Automotive transmission testing

Citation for published version (APA):Scheepers, B. T. M. (2002). Automotive transmission testing: a new concept. (DCT rapporten; Vol. 2002.018).Technische Universiteit Eindhoven.

Document status and date:Published: 01/01/2002

Document Version:Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can beimportant differences between the submitted version and the official published version of record. Peopleinterested in the research are advised to contact the author for the final version of the publication, or visit theDOI to the publisher's website.• The final author version and the galley proof are versions of the publication after peer review.• The final published version features the final layout of the paper including the volume, issue and pagenumbers.Link to publication

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, pleasefollow below link for the End User Agreement:www.tue.nl/taverne

Take down policyIf you believe that this document breaches copyright please contact us at:openaccess@tue.nlproviding details and we will investigate your claim.

Download date: 03. Apr. 2021

https://research.tue.nl/en/publications/automotive-transmission-testing(0ee77b80-ab64-4dfb-9b65-69e918f05b84).html

Automotive Transmission Testing, a new concept

Bart Scheepers 477079 DCT 2002 - 18 January 2002

Eindhoven Technical University Dynamics and Control Technology Automotive Engineering

Coaches: Prof. Ir. N. J. J. Liebrand Dr. Ir. W.J.A.E.M. Post

Summary

The increased demands for comfort and reliability in modern vehicles requires well-developed solutions. The characteristics of the vehicle powertrain are a major criteria in this regard. Testing is an important part of the transmission development. Transmission Test programs can be classified into three main areas:

component testing: static and dynamic tests

testing the complete transmission on a test rig: functional tests and road simu- lation tests

a vehicle testing: track, road and user test

The basic test rig setups for testing transmission components are based on the energy flow and energy loop (with or without energy modulator) process. The energy loop process is used most times because of the energy recovery. When testing complete transmissions only the engine flow or loop (with modulator) process is used while drive and brake can be controlled the best. Test rigs are driven by a combustion engine or electric motor. The combustion engine is an exact copy of the reality (the transmission is loaded exact the same way as in the vehicle) but is expensive to use. An electric motor is much easier to use but is not able to simulate the reality, mainly caused by the large rotor inertia.

An attempt is made to simulate the dynamic engine torque behaviour (caused by a non constant combustion and reciprocating parts) by using multiple electric motors in series. One large low dynamic motor generates a constant torque and two small high dynamic ones generate a smaller positive or negative torque. By connecting the motors, the added torque can be controlled in such way it corresponds to the dynamic torque a combustion engine generates. When accelerating, the inertia of all motors have to be accelerated. While the inertia-torque ratio of the e-motors is larger than the engine's, the e-motors accelerate slower. The test rig still doesn't always react the way the combustion engine does.

While transmission development has to become cheaper, better and quicker, more and more tests will be done on a test rig in stead of testing transmissions in a vehicle. Substituting the vehicle test into dynamic rig tests is called "Road to Rig". The transmission load during a road / track test has to be simulated on a test rig.

While current available test rigs can't simulate vehicle use exactly (mainly caused by the high electric motor inertia-torque ratio), a new test rig concept will be developed. With an hydromotor or an hydromotor in combination with an electric motor it might be possible to simulate a combustion engine without having the disadvantages an engine has.

Samenvat t ing

De toegenomen eisen voor comfort en betrouwbaarheid van moderne voertuigen vereist ver- gaande ontwikkelingen. De karakteristieken van de transmissie zijn wat dit betreft een belan- grijk criterium. Testen is een belangrijk onderdeel van de transmissie ontwikkeling. Testpro- gramma's kunnen verdeeld worden in drie hoofdcategorien:

component testen: statisch en dynamisch testen

testen van de complete transmissie op een testopstelling: functioneel testen en weg simulatie testen

0 voertuig testen: circuit, weg and gebruikers test

De drie basis testopstellingen voor het testen van transmissie onderdelen zijn gebaseerd op het energie flow en energie loop (met of zonder energie omzetter) principe. Het energie loop principe wordt het meest gebruikt omdat energie terug gewonnen wordt. Bij het testen van complete transmissies wordt alleen van het energie flow en energie loop (met omzetter) principe gebruik gemaakt omdat de motor en rem hierbij het best geregeld kunnen worden. Testopstellingen worden aangedreven door een elektro- of verbrandingsmotor. Een verbrand- ingsmotor is een exacte kopie van de werklijkheid (de transmissie wordt op exact dezelfde manier belast als in het voertuig) maar is duur en lastig in gebruik. Een elektromotor is veel makkelijker in gebruik maar is geen kopie van de werkelijkheid (vooral veroorzaakt door de grote rotor traagheid).

Er is een poging gedaan om het dynamisch motor koppelverloop (niet constante verbrand- ing en op en neer gaande onderdelen) te simuleren door meerdere elektromotoren in serie aan elkaar te koppelen. Een laag dynamische motor genereert een constant koppel en twee kleine hoog dynamische motoren genereren een kleiner positief of negatief koppel. Het totale kop- pel kan zo geregeld kan worden dat het overeenkomt met het dynamisch koppel dat door een verbrandingsmotor gegenereerd wordt. Tijdens versnellen moeten de traagheden van alle motoren versneld worden. Omdat de traagheid - koppel verhouding van de elektromotoren groter is dan die van de verbrandingsmotor versnellen de elektromotoren langzamer. De testopstelling reageert dus nog altijd niet op dezelfde manier als een verbrandingsmotor.

Omdat transmissie ontwikkeling goedkoper, beter en sneller moet worden steeds meer tests op testopstellingen uitgevoerd in plaats van in een voertuig. Het vervangen van voertuigtesten door dynamische tests op een testopstelling wordt "Road to Rig" genoemd. De transmissie belasting gedurende een weg / circuit test wordt gesimuleerd op een testopstelling.

Omdat de huidige testopstellingen niet in staat zijn het gebruik van een transmissie in een voertuig te simuleren wordt een nieuw concept ontwikkeld. Met een hydromotor of een hydro- motor in combinatie met een elektromotor is het misschien mogelijk een verbrandingsmotor te simuleren zonder de nadelen van een verbrandingsmotor en elektromotor.

Contents

Summary 1

Samenvat t ing 2

1 Introduction 5

2 Automotive transmission tests 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 Component testing 6

. . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Testing complete transmission 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.1 Functional testing 7

. . . . . . . . . . . . . . . . . . . . . . . . . . 2.2.2 Road simulation testing 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3 Vehicletesting 10

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4 Conclusion 11

3 Test rigs 12 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1 Component test rig 12

. . . . . . . . . . . . . . . . . . . . 3.2 Test rig for testing complete transmissions 14 3.3 Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.1 Combustion engine 15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.2 Electric motor 16

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3.3 Conclusion 18 . . . . . . . . . . . . . . . . . . . . . 3.4 Capability of an electric powered test rig 19

4 Alternative test rig 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1 Multiple electric motors 21

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Working principle 21 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.2 Motorproperties 22

. . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.3 Capabilities of the rig 22 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.4 Conclusion 23

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2 Hydromotor 23

5 Driving simulation 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1 Transmission loads 24

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2 From "Road to Rig" 25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3 Dynamic vehicle simulations 26

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4 Conclusion 26

6 Specifications new rig 2 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.1 Development criteria 27

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 Test rig specifications 28

Bibliography 30

List of Figures 3 1

Chapter

Introduction

The increased demands for comfort and reliability in modern vehicles requires well-developed solutions. The characteristics of the vehicle powertrain are a major criteria in this regard. It is important that the entire powertrain has been designed optimally with respect to dura- bility, dynamics, vibrations and acoustics under the most extreme conditions. Components such as the clutch, the torque converter and the differential gear must be able to withstand high dynamic loads over the entire life cycle.

Testing transmissions on a test rig shortens development time and avoids expensive road test runs. These test rigs have to be able to simulate a vehicle and driver exactly. When transmissions are tested on a test rig, most times electric motors are used for driving and braking. These are easy to use and to control. However, high dynamic torques like in a vehi- cle can't be simulated. These torques might increase wear so have to be simulated on a test rig.

During the last part of my study Mechanical Engineering, section Automotive Engineer- ing, at the Eindhoven Technical University i'm accomplishing my graduation project. Goal of the project is developing a transmission test rig on which a transmission can be loaded in exact the same way a vehicle does. This report deals with the first part of the project, the literature search.

The first part of the report describes in what ways transmissions are tested. Chapter 3 deals with the different types of test rigs and how the rig can be driven. While common used drives all have some disadvantages an alternative rig with multiple electric motors in series is developed by the Rostock University. Chapter 4 gives a critical examination of this rig. In the next chapter will be discussed how driving simulations on a test rig can replace vehicle tests; the "road to rig" concept. Finally the specifications of the new test rig are defined. How the specifications are translated in a new transmission rig can be read in the the final report.

Chapter 2

Automotive transmission tests

The test programs carried out during vehicle transmission development can be classified into three main areas:

component testing

testing the complete transmission on a test rig

vehicle testing

Testing on a test rig can be divided in functional testing and road simulation testing. Functional testing is not only carried out during development but also when a transmission is overhauled. In the next sections the three types of transmission tests will be discussed.

2.1 Component testing

Component testing is carried out at the early stage of the transmission development. Individ- ual subsystems (with their own function) or parts are tested. These tests are kept as simple as possible, so they can give principal information about the components. Influences from the environment like when the component is a part of a large system are excluded. A distinction is made between static and dynamic tests (figure 2.1). Static refers to the simple part test, dynamic to the individual subsystem test.

The use of component (subsystem) testing is shown by the next example. Pushbelt CVT's consist of a number of individual subsystems: the variator (belt and pulleys), an hydraulic system and an electronic control system (computer). When the belt breaks during a test (complete transmission testing), there can be several reasons. For example the hydraulic system could have failed; no belt lubrication or too much or too less pressure on the pulleys. It's difficult to determine what caused the failure exactly because the subsystems interact. If the belt breaks when the variator is tested separately from the other subsystems of the CVT, the failure must be caused by the variator subsystem. Some specific component part tests only give an approximation of the failure behaviour of the components for a typical type of fault. Fox example notch impact test for gearwheel fracture.

Extent of component testing

Static tests J E.g. Strength, deformation, elastic characteristic values

Density in the case of sintered materials

Chemical composition

Corrosion behaviour

Viscosity / temperature behaviour

Pressure viscosity

I Dynamic tests I Load profile or stress profile 1

I I Stress known.

Component calculable

Component calculable with

E.g. Gearwheels, shafts, bearings, bolts

3rd order

Clutches, ffeewheels, synchronisers, piston rings and similar wearing parts

All types of seals, ffiction linings, V-belts, pneumatic springs

Stress unclear. Component selection by

experience

Figure 2.1: Extent of component testing [I]

j

Automotive transmission suppliers (e.g. VDT) only use component tests. With these tests they can determine if the lifetime is sufficient. While component testing only gives information about one subsystem or part, testing complete transmissions as described in the next section provides the cooperation between the separate subsystems as a whole system.

2.2 Testing complete transmission

Testing complete transmissions on a test rig can be divided in functional testing and road simulation testing. Functional testing is not only carried out during development but also when a transmissions is overhauled.

2.2.1 Functional testing

Functional testing is the next step in transmission development. During these tests the engineer finds out if the individual subsystems and parts cooperate in the right way. The power conducting parts are initially tested with low stress. Functional testing consists of the next six items:

Temperature behaviour: Friction in the transmission causes heat, the temperature of the oil rises. Too high or too low temperatures in the transmission can result in damage due to bad lubrication behaviour (too high or too low viscosity) of the oil.

Power loss: Friction causes loss of power. This may not be too large (decreasing efficiency). Efficiency can be calculated by the next formula:

T o u t = Torque output shaft [Nml W O U ~ = Angular velocity output shaft [rad/s] %I = Torque input shaft Prn1 W i n = Angular velocity input shaft b u s ]

Hydraulic system: The hydraulic system is a very important part of the transmission. The next items are checked during the test: pressures (e.g. in CVT's pressure on primary and secondary pulley), flows and leaks.

Vibration tests: Due to imbalance there can appear vibrations in the transmission. This is not desirable because of irritating noise and wear (bearings and seals).

Ratio coverage: Only for CVT's. The variator is designed for a specific ratio coverage. Too large or too small ratios are caused by too large or too small pulleys. This can result in a damaged belt.

Shiftability: Transmissions have to shift smoothly and at the right moment. During this test the kick-down function is tested also.

Figure 2.2: P C screen functional transmission test 1101

Most times there is no specified cycle that has to be passed though. The functional tests have a general form: The transmission is connected to the rig. If all sensors are installed the test can start. The motor starts and if all axles turn smoothly on low speed of rotation, the motor will accelerate. If still everything works as expected the output axle can be loaded by using a brake (for more information about test rigs, see chapter 3). During the tests several variables are measured and saved so the engineer is able to determine if the transmission satisfies the prescribed demands.

Figure 2.2 shows the display used by a transmission overhaul company during a CVT functional transmission test. The X-axle in the large graph represents the speed of the motor (primary axle CVT) the Y-axle the speed of the secondary axle. The two straight lines represent the maximum and minimum prescribed ratio. During the test the speed of the primary and secondary axle is measured continuously and plot in the graph. If the measured ratio is larger than the prescribed maximum or smaller than the prescribed minimum ratio (easy to see in the in the graph) the transmission doesn't satisfies the demands. Further the plot gives information about the shifting characteristics (e.g kick down). The displays at the right side give information about oil pressure, flow etc .

2.2.2 Road simulation testing

During road simulation tests the complete transmission is mounted on a test bed comparable with the functional transmission tests. The main difference between functional and road simulation testing is the way of testing. With functional test the cooperation of the subsystems is worked out. During road simulation tests the emphasis is on the long term performance of the transmission. Functional tests normally don't take longer than an hour. During road simulation tests the engineer tries to create a situation similar to the use in a vehicle. Specific load time functions, comparable with vehicle use are defined (see chapter 5) . By simulating peak load situations (high torque and power) only, the test distance can be reduced to 40.000 km. After six weeks the engineer is able to predict the durability of the transmission.

Bench testing complete transmission

I Road simulation test Temperature behaviour load with sfress time functions

endurance tests -I G- I fatigue tests

Figure 2.3: General survey bench testing complete transmission [l]

This way of testing requires a higher performance of the test rig. The used motor has to act in the same way as the engine that will be used in the vehicle (same engine map, inertia etc.). Braking by using the motor must be possible; the energy moves in reverse way though the transmission. Road simulation test gives the possibility to do lifetime tests even before a test vehicle for which the transmission is developed is available. Engineering the different parts of the vehicle can be done at the same time so the development time for the whole vehicle decreases. A general survey of testing the complete transmissions on a test rig is given in figure 2.3.

2.3 Vehicle testing

Testing the transmission on a test bench only is not sufficient. That's why the transmission is mounted in a real vehicle so real life situations can be investigated. Vehicle testing involves testing the system as the last part of the testing and development phase. During this phase the reliability of the transmission is determined on various test routes and under different conditions. Finally the transmission influence on the vehicle comfort is investigated. Vehicle testing can be divided in the next three items:

Track test: At a track (most times closed for public) a large amount of kilometers is driven with test vehicles. Because of driving the same lap over and over again, the work is very boring for the driver. The test vehicles can be new ones, or old models equipped with new parts that have to be tested, e.g. suspension or transmission. Old models are used if the new model is not available yet. In Lommel (Belgium) such proving ground (Ford) is situated (Figure 2.4). The routes are so demanding as to produce faults after short running times. The test routes are so designed they include a mixture of motorways, hilly stretches and mountain roads. Especially the steep inclines are demanding for the transmission.

Figure 2.4: Ford Proving Ground Lommel. Left picture: overview of the ground. Right picture: the steep inclines [9]

Road test: One of the last parts of transmission testing are the road tests. The vehicle is used in different climate conditions to be sure it runs during all possible circumstances. Examples of test criteria:

The transmission oil temperature has to stay in a prescribed window to be assured of sufficient lubrication. During road tests in extreme hot or cold situations the tempera- ture behaviour of the oil is worked out.

In circumstances with high humidity or driving in rain the composition of the oil is checked. Too much water decreases the oil quality.

Cold starts at low temperature is bad for the bearings due to bad lubrication. After several cold starts the bearing wear is tested.

User test: In the context of testing for motor vehicles, the trend is now increasingly towards "fleet trials" in co-operation with specific user groups. These fleet trials provide an

extraordinary realistic profile in respect of various types of stress. This fact then implies that wear arising in the course of these fleet trials will also occur in actual day-to-day use. Examples of user groups are:

0 Police: many times maximum torque / power

0 Taxi driver: many kilometers in a short time

0 Hauliers: many start-stops

It is necessary to calculate beforehand how many test units permit a significant result for a given population. It should be mentioned at this point that tests have been carried out to show that when the number of test pieces is increased, their actual test duration of the test pieces can be reduced. This is naturally highly desirable with a view to saving time and money when developing new products.

2.4 Conclusion

Before a transmission is ready to use in a vehicle, a lot of work has to be done. Parallel at the designing the testing takes place. At first the components and subsystems, then the complete transmission on a test rig and finally the tests in a vehicle. It's important that the basic design is well thought-out so in the last phase only small adaptations have to be done.

Chapter 3

Test rigs

In this chapter different test rigs for testing rotating transmission parts are discussed. A division is made between component test rigs and rigs for testing complete transmissions. The last part of the chapter deals about the power plants used on current test rigs.

3.1 Component test rig

The component test rig serves to test individual parts or small subsystems so tolerable stresses and types of failure can be investigated. Component test rigs are most times specially de- signed. That's why they can be used for testing parts of similar design only.

Engine Component Brake A

Figure 3.1: Three basic test rig setups. Arrows represent the powerflow A) Energy flow process B) Energy loop process with modulator C) Energy loop process without modulator (torque test rig)

The design of the component test rig can be very easy. For testing the function of gear wheels, a motor and a set of gear wheels are enough. For investigating the efficiency of a CVT variator (pulleys and belt) a more intricate rig is necessary. The components have to be tested under realistic stress. Figure 3.1 shows the three basic setups for such tests. The transmission can be driven by an electric motor, hydro motor or internal combustion engine. The driving resistance and inertia is provided by a brake. This brake can be e.g. a brake-disk, water brake, hydro pump or electric generator.

The major disadvantage of the energy flow method (fig. 3.1 A) is the energy loss. This disadvantage can be circumvented by using an energy loop process. In this situation the brake converts the rotation energy into another type of energy like electric energy. This energy flows trough a modulator so the recovered energy can be used by the motor. Figure 3.1 B shows the energy flows in an energy loop process. Compared with figure A the power added to the test rig is much smaller. Only the losses in the tested component have to be compensated (losses in motor, brake and modulator are neglected in the figure, but are not zero in prac- tice). In the most common setup two electric motors are used. One to generate power, the other to dissipate the power. The electric energy generated by the brake is converted in a frequency modulator so it can be used by the motor. If this setup is used, the brake can act as a motor and the motor as a brake so braking on the engine can be simulated (reversing torque).

The use of energy loop rigs described before is still very expensive. This is due to the large motor and brake necessary for full test performance. Figure 3.1 C shows a test rig with energy flow, without brake and modulator. These are called torque test rigs. The necessary power is split up into torque and rotational speed. The torque is generated in the test rig itself. The rotational speed is caused by the motor. The output of the test gears is linked to the input of the crossover linkage and shafts.

Testgearwheel pair

Figure 3.2: Simple torque test rig for testing spur gears. [I]

Figure 3.2 is such a test rig for testing spur gears. For producing torque, the circuit is split up at a particular point, the bracing flange. A torque is caused by generating a angular displacement between the parts. The parts will be fixed when the required torque is reached. The transmission parts are now statically stressed. The electric motor only has to provide the power loss (10-20% of the test power) and is thus substantially smaller than in the energy

loop method with modulator. Figure 3.3 shows a torque test rig for testing high power CVT chains. The rig is equipped with two CVT's. The ratio's differ a little bit so torque is gener- ated. This causes microslip between the pulley and the belt. The larger the ratio difference, the higher the torque.

The disadvantage of torque test rigs is that they are, compared to the energy loop rigs with modulator, not that easy to manage (not flexible). Most times they are specially designed for testing one specific product like gears or pulleys. That's why these rigs are most times used by transmission part suppliers. Further it is not easy to control the torque magnitude and direction (positive or negative) during the test.

Figure 3.3: Torque test rig for testing high power CVT chains. Photo: Gear Chain Industrial, Nuenen, The Netherlands.

3.2 Test rig for testing complete transmissions

Test rigs used for testing complete transmissions are most times based on the energy loop principle (figure 3.1 B). This type of rig can be controlled relatively easy so simulating the vehicle can be done realistically. Energy recovery is possible (not possible with energy flow process) and the input and output side can be controlled independently (not possible with torque rig). One motor controls the speed of rotation, the other one the torque. Because there are different transmission lay-outs (for front, rear and four wheel drive cars and trans- verse and longitudinal placed engines) there are different test rig lay-outs for testing these transmissions (see figure 3.4).

When simulating road usage, the motor has to simulate the torque enforced at the trans- mission input shaft. The brake has to simulate the vehicle inertia and the resistance caused by air, rolling and hill climbing. The spring and damper behaviour of tires and drive shafts has to be simulated by the brake as well. Sometimes the rig is made cheaper by using a

Figure 3.4: Different lay-outs for testing complete transmissions.

smaller brake. A flywheel is mounted between the transmission and the brake. This wheel simulates the vehicle inertia (mass). The brake only has to simulate air, rolling and hill climbing resistance. A fly wheel is not usable when testing at constant speed and full power.

3.3 Power

Test rigs for testing rotating components (like gear wheels or variators) or complete transmis- sions (functional or road simulating) need one or more motors for power supply. There are several types of power suppliers suitable, e.g intern combustion engine, electric motor, hydro motor or gas turbine. Only the first two are used for test rigs. The aim of using test rigs is simulating the reality (using the transmission in a vehicle) as good as possible. The extend of reality, advantages and disadvantages of using an intern combustion engine or electric motor is discussed in the next two sections.

3.3.1 Combustion engine

At first sight using a combustion engine seems to be the easiest way of powering a test rig. However, there are some disadvantages.

+ Exact copy of reality: If the engine used on the test rig is the same as mounted in the vehicle, the simulation is an exact copy of the reality. The test engineer can be sure that testing the transmission on this rig gives the same result as testing the transmission in the vehicle. This test is very reliable.

- Difficult to control: A combustion engine can only be controlled by operating the throttle. This is not an easy way to control the desired torque and speed of rotation.

- Expensive: For every transmission tested on the rig, the corresponding engine has to be mounted. This requires a lot of effort because an engine needs cooling, exhaust pipes, fuel supply and connection to the operating room (Figure 3.5) .

- No energy recovery: The power produced by the engine can only be converted in heat, energy with a low value. This conversion can be done by a water brake, brake disk or electric brake.

- Engine not always available: During the development of a new vehicle, the devel- opment of the several components takes place at the same time. So it can happen that the transmission has to be tested while the engine is not available yet.

Figure 3.5: Test rig equipped with an internal combustion engine. [8]

When testing a transmission with a combustion engine, the most reliable results are achieved when using exact the same engine as will be mounted in the vehicle. While this takes a lot of effort, most times only one engine is used for testing different transmissions.

3.3.2 Electric motor

Nowadays most test rigs are equipped with an electric motor. The advantages and disadvan- tages of an electric motor used as power supplier are discussed below.

+ Simulating all combustion engines possible: If the capacity is large enough, the electric motor is able to simulate the engine map by using software.

+ Easy to control: By using frequency converters and software it is easy to control the electric motor concerning speed of rotation and torque.

+ Energy recovery is easy: Especially with electric motors it is easy to recover the power. Many times the motor is exact the same machine as the electric brake. The motor converts electrical energy in rotational energy, the brake does the opposite. By using frequency converters the electric power generated by the brake can be delivered to the motor. The electricity net only has to compensate the power losses.

f Reliable: An electric motor is very reliable. Only few maintenance (like changing bearings) is necessary.

- Not an exact copy of the reality: This is the only disadvantage of using an electric motor. There is one big difference between the combustion engine (reality) and electric motor used for transmission testing: the inertia - torque ratio. Using an electric motor results in two problems:

e Acceleration - decelerating: The electric motor inertia is much larger than the inertia of comparable combustion engine. See for example figure 3.6; the engine map of an 81 [kW] electric motor and an 1.6 [l] otto internal combustion engine. The e-motor is used on a test rig to test the transmission mounted on the engine. Especially this motor is used because at every speed of rotation the e-motor can generate more torque than the engine. At constant speed the e-motor has always enough power to simulate the engine. When accelerating a problem occurs. The electric motor has an inertia of 0.5 [kgm2], the engine only 0.1 [kgm2]. When the test engineer wants to simulate an acceleration of the engine (e.g. kick down), the e-motor has not enough torque to accelerate itself the way the engine can.

300

- 250 f t 200 E-motor 0,

$ 150 - 1.6 L Engine A00

50

0 2000 4000 6000

Speed of rotation [rpm]

Figure 3.6: Engine map electric AC motor and 1.6 1 combustion engine.

An example: the engine runs at 3000 [rpm], the operator shifts down and the engine accelerates as much as possible. At that time there is 150 [Nm] available (imagine all power is used for accelerating the engine). With an inertia of 0.1 [kgm2] this results in an acceleration of 1500 [rad/s2] because

T = Torque [Nml J = Inertia [kgm2] w = Angular acceleration [rad/s2]

If the e-motor has to accelerate the same way the combustion engine does, a torque of 0.5 - 1500 = 750 [Nm] is necessary. At 3000 [rpm] the motor generates 270 [Nm]; 480

[Nm] short. For generating so much torque, a larger motor can be chosen. However, if the generated torque increases, the inertia increases even more. An electric motor can never accelerate and decelerate the same way a combustion engine can.

Torque behaviour: A combustion engine doesn't generate a constant torque. This is caused by the non constant combustion in the cylinders and the movement of the recip- rocating parts like the pistons. The torque peaks can cause damage to the transmission parts. Figure 3.7 shows the torque of a small 3 cylinder diesel engine working at 2100 [rpm] and 50% full load. The dotted line indicates the average torq.de. The torque is measured between the engine and the flywheel. The frequency depends on the number of cylinders; more cylinders give an higher frequency. The amplitude depends on an number of parameters. At first the principle of the engine; Otto or Diesel. The diesel engine is based on self-ignition, resulting in higher pressures than the otto machine. Second parameter is the amount of fuel burned per degree crank angle (depends from speed of rotation, air temperature and density, fuel composition etc. [7]. A high amount results in high pressure (high torque). At last the number of cylinders and the way they are ordered is an important parameter. A large number of cylinders number results in an overlap in torque peaks. The peaks wipe out.

Crankangle []

Figure 3.7: Torque generated by a 3 cylinder diesel engine. n=2100 [rpm]. 50% full load.

For transmission testing, the torque offered by the flywheel to the transmission is im- portant. This torque pattern doesn't show such high torque peaks as figure 3.7, but these patterns are still too high and with a too high frequency to be simulated by an electric motor. This is caused by a the large rotor inertia.

3.3.3 Conclusion

The use of electric motors or combustion engines both have advantages and disadvantages concerning transmission testing. Nowadays almost all transmission test rigs are equipped with electric motors while combustion engines have much more disadvantages. For functional testing this power plant is perfect, for simulation the usage in a vehicle it is not ideal. The high motor inertia compared to a combustion engine causes the problem. Generating torque peaks and high accelerations and decelerations like a combustion engine is not possible with

an electric motor. In the next section is described to what extent simulating a combustion engine with an electric motor is possible. In chapter 4 a new transmission test rig concept is described.

3.4 Capability of an electric powered test rig

During a simple test the capabilities of a test rig, situated at the Eindhoven Technical Uni- versity, are investigated. This rig consist of two identical electric motors. One for driving, one for braking the transmission. A gear set (r=4) is placed between the transmission and the brake motor to invert the final reduction of the transmission. Figure 3.6 shows the e-motor map and the map of the combustion engine that has to be simulated.

I 180 5 181 181 5 182 1825 183

Time I*]

Figure 3.8: dynamic torque, frequency = 1 [Hz]

The target of the test is to determine to what frequency and amplitude the rig is able to generate a sinus. The maximum motor acceleration is not investigated. During the first test the prescribed sinus has the next specifications: nominal torque = 20 [Nm], frequency = 1 [Hz] and amplitude = 5 [Nm]. This is visualized in 3.8 by the dotted line. The solid line represents the torque the drive motor really generates.

Figure 3.9: dynamic torque, frequency = 2 [Hz]

The generated torque shows a sinus shape with a phase shift, an offset in the nominal torque and a larger amplitude than the prescribed. At this low frequency (1 [Hz]) the rig is still able to generate a sinus wave (not exactly the prescribed one, but a sinus can be recognized). During the second test the frequency is doubled: 2 [Hz]. Figure 3.9 shows the prescribed (dotted) and generated (solid) torque. It can be seen that the rig is not able to follow the prescribed sinus at all.

Because of the large rotor inertia the electric motor is not able to generate a sinus at frequencies a combustion engine does. Though, creating a sinus at 2 [Hz] should not be a problem with this e-motor. Probably the control system is not able to control the motor that fast. The generated torque is measured continuously. By comparing the measured value (dotted line) with the desired one (solid), the the error is known. The control system adapts the voltage to the motor in order to reach the desired torque. Unfortunately the system is to slow at high frequencies.

Chapter 4

Alternative test rig

In the previous chapter the advantages and disadvantages of the electric motor and combustion engine are described. In this chapter an alternative drive is presented.

4.1 Multiple electric motors

At the Rostock University a new test rig for testing transmissions or components is devel- oped. The rig is equipped with electric motors and should be able to simulate the torque a combustion engines generates.

4.1.1 Working principle

Figure 4.1 shows the working principle of the rig.

Induction motor

Servo motor

Brakemotor

Transmission / component

Figure 4.1: Test rig equipped with multiple electric motors. [5]

Instead of one motor (normal configuration) this rig is equipped with multiple electric motors in series. The torque that has to be generated is divided over the motors. The main (induction) motor is low dynamic due to high inertia and generates a constant high torque. The small servo motor is high dynamic and can only generate low (compared with the main motor) torques. The servo motor is able to change quickly the generated torque from positive into negative. The arrows in the figure represent the torque generated by the motors. The torques generated by both motors are added and represent the torque (power) supply. The

main motor and servo-motor are connected with a torsion spring. This spring gives the servo- motor an additional degree of freedom. It enforces the torque but not the rotation. The servo-motor is connected with the transmission (rigid) and executes the prescribed rotation. The transmission is loaded by an electric brake, also connected by means of a spring.

This rig is developed for generating high torques at high frequencies. This is made possible by using the springs. They don't only give the servo-motor and transmission an extra degree of freedom, but also let the servo-motor and transmission resonate at certain test frequencies. The springs make it possible to increase the test frequency.

4.1.2 Motor properties

The rig as described in the previous section is built in reality. What parts are used can be read in this section. The main motor is an induction motor with a nominal power of 185 [kW] and a torque of 610 [Nm] at 2900 [rprn]. Maximum speed of rotation is 4500 [rpm], rotor inertia equals 1.93 [kgm2].

The servo-motor inertia has to be small. That's why two small ones (act as one motor with long rotor) are used in stead of one large one. The motors are connected rigidly and controlled similar, so they may be interpreted as one unit. Each motor is able to generate a torque of about 100 [Nm] at 4500 [rprn], the inertia equals 0.029 [kgm2]. The servo-motors are specially designed for this rig, that's why these motors have such favorable inertia-torque ratio.

Finally the brake. This is an electric induction motor with 215 [kW] nominal power, 710 [Nm] at 2900 [rprn]. Maximum speed of rotation is 4500 [rprn].

4.1.3 Capabilities of the rig

As described before the motor section is able to generate a static torque of 610 & 200 [Nm]. By the sophisticated choice of the springs, at some frequencies the torque at the input shaft can reach a peak torque of 900 [Nm].

Figure 4.2: Acceleration of a 4 cylinder 4 stroke engine. [5]

Figure 4.2 shows the acceleration of a 4 cylinder 4 stroke engine. The speed of rotation plotted in this graph is measured between the engine and the transmission. The transmission has an inertia of 0.03 [kgm2]. The developed rig is still not able to generate the same torque the engine does. This is due to the frequency modulator. This part is not able to cope with such high currents necessary for generating this amount of torque.

This rig is designed for transient testing. However, the circumstances for investigating shifting behaviour are not realistic. When testing transient the accelerating and decelerating

(slowly) of the transmission occurs simultaneously with the motors. Simulating the torque peaks (speed of rotation increase and decrease constantly) generated by a combustion engine is possible while the generated motor torque is detached from its inertia. The main motor runs at constant speed or slowly increases or decreases like the average speed of rotation of figure 4.2.

Simulating road use is not possible. In that case the motors (induction and servo) have to be able to change their own speed of rotation quickly (i.g in case of kick down). A combustion engine can accelerate at 1500 [rad/s2] (subsection 3.3.2). The total inertia of all electric motors equals 1.93 + 0.029 + 0.029 = 1.988 [kgm2]. The maximum torque available for acceleration is 610 + 100 f 100 = 810 [Nm]. This results in maximum acceleration of 810 / 1.988 = 407 [rad/s2]. This is factor 3.7 less than the combustion engine.

4.1.4 Conclusion

With the set-up describes above, a move in the right direction concerning simulating engines with electric motors is made. Simulating the torque peaks caused by the explosions in the cylinders and the movement of the reciprocating parts is possible. The large inertia of the motors is still a problem if the motors have to accelerate or decelerate quickly. Another disadvantage is the high price that has to be paid for the rig. This is caused by the use of four motors and a large frequency modulator.

4.2 Hydromotor

The main problem in powering a transmission test rig with an electric motor is the large inertia of the motor. This is tried to solve by using multiple motors. The inertia and torque are detached partial.

A hydromotor doesn't have this disadvantage. Compared with an electric motor, an hy- dromotor has a much higher power-inertia ratio. For generating the same power (torque), a much smaller hydromotor can be used. While the hydromotor power-inertia ratio is compara- ble with a combustion engine's, an hydromotor might be suitable for simulating a combustion engine.

More about driving a transmission rig with an hydromotor can be read in chapter 6.

Chapter 5

Driving simulation

Nowadays vehicle development has to satisfy some contradictory demands.

e Development times have to decrease. Development has to be done cheaper. The product has to be better.

Concerning the test period the first two demands can only be reached by decreasing the number of prototypes and reduction of the total test period. The last item can be reached by using more realistic tests. For transmissions, more and more tests will be done on a test rig in stead of testing the transmission in a prototype vehicle on a track or on public roads. Substituting the vehicle test into dynamic test rig tests is called "Road to Rig". The concept described in this chapter is developed by Ford [3]. In this chapter will be explained how the transmission load during a road / track test can be simulated realistically on a test rig.

5.1 Transmission loads

For a considerable extent, the transmission life expectancy depends on the imposed loads. These loads are caused by several factors:

Engine: The engine causes load acting on the transmission by enforcing torque. The load extent mainly depends on the engine concept (Diesel / Otto) and engine size. A larger engine causes higher torques. A diesel engine reaches high torques at low speed of rotation, an Otto engine at high rpm. The low speed range is always passed through, so diesel engines cause higher loads on the transmission.

Driver: The driver influence is divided in two phases: shifting phase and non shifting phase. When shifting, the driver loads the transmission by operating the clutch (torques on the transmission by his clutching and declutching behaviour), throttle and gearlever (shift forces). During non shifting phase the driver influences the transmission load by the way he operates the throttle. Drivers are divided in different categories, depending on driving behaviour. For example "hard" drivers change gear quickly. This causes wear in the transmission due to high shift forces and bad synchronization (not enough time).

Vehicle: Transmission load depends on the drive train concept (front / rear), vehicle class and payload.

Road: Driving in mountainous environments (use of gear 1-3) causes different wear than driving on motorways (high oil temperatures). The type of road is an important factor concerning transmission load.

Environment: On slippery roads in winter the transmission is loaded different than driving on smooth roads in summer. Imagine the transmission load if the wheels suddenly reach traction after slipping.

Figure 5.1 shows a summary of loads acting on a transmission.

I Stress Time Function I

e.g. e.g. e.g. e.g. e.g. Unladen Overloading Power unit Uneven road I weight I I Payload I I I I vibration I I surface I

Figure 5.1: Causes of the stress-time function.[l]

To be able to develop realistic test cycles for testing transmissions on a test rig, it has to be known what impact the factors described in the previous section have on the transmission load. That's why a lot of vehicle measuring in different vehicles and circumstance had to be done. Not all combinations had to be tested. Interpolation is possible if specific vehicle / engine combinations are investigated. Selection criteria were: vehicle class, drive train type, engine concept and engine size.

During vehicle tests Ford measured the following variables:

Torque and speed of rotation at transmission entrance Torque and speed of rotation at drive shaft Throttle position Clutch pedal position and force Brake pedal position Shift forces Vehicle acceleration Transmission and engine oil temperature

Most test rides are carried out at the Ford Proving Ground in Lommel. Test rides result in a user profile (stress-time function). These user profiles can be simulated on test rigs easily.

5.3 Dynamic vehicle simulations

Dynamic vehicle tests are carried out in stead of vehicle lifetime tests. Testing on a test rig takes only 40% of the time needed for vehicle testing. This is possible because of the next reasons:

a During the test only the transmission is involved, no other prototype parts 0 Testing goes on 24 hours a day, 7 days a week. No outside influences, e.g.

weather and test driver. Low load parts in the vehicle test are skipped in the test rig test.

The test rig used at Ford is equipped with an internal combustion engine. The measured throttle data (vehicle test) can be used directly for actuating the throttle of the test rig engine. When using an electric motor for powering the rig, the measured torque data has to be used. The dynamic vehicle simulation is divided is the following parts:

Driver: Three actuators simulate the driver: clutch, throttle and gearlever. The way these actuators operate the handles depends on the driver type: "hard", "middle" and "soft". When testing an automatic transmission only the throttle is actuated.

Vehicle: An electric brake simulates the resistance caused by air, rolling and hill climbing. Vehicle inertia is simulated by a flywheel.

Road: From all test tracks the topological values are measured. During the simulation, the gradient can be calculated constantly. The power necessary for riding up the slope is simulated by the electric brake.

5.4 Conclusion

Once the transmission-load behaviour is know, "Road to Rig" simulations are able to replace testing transmissions during vehicle tests. Less expensive prototypes are necessary to deter- mine life expectancy. Testing doesn't take so much time anymore so transmission testing becomes cheaper.

Chapter 6

Specifications new rig

As described in chapter 3 current available rigs for testing transmissions all have their ad- vantages and disadvantages, most times determined by the way the rig is driven (e.g. engine, electric motor). With an hydromotor or an hydromotor in combination with an electric motor it might be possible to simulate an combustion engine without having the engine and e-motor disadvantages. Before developing such rig, it has to be known what demands it has to satisfy.

6.1 Development criteria

The next criteria prescribe the demands:

Test type: As explained in chapter 2, transmission testing on a test rig can be divided in component testing and testing complete transmission. Testing complete transmissions consists of functional testing and road simulation tests. The necessary test rig capabili- ties dependent on the type of test. The development criterion is the degree the test rig has to be to able to simulate a vehicle (combustion engine and load).

Demands required for different test types:

Component test During dynamic component tests, only parts or small systems are tested. The drive only has to be able to generate a constant power at constant speed of rotation. A simple brake (brake disk or eddy current brake) is sufficient for loading the transmission.

Complete transmission testing

- Functional test For this type of test a relative simple test rig suffices. The motor has to be able to accelerate and decelerate smoothly for testing the shifting behaviour. Except loading the transmission (braking the motor), the brake system also has to be able to generate power. The transmission experiences negative torques like using the engine for braking. This can be realized by using a flywheel.

- Road simulation test This type of test requires a high performance test rig. The rig has to be an exact copy of the reality. The motor has to act like a combustion engine (same engine map, dynamic torque behaviour and inertia), the brake like the torque the transmission "feels" at the output shaft (vehicle inertia, air, rolling

and hill climbing resistance). In other words: on the test rig, the transmission has to be loaded the same way the vehicle does.

Transmission application: This determines the dimensions of the rig. Transmissions used in wind turbines are larger than motorcycle transmissions. The application also deter- mines the power that has to be generated by the drive and absorbed by the brake. Also important for the drive choice is the engine map that has to be simulated.

Recovering energy: Recovering energy saves money when using the rig. However, this makes the rig more complex and more expensive.

The first two criteria determine the basic development parameters of the test rig. Re- covering energy is optional, it improves (and complicates) the rig. These criteria only deal with the powering. All rigs need equipment for measuring variables like torque and speed of rotation. The type of test (what do you want to know) and the accuracy prescribe the type, number and accuracy of your measurement equipment. A computer operating system for controlling the test is always necessary.

6.2 Test rig specifications

The graduation project target is to develop a test rig that acts like a vehicle. For the new rig the next choices are made concerning the previous criteria:

Test type: Complete transmission testing Developing a test rig acting like a vehicle is only useful when testing a complete transmission. More specific: Road simulation test.

Transmission application: Volkswagen Bora 1.6L At first will be tried to simulate a Volkswagen Bora with an 1.6L7 4 cylinder, four stroke petrol engine, front wheel drive. For this specific vehicle is chosen because of the available data. This vehicle is examined accurately during the ZI-project. When the rig is able to simulate the Bora, it shouldn't be that difficult so simulate other vehicles. Another reason for choosing this car is it's engine map. The hydropumps available at the TU/e Fluid Power laboratory are able to generate enough power for simulating the Bora engine is.

Recovering energy: While energy recovery saves a lot of money the test rig will be equipped with an energy recovery system. This can be made electric (when an electric brake is used) or with fluid power (when an hydropump is used as brake). An hydraulic system is preferable. The energy recovery system has to save as much power as possible.

The choices made before lead up to the list of demands:

drive unit

Power: max. 75 [ k W ] Speed of rotation: max. 6000 [rpm] Torque: nominal max. 155 [Nm]

dynamic * [Nml

Inertia frequency x [Hz]

0,14 [Kgm2]

*) The torque magnitude and frequency depend on several parameters. During the grad- uation period, an engine - transmission model will be developed. With this model the user is able to determine the torque (dependent on crank angle) generated by the engine at different locations in the driveline. The torque pattern mainly depends on the engine speed and aver- age torque.

brake unit

Brakepower: Speed of rotation: Braking torque: Inertia

min. 75 [kW] max. 1700 [rpm] max. 1730 [Nm]

132 [Kgm2] vehicle mass and inertia wheels

The torque at the drive output (brake input) shaft can be decreased by using a gear transmission. The speed of rotation will increase.

When using a hydropump for braking there will also be a minimum speed of rotation. When the speed is too low, the oil circulates in the pump and becomes too hot after a while. Low speed for a short time is not pernicious.

Bibliography

[I] Lechner, G.; Naunheimer, H.: Automotive Transmissions, Fundamentals, Selection, De- sign and Application, Springer-Verlag, Berlin, 1999, ISBN 3-540-65903-X, blz. 381-418

[2] Naunheimer, H.; Zelssmann, H.; Lechner, G.: Satellitengestutzte Aufieichnung von Streckendaten als Vorabegrossen fur Fahrsimulationsprogramme, ATZ Automobiltech- nische Zeitschrift 95, 1993, No. 12, blz. 650-654

[3] Sciuto, M.; Hellmund, R.: "Road to Rig" - Simulationskonzept an Powertrain- Prufstanden in der Getriebeerprobung, ATZ Automobiltechnische Zeitschrift, 2001, vol 103 afl 4, blz. 298-307

[4] Abels, H.: Prufstand fur Pkw-Schaltgetriebe, Antriebstechnik 33, 1994, No. 6, blz. 44-47

[5] Falkenstein, J.; Hirschmann, K.H.: Elektrischer Priifstandsantrieb ersetzt Verbren- nungsmotor, Antriebstechnik 40, 2001, No. 9, blz. 70-72

[6] Shkolnikov, A.; Shkolnikov, A.: Mechanical Transmission Testing with the Aid of Syn- chronous Machinery, Chelyabinsk State Technical University, 1996, SAE Technical Pa- pers 961769

[7] Aken, Prof. Dr. Ir. Ch. van: Zuigermotoren, Lecture notes, Universiteit Gent, Fakulteit van Toegepaste Wetenschappen, 1996 - 1997, blz. 143-147, 154 - 155

[8] Internet site Schenck: http://www. carlschenck.de

[9] Internet site Lommel Prooving Ground Ford: http://users.skynet. be/lpg/lpg/indexZ.htm

[lo] Leaflet EUR-0-MATIC Nederland B. V., Tilburg, The Netherlands

List of Figures

2.1 Extent of component testing [l] . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 PC screen functional transmission test [lo] . . . . . . . . . . . . . . . . . . . . 2.3 General survey bench testing complete transmission [I] . . . . . . . . . . . . . 2.4 Ford Proving Ground Lommel . Left picture: overview of the ground . Right

. . . . . . . . . . . . . . . . . . . . . . . . . . . picture: the steep inclines [9]

3.1 Three basic test rig setups . Arrows represent the powerflow A) Energy flow process B) Energy loop process with modulator C) Energy loop process without modulator (torque test rig) . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3.2 Simple torque test rig for testing spur gears . [I] . . . . . . . . . . . . . . . . . 3.3 Torque test rig for testing high power CVT chains . Photo: Gear Chain Indus-

trial, Nuenen, The Netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4 Different lay-outs for testing complete transmissions . . . . . . . . . . . . . . . 3.5 Test rig equipped with an internal combustion engine . [8] . . . . . . . . . . . 3.6 Engine map electric AC motor and 1.6 1 combustion engine . . . . . . . . . . . 3.7 Torque generated by a 3 cylinder diesel engine . n=2100 [rpm] . 50% full load . 3.8 dynamic torque, frequency = 1 [Hz] . . . . . . . . . . . . . . . . . . . . . . . 3.9 dynamic torque, frequency = 2 [Hz] . . . . . . . . . . . . . . . . . . . . . . .

4.1 Test rig equipped with multiple electric motors . [5] . . . . . . . . . . . . . . . 4.2 Acceleration of a 4 cylinder 4 stroke engine . [5] . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . 5.1 Causes of the stress-time function . [1]