See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/296057735

Influence of Tolerances on Characteristic Manufacturing Deviations in Soft

Gear Machining

Conference Paper · October 2015

DOI: 10.13140/RG.2.1.2874.5362

CITATIONS READS

3 793

4 authors, including:

Markus Krömer
Gleason Switzerland AG
19 PUBLICATIONS   49 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SPARTApro - A novel manufacturing simulation for gear machining View project

All content following this page was uploaded by Markus Krömer on 17 March 2016.

The user has requested enhancement of the downloaded file.

Influence of Tolerances on Characteristic Manufacturing
Deviations in Soft Gear Machining

Analyzing soft gear machining with a manufacturing simulation

Prof. Dr.-Ing. Dr.-Ing. E.h. Dr. h.c. Dr. h.c. F. Klocke,

Prof. Dr.-Ing. C. Brecher, Dipl.-Wirt.-Ing. C. Löpenhaus,
Dipl.-Ing. M. Krömer, Laboratory for Machine Tools and Production
Engineering (WZL), RWTH Aachen University

Abstract
Designing gear soft machining processes is often based on characteristic values such as the
chip thickness. These characteristic values are calculated using analytic formulas based on
geometrical or empirical studies. These formulas have in common that they neglect the pos-
sibility of inaccuracy of the tool shape, of its tolerances as well as the clamping. However,
research has shown that these tolerances have an influence on the tool wear and the char-
acteristic values for the gear hobbing process. This leads to the conclusion that it is neces-
sary to consider tolerances of tool and clamping in the calculation of characteristic values for
designing the manufacturing process. These tolerances affect the machined gear quality and
in particular the machined root geometry, because while the flank is ground during most pro-
cess chains, the root geometry is mostly defined by soft machining.
This paper presents a method for simulating tolerances of the tool and the clamping in differ-
ent soft machining processes. The simulated gear geometry is then analyzed with the help of
a virtual measurement machine which calculates topographic surface plots as well as meas-
urements of the profile and the lead line of the gear.

The authors gratefully acknowledge financial support by the German Re-

search Foundation (DFG) [KL 500/134-1] for the achievement of the pro-
ject results. For the software-based achievement of simulative results the
authors gratefully acknowledge the financial support for the software de-
velopment by the WZL Gear Research Circle.
Introduction and Objective
Because cost efficiency has to increase and the time to market has to be shortened, simula-
tion tools become more and more important in each step of the product development pro-
cess. Simulation tools can avoid iterative steps based on trials and minimize the time for de-
velopment as well as costs [1]. At the Laboratory for Machine Tools and Production Engi-
neering (WZL) of the RWTH Aachen the WZL Gear Toolbox is developed. This software suit
combines manufacturing simulations for cylindrical [2] and bevel gears [3], tooth contact
analysis [4] as well as the calculation of gear geometries [5].
With the manufacturing simulation for continuous gear hobbing SPARTApro it is possible to
simulate the resulting macro and micro geometry of the work piece and the tooth flanks. Also
all undeformed chips of the cutting process are simulated and afterwards used to calculate
characteristic values such as the chip thickness, cutting length or specific cutting volume.
Within SPARTApro it is also possible to calculate the resulting cutting forces for different
work piece materials according to the cutting force model of Gutmann [7]. Additionally it is
possible to design cost optimized hobbing tools within an integrated variant calculation. Input
parameters of the simulation are the geometry and gearing data of the work piece and tool
as well as the process data like cutting speed and feed rates. These inputs are used to gen-
erate models of work piece and tool. The manufacturing simulation is based on a geometrical
penetration for calculating. Results of this simulation are shown in Fig. 1.

supported by the Macro geometry

WZL Gear Research Circle

Simulation Vorverzahnen Soft Gear Machining

Micro geometry Chips Key Values Cutting Forces Costs

Fc, Fr, Fa, M

hcu, lmax, V, … Rotation

Fig. 1: Process simulation for soft gear machining SPARTApro

Because of its high productivity gear hobbing is the most common machining process for
cylindrical gears, but due to restrictions in the working area, like interfering contours or lim-
ited, it is not possible to use gear hobbing for all cylindrical gears. Therefore, the scope of the
manufacturing simulation SPARTApro is extended to gear shaping [6] and gear profile mill-
ing. While a prototype of the gear milling and gear shaping simulation is integrated in the
SPARTApro the development is still in progress and not finished yet.
Fig. 2 shows the three machining processes. Because of the different kinematics each pro-
cess results in different characteristic manufacturing deviations. While in gear hobbing there
are deviations in direction of the gear width as well as the height of the tooth, the other two
processes create only deviations in one direction. In gear profile milling there are only feed
marks caused by the axial feed of the tool and in gear shaping only generated cuts in orien-
tation of the work piece face occur.

Gear hobbing Gear profile milling Gear shaping

vc
vf
feed mark

feed mark

generated cut generated cut

Fig. 2: Characteristic manufacturing deviations of different machining processes

These characteristic deviations are superposed by tolerances of tool, clamping and machine
kinematics. The total deviation is a combination of the described characteristic deviations as
well as the deviations caused by tolerances [8]. With the knowledge of the total deviation in
the process, the resulting gear quality can be determined. But because of the complex ma-
chining setup and kinematics the deviations can not be calculated with a simple analytical
approach.
Simulation of Soft Gear Machining Processes
Therefore it is comprehensible that especially for a manufacturing technology with such a
high importance and complexity as the soft gear machining, a manufacturing simulation is
needed [1]. Besides the knowledge of the work piece and chip geometry, a manufacturing
simulation also has to be capable of calculating characteristic values with a technological or
an economic background referring the specific machining process.
In the presented software it is possible to simulate different work piece and tool designs and
combine these with kinematics represented by a mathematical model. The machining pro-
cess is simulated in a geometrical penetration between tool and work piece over the com-
plete manufacturing process. The results like the undeformed chips or the calculated charac-
teristic values are afterwards displayed along the unrolled profile edge, to analyze tool wear
at any given position at the tool cutting edge. SPARTApro is upgraded to be capable of simu-
lating the surface topography of the work piece. Therefore, the resulting gap geometry is
compared with an ideal gear geometry, like shown in Fig. 3. This ideal geometry can be cal-
culated by other software products like the GearGenerator [5] developed at the WZL. The
GearGenerator uses the basic requirements of a gear tooth system [10] to calculate the ideal
gap geometry. The simulated surface can be measured and analyzed in any direction with a
virtual measurement machine based on the VDI/VDE 2612 [11] standard.

SPARTApro GearGenerator Comparing gaps Virtual measurement

Fig. 3: Approach for calculating the gear quality

The described method for analyzing deviations is integrated in the simulation program
SPARTApro. The output of the results is based on the design of measurement sheets of gear
measurement machines. While the gear measurement machines measure a specific number
of teeth of the gear, the software measures only one tooth on different heights. By analyzing
one tooth it is not possible to measure radial run-outs or pitch errors but it is possible to ana-
lyze modifications of the profile and the lead. In future versions it will also be possible to ana-
lyze a various number of teeth of the work piece.

Influence of Tolerances on the Work Piece Topography

Deviations in the machining process can occur as tolerances of the tool and work piece or
due to displacements in the clamping of tool and work piece [14]. The allowable tolerances of
the cutting tools are defined in different standards. For hobs the tolerances are defined in
DIN 8000 [15]. These are deviations of the pitch, the concentricity or the flank profile. In the
following these three tolerances will be investigated. The flank profile can differ from the ideal
profile because of tolerances in manufacturing of the hob or due to tool wear. There are sev-
eral types of pitch deviations, which can be consolidated to the axial pitch between two teeth
of the hob. On the tool clamping two different types of deviations have to be taken into ac-
count. These are tumbling and eccentricity deviations.
After identifying possible tolerances and deviations of tool and the clamping of the tool, the
maximum of each tolerance within the tolerance field has to be determined. The tolerances
of the tool are defined in DIN 3968 [16] and depend on the module and the tools quality cat-
egory. There are five grades for the quality but only the two best are relevant in praxis nowa-
days. In the further investigations a gear with a module of mn = 2.56 mm will be simulated.
Because the profile eccentricity and the deviations of the tip run-out have the highest maxi-
mum values, these two deviations will be further analyzed.
In Fig. 4 the surface topography as well as measurements of the work piece profile and lead
line are depicted. In the top left section the micro-geometrical surface of an ideal work piece
is depicted. Below the simulated surface of a work piece manufactured with an eccentricity of
Frk =25 µm is shown. It is visible that the surface topography of the tolerance affected work
piece differs from the ideal surface because of high deviations in direction of the tooth height.
At the right section of Fig. 4 the simulated profile and lead measurements of the two gears
are shown. The red, solid-dark line represents the ideal machined work piece while the light
graph represents the tolerance affected gear.
The measurement results of the profile as well as of the lead of the work piece simulated
without tolerances show a maximum deviation of Ff = 5 µm to the ideal flank. According to
DIN 3961 [12] this complies a gear quality of IT 5 which is close to the best quality which can
achieved in the hobbing process. As shown in Fig. 4 the surface topography of the second
work piece differs from the simulated surface without deviations due to tolerances. Analyzing
the measurements show nearly the same deviations in direction of the work piece width as in
the simulation without deviations due to tolerances but much higher deviations of the profile.
These profile deviations have a maximum value of Ff = 12 µm. A profile deviation of 12 µm
leads to a gear quality of IT 7 for a gear with a module of mn = 2.56 mm according to the
standard.
Topography without deviations due to tolerances Profile line
15
10 20
gear width

μm
5
0 5
mm
-5
μm
left flank root right flank

Tolerance affected topography Lead line

15
10 5
gear width

μm
5
0
5
-5 mm
μm
left flank root right flank

Fig. 4: Surface topography of an ideal and tolerance affected work piece

Influence of Tolerances on Characteristic Values

After calculating the work piece geometry and analyzing the resulting gear quality, the influ-
ence of tool deviations to the characteristic values in the hobbing process will be discussed.
In this paper the focus is on the maximum chip thickness and the chip volume. The chip
thickness is one of the most important characteristic values for process design in machining
with defined cutting edges [1], [2] and [13]. The chip volume is a combination of cutting
length and chip thickness and correlates with the cutting forces and the tool load.
In the top-left picture of Fig. 5 the surface topography of a simulated gear is shown. The
characteristic line of contact of hob and work piece is visible but the structure of the surface
is intermittent by small ‘buckles’. These buckles are a result of a tip concentricity at one hob
tooth of frK = 12 µm.
In the diagrams of the right the characteristic values chip thickness and chip volume are plot-
ted versus the unrolled profile edge. In each diagram three graphs are shown. The red, dark-
solid graph has the highest values for the hobbing tooth with a defined deviation of
frK = 12 µm. The second solid-light graph displays the following profile in contact after the first
one. The last graph represents all the rest of the tool profiles, which have all nearly the same
values. As it is shown the values of the tooth with a radial run-out are the highest for both
values, while the following tooth has the lowest. The chip thickness as well as the volume of
the profile with a run-out deviation is around 20 % higher than of all other profiles. As ex-
pected, the values of the following profile are in both cases the same amount lower as the
values of the deviation profile are higher than the other teeth of the hob. This observation
shows that there is a significant influence of tool deviations on characteristic values.
Topography Characteristic values
15 0.3

chip thickness [mm]

7.5 0.2 +20%
gear width

0
0.1
-7.5
0
-8 -6 -4 -2 0 2 4 6 8
-15
μm 1
left flank root right flank chip volume [mm³] 0.8 +20%
0.6
work piece: tool: process:
0.4
mn = 2.56 mm da0 = 80 mm climb cutting
z2 = 40 z0 = 1 fa = 2.4 mm 0.2
αn2 = 17.5° ni0 = 16 0
profile with deviations -8 -6 -4 -2 0 2 4 6 8
β2 = 23° tip run-out
following
b = 24 mm frp = 12 μm unrolled profile edge [mm]
other tool profiles

Fig. 5: Gear geometry due to a tip run-out of a tool profile

Comparison of Gear Manufacturing Processes

To compare the productivity and the resulting gear geometry of the different machining pro-
cesses, SPARTApro can simulate the manufacturing of the same work piece for different tool
and process concepts. For comparing the processes to each other it is necessary to define
an upper limit for the process parameters cutting speed and axial feed. These limits can be
determined base on empirical knowledge or by simulations, while calculating the maximum
chip thickness or cutting force for example. SPARTApro can afterwards calculate the number
of cuts as well as the working area of the tool and use this information to model the process
main time.
Another limit for the variation of the tool design and feed rate can be the characteristic manu-
facturing deviations. By simulating and analyzing the surface topographies of the work piece
after each process the maximum deviations can be determined. Within a variant calculation
an optimized tool and process design for each manufacturing process can be defined. Fig. 6
shows the simulated characteristic manufacturing deviations of the same work piece after
gear hobbing, milling and continuous gear shaping. As visible the characteristic deviations fit
to the theoretical deviations in Fig. 2. Beside the simulation of the ideal processes it is also
possible to simulate tolerances of the tool and clamping in each manufacturing process as
presented in in Fig. 5 exemplary for the gear hobbing process.

Gear hobbing Gear milling Gear shaping

Left flank Left flank Left flank

Right flank Right flank

Right flank

Fig. 6: Comparing surface topographies of different machining processes

Summary and Outlook

In this paper a manufacturing simulation for soft gear machining was developed and pre-
sented. For that purpose shortly the operating mode of a manufacturing simulation software
for gear hobbing was explained. In addition the method for a virtual measurement machine
was presented and applied on different simulated processes. This software is also capable of
simulating the soft machining processes gear profile for milling and continuous gear shaping.
It is possible to calculate characteristic values for the processes as well as process times and
cutting forces.
Furthermore the method for calculating surface topographies and measurements of the gear
was presented in in detail. Afterwards two examples of simulated hobbing processes were
shown. The examples show the effect of tolerances of the tool and the tool clamping on the
profile and lead line of the work piece. Fig. 7 shows the iterative process adjustment in case
of unknown deviations of the gear profile and gear lead. After machining and measuring of
the work piece the measured deviations can be analyzed with the help of the simulation
software. Therefore, various simulations with different deviations in the clamping and the tool
profile can be executed and afterwards compared to the real measurement sheets. This
comparison shows which tolerance causes the deviations of the work piece and which ad-
justments have to be done in the manufacturing process.

Machining Measuring Analyzing

Tool Clamping

Source: Klingelnberg Process Machine

 Machining the work piece  Measuring the work piece  Reason for deviations

 Calculating possible
adjustments

Process adjustments

Fig. 7: Iterative process adjustment in case of unknown profile and lead deviations

Literature
[1] Klocke F., Gorgels C., Schalaster R., Stuckenberg A.: An Innovative Way of Design-
ing Gear Hobbing Processes; International Conference on Gears 2010; Munich
[2] Brecher, C., Brumm, M., Krömer, M.: Design of Gear Hobbing Processes Using Simu-
lations and Empirical Data, In: 9th CIRP Conference on Intelligent Computation in
Manufacturing Engineering - CIRP ICME '14; 2014; Naples Italy
[3] Brecher, C., Klocke, F., Gorgels, C. Hardjosuwito, A.: Manufacturing Simulation of
Bevel Gear Cutting; Proceedings of 1st International Conference on Simulation and
Modeling Methodologies, Technologies and Applications- SIMULTECH 2011; Noord-
wijkerhout Netherlands
[4] Brecher, C., Brumm, M., Henser, J.: Calculation of the Tooth Root Load Carrying Ca-
pacity of Beveloid Gears; AGMA Fall Technical Meeting Papers 2013; Alexandria
USA
[5] Brecher, C., Brumm, M.; Hübner, F.; Henser, J.: Influence of the manufacturing meth-
od on the running behavior of beveloid gears. In: Wissenschaftliche Gesellschaft für
Produktionstechnik (WGP) Volume 7, Issue 2-3; 2012, Berlin
 [6] Klocke, F., Brecher, C., Krömer, M.: Untersuchung des Wälzstoßens mittels geomet-
rischer Durchdringungsrechnung; In: Aktuelle Entwicklungen beim Vorverzahnen;
12.-13.November 2014; Aachen
[7] Gutmann, P: Zerspankraftberechnung beim Wälzfräsen. Diss. RWTH Aachen 1988
[8] Borchert, W.: Auswirkung Wälzfräserfehlern Flankenform Stirnradverzahnung. Disser-
tation, RWTH Aachen, 1972
[9] Klocke, F.: An Innovative Way of Designing Gear Hobbing Processes. In: 1st CIRP
Conference on Surface Integrity, Procedia Engineering, Bremen, 2011
[10] Litvin, F. L.; Fuentes, A.: Gear Geometry and Applied Theory, 2. Auflage, Cambridge
University Press, Cambridge, 2004
[11] VDI/VDE 2612: Profile and helix checking of involute cylindrical gears, 2000
[12] DIN 3961: Tolerances for Cylindrical Gear Teeth – Tolerances for Deviations of Indi-
vidual Parameters, 1978
[13] Weck, M.; Winter, W.; Klocke, F.; Winkel, O.: Analysis of Gear Hobbing Processes by
Manufacturing Simulation. Wissenschaftliche Gesellschaft für Produktionstechnik,
2003
[14] Gravel, G.: Simulation of Deviations in Hobbing and Generation Grinding, VDI Inter-
national Conference on Gears, Munich, 2013
[15] DIN 8000: Design dimensions and tolerances of hobs for involute spur gears; defini-
tions, 1962
[16] DIN 3968: Tolerances for single-start hobs for involute spur gears, 1960

View publication stats