location of a dynamic load.
‘The load sharing factor X; depends on the type of gear transmission and on the profile modification. In case of
buttressing of helical teeth (undersized profile modification), the load sharing factor is combined with a buttressing
factor Xout
Formulae for profile modification (tip relief) are given in annex B.
9.1 Buttressing factor
Helical gears may have a buttressing effect near the end poinis A and E of the line of action, due to the oblique
contact lines. This applies to both cylindrical and bevel gears with tip relief less than an optimum value, Ca < Catt.
Souta Xoutt
1
A UB o we
Figure 4 — Buttressing function
The buttressing is expressed by means of a factor Xp, simplified as linear functions within the ranges A-AU,
AU-EU, EU-E; see Figure 4, marked by the following values:
© 1502000 Airighisresewed 7 ISO/TR 13989-1:2000(E)
Tau Ta = Te Fey = 02-8i0By tor cylindrical gears
Tyy- T= Te ~ Tey =02-8inByy for bevel gears
Xow = Xoue = 13 when a 21
Xoua = Xouie when g <1
Xpurau = Xputeu
Xout for Ta < Fy < Tau
Nout = for Tau < Fy < Tu
for Feu < ly < le
9.2. Spur gears with unmodified protiles
(49)
(50)
(st)
(82)
(53)
(54)
(55)
(58)
The load sharing factor for a spur gear with unmoditied profile is conventionally supposed to have a discontinuous
trapezoidal shape; see Figure 5. However, due to manufacturing inaccuracies, in each path of double contact the
load sharing factor will increase for protruding flanks and decrease for othor flanks [34]. Tho representative load
sharing factor is an envelope of possible curves; see Figure 6.
18
23
wa
Figure 5 — Load sharing factor for cylindrical spur gears with unmasiled profiles and
(a - 215,
Figure 6 — Load sharing factor for cylindrical spur gears with unmodified profiles and
quality grade 8 or coarser
0-21 K-Tn
for Ths Tye
for Ia< ly < lo
(87)
(58)
SO 2000 - Allsights reserved ISO/TR 13989-
for p< Ky 0, 10" Tk < Fy Cot (70)
2 Ct Cot
Tew= Te tor Can < Cen (ry
Cao ~ Cot): Te + (Cao +2 Cot) Fi
Top = G2 GetTE(Ga+2 CHT tg ox 5 Oxy v7
B-Caa + Con
tor Cao < Cen (73)
tor Cay > Cort (74)
for Can < Cen (75)
lical gears with unmodified profiles
Helical gears with a small total contact ratio, ¢, < 2, have stil pure single contact of tooth pairs. Hence, they can be
treated similar to spur gears, considering the geometry in the transverse plane, as well as the buttressing effect.
See Figure 9.
Figure 9 — Load sharing factor for narrow helical cylindrical gears with unmodified profiles, including the
buttressing effect
The load sharing factor is obtained by multiplying the Xin 9.2 with the buttressing factor Xbut
9.5 Narrow helical gears with profile modification
Helical gears with a small total contact ratio, «, <2, have stil pure single contact of tooth pairs. Hence, they can be
‘treated similar to spur gears, considering the geometry in the transverse plane. See Figures 10 and 11
N
AVAA AB BBB. DOD DE EENE
Figure 10 — Load sharing factor for narrow helical cylindrical gears with optimal or oversized profile
‘modification
20 6 180 2000 ~All rights reserved ISO/TR 13989-
A AUAB OB Oo oF UE
Figure 11 — Load sharing factor for narrow helical cylindrical gears with undersized profile modification
‘The load sharing factor is obtained by muttiplying the Xr in 9.2 with the buttressing factor Xeut
9.6 Wide helical gears with unmodified profiles
The buttressing effect [35] of local high mesh stiffress at the end of oblique contact lines for wide helical gears,
> 2, is assumed fo act near the ends A and along the helix teeth over a constant length, which corresponds io a
transverse relative distance 0,2 sinfly: see Figure 12. See also 9.1 and Figure 4.
Mew
A AU OB Oo
Figure 12 — Load sharing factor for wide cylindrical helical gears with unmodified profiles,
The load sharing factor is obtained by multiplying the value 1/c,, representing the mean load, with the buttressing
factor Xeut
(78)
9.7 Wide helical gears with profile modification
The extensions of tip relief al both ends A-AB and DE-E of the path of contact are assumed to be equal and to
result in a contact ratio «, = 1 for unloaded gears; see Figure 13. The load sharing factor for wide cylindrical helical
gears, «22, with undersized or with oversized profile modification follows {rom interpolation or extrapolation,
respectively, between the factor for unmodified profile with buttressing effect and the factor for optimal profile
modification; see Figure 14.
A AB OB oO 0 E
Figure 13 — Load sharing factor tor wide cylindrical helical gears with optimal profile modification
© 1502000 Airighisresewed 24 ISO/TR 13989-1:2000(E)
ai) 4B 8 0
t
Figure 14 — Load sharing factor for wide cylindrical helical gears with oversized profile modi
OE EU E
near A, and undersized near E
The ranges are marked by the following points
Tyg =05-(Ta + Ta)
Toe =05 (p+ Te)
(6a +1)-(€q =1)- (Coo ~ Con)
TTR eA bay (C0 +) Cae
We-Ta)
(a +1)-(€u = 1)-(Car~Cot)
neR
BEE gai Cap *(8ea + Cat
(Te-Ta)
1): Car + (Bea +1-Can Ty Ta
€a-Cot 260 (aD Co Tan —T
(ca-1)
a Gad Con
6 -Cap + (Bea
2-£y (Ea +W Con Te-Toe
9.8 Narrow bevel gears
for Cag > Coit
for Car = Con
for Fan < Fy < Tap lt Cap Cot
for Tan < Ty < Fant Cae > Cat
for Tag < Ty < bE
for Me < My < Te if Cay < Con
for pe < Sy < Tee it Cas > Cort
for Tbe < Sy < Tee tt Ca > Cosh
(77)
(78)
(78)
(80)
(81)
(82)
(83)
(84)
(85)
For narrow bevel gears, ¢,<2, with profile modification C4 < Coy. the load sharing factor Xr is found by linear
interpolation between Xr as calculated (for Ca =0) in 9.4, and Xp as calculated (for Ca = Cai) in 8.9. Remember
Xeut
For narrow bevel goars, ¢ <2, with profile moditication Ca > Cer, the load sharing factor Xj is calculated as in 9.9.
SO 2000 - Allsights reserved ISO/TR 13989-
9.9 Wide bevel gears
For wide bevel gears, ¢,2 2, with optimal profile modification, Cai = Cet, Cae = Cot, the load sharing factor is
assumed to be parabolic [35]; see Figure 15.
A M E
Figure 15 — Load sharing factor for bevel gears with optimal profile modification
‘The midpoint M is defined by
Tale
ry Za )
The load sharing factor for optimal profile modification is
1s_(Ty-lin? 6
for Cat = Cots Can = Cot 87
ges 7)
If the profile modification Cay differs from Cao, then the sections AM and ME shall be calculated separately with a
discontinuity at point M; see Figure 16.
For undersized profile modification an interpolation is made between the factor for unmodified profile with
buttressing effect according to 9.6, and the parabola for optimal profile modification.
For oversized profile modification the parabola has a new end point AA or EE.
A AU ™ EE
Figure 16 — Load sharing factor for bevel gears with undersized profile modification near A and oversized
profile modification near E
For undersized profile mosification, Xr; is found by linear interpolation between Xr for optimal profile modification,
equation (87), and Xr for unmodified profile according to equation (76). This interpolation is to be made stepwise
from Ato M with the influence of Cao, and from M to with the influence of Cat-
For oversized profile modification, new end points AA and EE are found as
(88)
© 1502000 Airighisresewed 23 ISO/TR 13989-1:2000(E)
Teg = Te ~“ (re-ro(S- a
Xp=0 tor Jay < Tan (90)
a [,_ eer | for rane ta< ta o)
6a © Cae Cer | an =F)
15 3 {yr s
oe Fea Gt for Fin < Fy < Tig (92)
Xp=0 tor Fee < Fy < Fe (93)
10 Scuffing temperature and safety
10.1 Scufting temperature
‘The scuffing temperature is the contact temperature at which scutfing is likely to occur with the chosen combination
of lubricant and gear materials. The scuffing temperature is assumed to be a characteristic value for the
material-lubricant-material system of a gear pair, to be detetmined by gear tests with the same
material-lubricant-material system [36],
When using a low-additve mineral oil, the scuffing temperature is assumed to be independent of operating
conditions in a fairly wide range.
When using a mineral oil or a synthetic oil with anti-scutt or friction-reducing additives, extended research is still
needed to determine the nature of a possible non-constancy of the scutfing temperature for the materials and the
‘operating conditions concerned. Special attention shail be paid to the correlation between test conditions and actual
(F design conditions. The correlation may be strongly influenced by properties shown in the transition diagram; see
Figure 1
10.2 Structural factor
The scuffing temperature of low-additive mineral oils that is determined from test gears may be extended to
different gear steels, heat treatments or surface treatments by introducing an empirical structural factor.
05 = Onan + XW Ctinaxt (94)
where
Our is the bulk temperature of test gears, in °C;
pax 's the maximum flash temperalute of test geats, in K;
Xw Is the structural tactor (see Table 2).
24 6 180 2000 ~All rights reserved ISO/TR 13989-
Table 2— Structural factor
Material Xw
Through-hardened steo! 1,00
Phosphated steel 1,25
Copper-plated ste! 1,50
Bath or gas nitrided steel 1,50
Hardened carburized steel, with austenite content:
—less than averago 415
— average (10 % to 20%) 4,00
= greater than average 0.85
Austerite steel (siainless steel) 0,45
However, this approximation is restricted to methods using the coefficient of friction for common working conditions
(see 6.3) together with an average value of the thermo-elastic factor (see clause 8). The structural factor may be
superiluous if methods are used considering realistic values of the coefficient of friction and the thetmo-elastic
factor.
10.3 Contact exposure time
Itwas shown by tests [97] that the scuffing temperature of gears lubricated with anti-scuft oils may be influenced by
the contact exposure time, that is the time curing which a point on a tooth flank is exposed to the Hertzian contact
band of the meshing tooth,
‘The decisive contact exposure time imax for a pair of tooth flanks is the longest of «and 2
by
91
fra > (95)
lnax ty = 2 (98)
‘e
‘The dependence of the scuffing temperature @s on the contact time is approximated as shown in Figure 17 by a
curve consisting of two straight lines
5 = O56 +Xo-Xw (lo !nax) for max to (98)
where
gq is the scuffing temperature at long contact tir
es, in
Xo. is the gradient ofthe scutfing temperature, in Kis:
xy is the structural factor
‘eis the contact exposure time at the bend of the curve, in is
Iyax is the contact exposure time of meshing teeth, in us.
© 1502000 Airighisresewed 25 ISO/TR 13989-1:2000(E)
The following values may be applied for oils:
without anti-scuff adgitives: Xo = 0 Klis, te = 0 nst
— with antisculf additives: Yq = 18 Klus, ic = 18 us,
0 14s tras
Figure 17 — Influence of contact exposure time on the scuffing temperature for anti-scut! oils
10.4 Scuffing temperature in gear tests
The sculfing temperature may be determined in gear tests, such as Ryder [38], FZG-Ryder [39], FZG L-42 [40],
FZG A/8,3/90 [41]
The test result shall be expressed in a scutfing temperature, together with the test conditions. If the test result is
expressed in other terms, then a relation shall be given as, for instance
Og =80+(085+14- Xw) XL -(Spz5)* (99)
whore
Xw is the structural factor (see Table 2);
X___ is the lubricant factor, [see equation (27)];
Sez@_ (is the load stage according to FZG A’8,3/90 test. This is the load stage where scuffing occurs.
However, oil data tend to vary much with regard to Seza, a load stage variation of 4 1 is common, and it is assumed
‘that the oil somewhat dotoriorates during an oil shift interval. Thorotore, calculations may be made with one load
stage less than the specification.
10.5 Safety range
In contrast to the long time for development ot fatigue damage, a single momentary overload can initiate scutfing of
such severity that affected gears may not longer be ‘it for use. This should be carefully considered when choosing
an adequate safety range, especially for gears required to operate with high pitch line velocities.
In cases with a short contact exposure time fax and Safety conditions based on an increased scuffing temperature
5 > @_ (see 10.3) that contact exposure time Zae Shall not increase, unless the transmitted power is lowered
adequately
A safety factor may be defined by
Os Por
max ~ Foi
58 (100)
26 6 180 2000 ~All rights reserved ISO/TR 13989-
whore
@s is the scuffing temperature. in °C;
@emax_ is the maximum contact temperature, in “Ci
jis the oll temperature, in °C.
However, the relation between the actual gear load and the decisive contact temperature is very complicated, and
the use of a safety factor expressed in any quotient of temperatures may cause confusion.
‘Therefore, in addition to the specification of the test load stage (see 10.4) it is advised to express the concept of
safety as a demanded minimum difference (for instance > 50 K) between the scuffing temperature and the
estimated maximum contact temperature.
ar
@1S0 2000 ~ At rights reserved ISO/TR 13989-1:2000(E)
Annex A
(informative)
Flash temperature formula presentation
Since the first publication of the original flash temperature formula [12] [14] Blok made the following conversions:
step from width to semi-width of Hertzian contact band and substitution of 0,83: V2 -1,17 for parabolic friction
heat distribution by 1,11 for eliptic friction heat cistnbution [16];
— extension to unequally directed tangential velocities [32]; see equation (3).
For convenience, exact conversions were made:
‘some quantities were expressed in other quantities, for instance the semi-width of Hertzian contact band and
the radii of curvature:
— some parts of the formula were concentrated in separate factors, for instance the thermo-elastic factor; see
clause A.3.
For practical applications adaptive conversions were made:
— redefinition of factors, for instance the load sharing factor; see clause 9;
— addition of empirical factors, for instance the approach factor; see clause 8
A. General case
In a most general case of tooth contact (e.9. hypoid gears) the successive contact areas will assume the shape of
tapered bands; see Figure A.1. Moreover, the two tangential velocities, vey and vga, are directed at unequal angles,
ny and /2, with respect to the longitudinal axis of such an area. in simpler cases (¢.g. cylindrical gears) the angles
Teduce into 1 =
The distribution of the contact pressures over some cross-section in a tapered contact area may be approximated
by the semi-elliptical distribution that would occur over a substitute band-shaped contact area interposed between
‘two parallel surfaces, whilst naving a uniform width equal to the aforementioned local width; see Figure A.1
Yo
‘igure A.1 — Substitute band-shaped contact area, with two tangential velocities in different directions
28 6 180 2000 ~All rights reserved ‘The actual Hertzian contact zone of hypoid gears may be supposed to be elliptical and the tangential velocities are
noither coincident nor perpendicular to a major axis of the contact zone. However, the ollistical contact may be
rather elongated in having a sufficiently high elliptical ratio, or it may have the shape of a somewhat tapered band.
Either tangential velocity is to show a direction deviating not too much from that of the minor axis. In other words, let
either velocity have a component along the major axis.
For determining the maximum flash temperature sought, the actual elliptic contact zone may be substituted by a
band-shaped contact zone of which the width 2}; equals the length of the minor axis of the ellipse; see Figure A.1
Note that the maximum contact pressure here, like the minor axis, may be directly proportional to the cubic root of
the load, instead of the square root. In some cases, the Herizian formulae nave to be adapted for elongated point
contact [42].
‘To summarize, the present procedure would appear justifiable to a reasonable approximation. A major reason lies
in the feature that, for the actual sufficiently elongated elliotic contacts under the above-defined Kinematic
conditions, one may expect the actual maximum flash temperature to occur at a point fairly close to the minor axis
concerned.
The flash temperature formula of Blok [12](14][16][32] for substitute band-shaped contact and tangential velocities,
differently directed is
Hm Xs°Xp-Wen 808(¥g1 ~ ¥go)
Oa ber tr a esa) mo
where
iim i the mean coefficient of friction:
Xy is the approach factor, see clause 8;
Xj Is the oad sharing factor, see clause 9;
‘wen is the normal unit load, in Nimm
oa aa)
where
vax isthe transverse unit load, see 6.3, in Nimm:
un _ isthe normal working pressure angle, in degrees
ey ~ Arcsin (Sineeyy 0084) (aa)
By _ is the working helix angle, in degrees:
Bu roan neo | aa
by is tho semi-width of Hortzian contact band, in mm;
¥qt__ is the tangential velocity (vector) of pinion, in mis;
© 1502000 Airighisresewed 29 vg2__ isthe tangential velocity (vector) of wheel, in mis;
Byty_ is the thermal contact coef
ient of pinion, see clause A., in NAmm"®n"®.5"%K);
By2__ isthe thermal contact coefficient of whee', see clause A.3, in N/(mm'*m"*s"*K);
v4 Is the angle of direction of tangential velocity of pinion, in degrees;
2 is the angle of direction of tangential velocity of wheel, in degrees.
A2 Cylindrical gears
The flash temperature formula adapted? for cylindrical gears reads
aia yt?
in XX Nor we eg (As)
oT
whore
Jy is the mean coefficient of friction (see clause 6);
Xqt_ is the thermo-elastic factor (see clause A.3);
Xj Is the approach factor (seo clause 8);
Xg__ is the geometry factor
for an external gear pair:
ue abs(1+ Fy -1- 7 /u)
NXg = 051 Xyg (ue t)?,— (A8)
: OR Wy
for an internal gear pair (practical sign convention):
abs (Ie 7-14 75
Xq=051 Xap -(u— I"? sealtey-te 4) an
(1455) (ue ry)
X; is the load sharing factor (see clause 9);
wat is the transverse unit load (see 5.3)
% is the pitch line velocity;
a isthe centre distance.
4) To avoid possibie misinterpretation of the unit of the rotational frequency, the formula is expressed in pitch tine velocity and
centre distance, instead of more logically the rotational frequency and cenire distance. The old-iashioned interpretation of » is
revolutions per minute withthe unit rimin. Any attempt to redefine ‘revolutions per time" in order to obtain a coherent system of
units fails, since the unit 1/s has a double meaning, either 260°-angla/s or radian’s. The deeper cause of this ambiguity in the
international system of units is @ lacking dimension for the quantity angle and an ill-considered omission of the unit radian in
too many cases. The solution is to reduce the “quantity revolution into the "phenomenon" rotational frequency with the
requeney unit Hz
30 6 180 2000 ~All rights reserved ISO/TR 13989-
‘The influence of the normal pressure angle and the helix angle is given by
Xyg —122-sin™ ary, COS" C08"? fy (Aa)
However, the factor Kay proves to be so close) to unity, see Table A.1, that it can be approximated by the value
1,00.
Table A.1 — Factor X.9
Xap
a p=0° A= 10° B= 20° B= 30"
= 20,000° = 20,284° = 21,178° = 22,796°
18 0947 = = =
20° 0978 0975 0.966 =
2 1,007 1,004 0,995 0,981
24° 4,095 4,032 4,028 1,008
26° 1,064 1,060 1,051 1,096
28° = = = 1,063
A3 Thermo-elastic factor
‘The thermo-elastic factor accounts for the influence of the material properties of pinion and wheel
ve Ce )? alter a)
_ (as)
Baar (1+ Fy) + Be (1 Fy 10)
whore
is the reduced modulus of elasticity, in Nimm?
— (10)
(109) y+ (1-8)
whore
Ey is the modulus of elasticity of pinion material, in Nimm2;
F2_ is the modulus of elasticity of wheel material, in Nimm2;
¥ _ is the Poisson's ratio of pinion material
¥p_ is the Poisson's ratio of wheel materia!
Byy is the thermal contact coefficient of pinion, in Ni(mm"”®.m"®.s'"K);
5)_ The constant 1,22 was introduced in the numerator of (A.8) and in the denominators of (A.6) and (A.7) as 0,51 = 0.6211,22
‘wath no other purpose than to simpily (8.8)
© 1502000 Airighisresewed 3t (ast)
thie = (0001 Aa me ove)!” (aig)
where
Aun; is the heat conductivity of pinion material, in Nis:
Aye is the heat conductivity of whee! material, in Ni(s-K);
‘>M1 the density of pinion material, in kg/m?
Aue Is the density of wheel material, in kg/m;
cums the specific heat per unit mass of pinion material, in Jitkg):
cuz ‘the specitic heat per unit mass of whee! material, in uitkg K).
In most cases where the thermal contact coefficients are the same for pinion and wheel, the thermo-elastic ‘actor
depends solely on the material characteristics.
5M
Bu
For martoneitic stocls the range of ay is 41 N(s K) to 62 Nils K),and py-oy is about 3,8 N/(mm2K), ¢0 that for
such steels the use of an average value By = 435 Ni(mm'm'”s'@-K) = 13,8 Nmm-s'~K) will not introduce 100,
great an error when the thetmo-alastic coefficient is not known. For gears made of representative steels, with
Ey = Eg = 206 000 Nimm®, 14 = 12 = 0,3 follows
Xu (a13)
y= 50,0 KN. 812.92 mm (aia)
AA Bevel gears
In bevel gears the successive contact areas will assume the shape of somewhat tapered bands. However, in most
cases a parallel Dand-shaped contact may be a good approximation, and together with both tangential velocities
coincident and directed perpendicular to the major axis, the original formula (A.1) may be used directly, especially
when correct data of the radii of curvature and the Hertzian contact band are known. For reasons of convenience
the formula may be rewritten with the octoid line of action approximated by a straight line and the factors expressed
in common quantities of bevel gears.
For the rewritten kinematic formulae the following assumptions®! were made [43]; see Figure A.2.
— pinion and wheel have a common apex and an arbitrary shaft angle:
2a d\+82 (CSE)
— all calculations concem the mean cones:
— the line of action is approximated by a straight line:
6) These assumptions produce a convenient approximation of the racii of curvature, but it does not invoNe that the bevel
gears are substituted by a virtual eylinarical gear pair.
32 6 180 2000 ~All rights reserved ISO/TR 13989-
the surface of action is taken as a plane surface.
Fay (eos /cos6;| = Re, tansy cosay
Figure A.2— Approximate line of action
‘The tlash temperature for bevel gears may be calculated either for the virtual oylindrical gear pair, or with the direct,
formula
n= tn Xn XX oT -wor)* (A.16)
where
Jhy is the mean coefficient of friction;
Xq_ is the thermo-elastic actor (see clause A 3)
Xy_ Is the approach factor (see clause 8);
Xq_is the geometry factor:
abs( i+ Ty - t+ 7; land; Ftands)
NXg =051- Xap (cots; + cotdg)"* i — 2 (At?)
(1+ 5)" A= Fy -tandy Mandy)"
where
Xap _ is the angle factor (see clause A2);
© 1502000 Airighisresewed 33 ISO/TR 13989-1:2000(E)
X
wet
4
fs the load sharing faotor (see clause 9)
is the transverse unit load (see 5.3)
is the pitch line velocity;
is the cone distance of mean cone.
SO 2000 - Allsights reserved ISO/TR 13989-
Annex B
(informative)
Optimal profile modification
If tooth profiles are modified, the modifications shall be designed and manufactured to answer a demanded load
sharing function, for instance like Figure 6. The optimal tip relief for pinion and wheel approximates.
Kx Kmp “Fi
(B.1)
b-e0804 cy 1)
“eft
where
Ka 18 tho application factor;
Kmp__ is the multiple path factor (see 5.3);
FIs the tangential force, in N;
is the tacewiath, in mm;
cq _ is the transverse pressure angle, in degrees;
¢; isthe mesh stiffness, in Niimm-ym).
‘The tip relieved height for cylindrical gears is neither to reach the area ot single pair tooth contact, nor to resutt in a
contact ratio << 1 when the gear is unloaded (.e. «calculated with fictive tip diameters equal to the diameter
where the relieved area starts).
If on a mating gear root relief applies, then the tip relief shall be replaced by an equivalent tip gear relief, being the
‘sum of the tip relief and a reduced root relief of the mating gear.
; 2
i
aCe -) (2)
7 2
Cage = Caa +E {# = (8.3)
for cylindrical gears
Hy