477

The Influences of the Wheel Profiles on the Wheel Wear and

Vibrational Characteristics of the Passenger Cars Running on

the Seoul-Pusan Conventional Line∗

Bu-Byoung KANG∗∗ and Chan-Woo LEE∗∗

Wheels of the railway vehicle play the important role for driving train through wheel-rail interaction. Especially wheel profile is one of the most important design factors to rulethe running stability and safety of train. Accordingly maintenance of wheel like wheel pro-file control is also very important for securing safety and stability of train operation. Thisstudy presents the wheel wear measurement results of Saemaeul running on the conventionalline. The train set included three different cars which have different shape of wheel profileincluding KNR profile currently used in Saemaeul. Train set was operated on Seoul-Pusanline with fixed train set formation for commercial service. Wheel wear measurements wereperformed periodically. We can find the influence of wheel profile on the wheel wear of thetrain running on the conventional line through the measurement results.

Key Words: Wheel Profile, Flange Wear, Arctype Profile, Carbody Vibration

1. Introduction

KTX, which is planned to get introduced and oper-ated in Korea, consists of 20 cars for 1 set. The technicalrequirements of these high speed trains are being provedthrough design, manufacture and performance test basedon the contract basis. But there is a little difference inoperation-environment and railroad-conditions for a highspeed train between Korea and France.

Currently, KTX is designed to operate in existing rail-roads as well as new high speed railroads. For that, wheelsfor a high speed train for both of new high speed railroads& old existing railroads and examination for the interfaceproblems, which can be occurred with old existing rail-roads, are needed(1), (2). Especially for KTX, train-rail in-terface examination in new high speed railroads is almostcompleted in O-Song and Chun-An rail. Service routes ofKTX consists of new high speed railroads and old exist-ing railroads, but suitability test of the train-rail interfacecannot be conducted in current conditions on the existingrailroads although it is important and necessary.

Therefore, for the preliminary inspection of train-

∗ Received 9th January, 2004 (No. 04-5008)∗∗ Rolling Stock Research Department, Korea Railroad

Research Institute, #385, Woulam-Dong, Viwang-city,Kyonggi-Do 437–757, Korea. E-mail: bbkang@krri.re.kr,cwlee@krri.re.kr

safety while operating KTX in the existing railroads, theeffect on operation by wheel wear profile of a train whichhas an identical suspension is studied by examining thewear properties of Saemaeul Ho train wheel with appli-ance of KTX train wheel profile, Heumann (1/20) wheelprofile and KNR (1/40) wheel profile.

2. On-Line Test

2. 1 Test method and procedureThe vehicle, used in the test, is Saemaeul passen-

ger train, operated in Kyeongbu Line between Busan andSeoul, with attachment of ASEA truck. Car No.430,No.552 and No.639 of Saemaeul Train are used in the test.Three different shapes of wheels are attached to 8 wheelsof each vehicle to observe wheel wear property.

KTX wheel (NF F01-112, 1/40), Saemaeul Ho wheel(KNR 1/40) and new model of Mugunghwa Ho wheel(1/20 modified heumann) are applied to No.430, No.552and No.639 as wheel profile by each testing-car. Originalprofile of each wheel is compared and shown in Fig. 1. Fora testing period, wheel wear status is tracked down and in-vestigated on travel up to more than 100 000 km that is thestandard distance covered for inspection after profiling ofeach wheel(3). During testing period, wheel profile & wearamount is periodically measured by Miniprof system.

2. 2 The type of test carASEA truck for Saemaeul passengertrain is shown in

JSME International Journal Series C, Vol. 47, No. 2, 2004

478

Fig. 1 Comparison of the initial wheel profiles

Fig. 2 ASEA bogie

Fig. 3 Wheel number arrangement

Figs. 2 and 3 shows wheel arrangement and wheel number.Figure 2 shows ASEA truck description that is used

in the test. ASEA truck like figure has shebron rubberspring as primary suspension and coil spring as secondarysuspension. Generally, it is known that there is less wearamount for applying shebron rubber spring on primarysuspension than coil spring(4), (5).

3. Results and Discussion

3. 1 Wear rateThe measurement is for thickness (Sd), gradient (qR)

and height (Sh) of wheel flange, which is used as a cri-terion of inspection. Figure 4 shows a standard parame-ter, used for inspection of wheel wear amount and Table 1shows an initial profile dimension of three different pro-files used in the test.

Figures 5 – 7 and Table 2 shows wear rate per unitdistance of all 8 different wheels after running test withtrains that have three different wheel profiles one another.

Figure 5 shows that KNR 1/40 wheel has the mostwear rate for thickness of wheel flange, then KTX 1/40wheel, and 1/20 h wheel was even getting thicker insteadby having minus (−) value. For 1/20 h wheel, thicknessand gradient of wheel flange even get increased due to

Table 1 Initial profile dimension (unit: mm)

Fig. 4 Criteria of Sd (flange thickness), qR (flange gradient),Sh (flange height)

the lowered base surface by wear of tread surface whengetting the wheel profile through measurement followingthe criteria of Fig. 4. In other words, wear rate aroundtread surface area is bigger than around wheel flange area.From wear rate for the height of wheel flange in Fig. 6,1/20 h wheel has a little higher wear rate and there is al-most no difference of wear rate between KTX 1/40 wheeland KNR 1/40 wheel can be shown. Figure 7 show thatwear rate of gradient of wheel flange has a similar phaseto that of thickness of wheel flange and KNR 1/40 wheelhas the biggest and 1/20 h wheel has even increased gradi-ent of wheel flange instead by having minus (−) value.

Figure 8 show the average of wear rate for each 8wheels with three different testing vehicles. Compared toKTX 1/40 wheel, KNR 1/40 wheel has 3.4 times higherwear rate in thickness of wheel flange and 3.3 times higherwear rate in gradient of wheel flange. For wear rate inheight of wheel flange, KTX 1/40 wheel and KNR 1/40wheel have similar rates, and 1/20 h wheel has 2.2 timeshigher rate.

The wheel number (No.) shown in Figs. 5 – 7 and Ta-ble 2 is gotten by arrangement criteria in Fig. 3. Althoughthere is a change in wear rate by location of vehicles, anapparent difference in wear rate by wheel profile besidesheight of wheel flange can be shown.

3. 2 Transition of the wear ratesFigures 9 – 11 show the change of wheel profile as

wear for each wheel profile proceeds, and Table 3 showaverage wear amount and wear transition of thickness(Sd), height (Sh) and gradient (qR) of wheel flange indifferent traveling distances covered. From Table 3, thatchange of wear rate for thickness (Sd) and gradient (qR)of wheel flange tends to decrease for KNR 1/40 wheel aswear proceeds is shown. Also, it tends to increase littlefor KTX 1/40 wheel, but it changes by only small amount

Series C, Vol. 47, No. 2, 2004 JSME International Journal

479

Fig. 5 Flange thickness wear rate

Fig. 6 Flange height wear rate

Fig. 7 Flange gradient wear rate

compared to KNR 1/40 wheel.In Fig. 12, for worn wheel profile by each wheel pro-

file after traveling about 90 000 km, that KTX 1/40 wheeland KNR 1/40 wheel have similar wheel flange profilesas wear proceeds is shown. This is assumed by the rea-son that wheels get adapted to the similar wear profile bywheel and rail interaction as wear proceeds. However, for1/20 h wheel, there is almost no change in wheel flangeprofile and that large tread wear from bottom area of wheel

Table 2 Wear rate of each wheel

Fig. 8 Average of wear rate

Fig. 9 Evolution of wheel profile (1/20Heumann) until runningdistance 105.478 km

Fig. 10 Evolution of wheel profile (KTX 1/40) until runningdistance 82.073 km

Fig. 11 Evolution of wheel profile (KNR 1/40) until runningdistance 105.478 km

JSME International Journal Series C, Vol. 47, No. 2, 2004

480

Table 3 Transition of the wear rate during the running

Fig. 12 The comparison of worn profiles after running about90.000 km

Fig. 13 Locations of accelerometers

flange to the center of tread area occurs is shown.Consequently, for KNR 1/40 wheel, it has a extreme

change in wheel profile depending on the traveling dis-tance, especially around wheel flange area, and there isa little profile change in bottom area of wheel flange forKTX 1/40 wheel and there is no change in wheel flangearea with a little wear in tread surface area for 1/20 hwheel. For 1/20 h wheel, as wear proceeds, there is pos-sibility that adversely affects operation-safety in a highspeed running due to the increase of equivalent conicityof worn wheels, comparing to KTX 1/40 wheel and KNR1/40 wheel. So we should examine the running perfor-mance of worn profiles.

Wheel profile relates to operation-safety and stabil-ity of a train(6). As wear proceeds, equivalent conicityof wheel tread surface increases, and the increased conic-ity causes unstable operation of a train in a certain speed.Therefore, suspension feature must be considered in de-sign phase for operation-safety of a train. But, there isdifficulty in designing suspension because it is hard toguarantee operation-safety in a high speed traveling when,especially, equivalent conicity of wheel tread surface in-creases due to a extreme change in wheel profile by wearcaused in traveling. So, research for effect on operation-safety of a train by a change in wheel profile and researchfor designing suspension to maintain safety within a mer-cantile operation speed range need to be proceeded.

Fig. 14 Vertical acceleration (Ockchun→ Kimchun)

Fig. 15 Lateral acceleration (Ockchun→ Kimchun)

3. 3 Vibration characteristicsTo investigate the influence of wheel profile on vi-

brational characteristics, running test was conducted atthe initial stage of the wear test. The locations of mea-surement in the train with 16 cars for 1 set were shownin Fig. 13. Acceleration was measured in the car 10, 12and 13. Carbody vertical acceleration was filtered with30 Hz low pass filter and transferred to RMS to comparethe magnitude by dB. Carbody lateral acceleration wasfiltered with 2 Hz filter for low frequency characteristics.

Figures 14 and 15 shows vertical and lateral accel-eration measured on the line between Ockchun and Kim-chun. Table 4 shows accelerations measured on the Seoul-Busan line on the three different sections Suwon-Chunan,Ockchun-Kimchun, and Dongdaegu-Samrangjin.

Table 4 shows that lateral vibration is higher in thecase of KTX 1/40 wheel, and KNR 1/40 wheel and 1/20 hwheel have similar value, vertical vibration is higher in thecase of KNR 1/40 wheel, and KTX 1/40 wheel and 1/20 hwheel have similar value.

As a result, 1/20 h wheel profile shows better vibra-tional performance than other profiles. It seems that theconicity of 1/20 h wheel is acceptable to the Saemaeulpassenger train in the operating speed range lower than

Series C, Vol. 47, No. 2, 2004 JSME International Journal

481

Table 4 Measured vibration level

140 km/h. The vibrational characteristics of the train withworn 1/20 h profile will be examined right before the re-profiling work.

4. Conclusion

Running test is conducted on the Saemaeul passen-ger train in Kyeong-Bu Line between Seoul and Busanby attaching KTX wheel (NF F01-112, 1/40), KNR 1/40wheel and 1/20 heumann wheel on it; as results, effect ofwheel profile on wheel wear and carbody vibration can beanalyzed under the identical test-environment to the lineconditions of the existing railroads and the suspension oftrains. Below results are obtained from the test.

1. KNR 1/40 wheel has the largest wear of wheelflange thickness and flange gradient, KTX 1/40 wheel isthe next and 1/20 h wheel has mere wear of wheel flange

occurrence.2. For KNR 1/40 wheel, enormous profile change oc-

curs due to wheel flange area wear, but identical wear pat-tern is detected to that of KTX 1/40 wheel after travelinga certain distance.

3. For 1/20 h wheel, larger wear rate is found on treadsurface area than wheel flange area. 1/20 h wheel profileshows better vibrational performance than other profiles.

In the future, we should examine the safety and sta-bility of train with worn 1/20 h profile for the more reliableevaluation of 1/20 h wheel applicability in the future.

References

( 1 ) Kang, B., Lee, H. and Wang, Y., Study on Wheel WearCharacteristics of High Speed Train Running on theConventional Line, Proceedings of Korean Society forRailway Autumn Conference, (2000), pp.120–125.

( 2 ) Lee, H. and Kang, B., Study about Prediction ofWheel/Rail Wear & Maintenance Standards, KRRI Re-port, (1999), pp.24–56.

( 3 ) Huh, H., Lee, C. and Kim, H., Analysis on WheelFlange Wear Property of Passenger Trains, Proceed-ings of Korean Society for Railway Autumn Confer-ence, (2002), pp.115–118.

( 4 ) Huh, H. and Lee, C., Research on Operational En-hancement of NT21 Truck, KRRI Report, (1998),pp.40–43.

( 5 ) Huh, H., Lee, C. and Kang, B., Development on Op-timized Management System of Wheel Tread Surface,KRRI Report, (2002), pp.133–136.

( 6 ) Aea Technology, Introduction to Railway Vehicle Dy-namics — Ch9 Wheel and Rail Profiles, Vampire Sem-inar, (2002).

JSME International Journal Series C, Vol. 47, No. 2, 2004