Effect of wheel slips on the durability of surace layer in the wheel-rail system

M IĘDZYNARODOW A K O N FEREN CJA ZESZYTY NA UK O W E N A UK O W O -TEC H N ICZN A _______________________ POLITECHNIKI ŚLĄSKIEJ 1999 POJAZDY S ZY N O W E NO W Y CH G ENERACJI ’99 TRA N SPO RT z.35. nr kol. 1415

Laura PERCH U Ć Instytut Fizyki, PAN Marek SITARZ Politechnika Śląska

EFFECT OF WHEEL SLIPS ON THE DURABILITY OF SURFACE LAYER IN THE WHEEL-RAIL SYSTEM

S u m m a ry . R elationship between a force on the drive wheel and a slip of the wheel on rail is one o f the m ajor factors dealing with the wheel/rail system. From theoretical considerations and investigation carried out it is clear that a slip at the circum ference o f drive wheels always appears during the rail vehicle running.

Under operation conditions a phenom enon o f macro slip and side-slipping is especially undesirable and harm ful, since it causes an excessive w ear o f rolling wheel surfaces ow ing to generation o f such defects as spallings, shellings, flat places etc. Therm al aspects of the m artensite form ation have been taken into consideration too.

WPŁYW POŚLIZGÓW KOLA NA TRWAŁOŚĆ WARSTWY WIERZCHNIEJ W SYSTEMIE KOŁO-SZYNA

S treszczen ie. Jednym z w ażniejszych czynników , które dotyczą problem u sprzężenia kói z szynam i, je st zależność m iędzy siłą na kole napędow ym a poślizgiem tego koła po szynie. Z przeprow adzonych badań laboratoryjnych i eksploatacyjnych oraz rozw ażań teoretycznych wynika, że poślizg na obw odzie kół napędow ych w ystępuje zaw sze w czasie ruchu pojazdu szynowego. W warunkach eksploatacyjnych szczególnie niepożądane i szkodliw e je st zjaw isko m akropoślizgu i zarzucania, które pow oduje nadm ierne zużycie powierzchni tocznych kół kolejow ych poprzez m.in. pow staw anie defektów takich jak wykruszenia, w yłuszczenia, płaskie m iejsca. W pracy om ów iono term iczne aspekty tw orzenia się martenzytu.

76 L. Perchuć, M. Sitarz

1. INTRODU CTION

D evelopm ent o f the railway transport involves a reduction in running times and there by an increase in running speeds. In the consequence cooperation conditions of the w heel/rail couple change for the worse. Basing on the previous laboratory and operation tests perform ed one can state that costs, resulting from railway set repairs due to forming of flat places and spalling of rolling wheel surfaces, will have been growing up considerably and actually they am ount to several hundred million dollars for the international railway industry annually [3].

The problem o f surface layer durability o f the system has remained still insolvable.

Research works, devoted to questions o f accelerated w ear of the rolling wheel surfaces have been developed [1, 2, 3], It is already known that a direct cause o f occurring that type of surface defects during the operation period is the following:

• Badly adjusted, frozen or dam aged brakes,

• M alfunctioning anti-slip devices,

• Relatively high braking forces as com pared with adhesion forces o f w heel-rail contact,

• Heavy proneness o f steels, used for rolling w heels, to forming m artensite structures.

Under real conditions of the wheel set operation on a track there appear wheel slips in relation to a rail at contact points [5], In conjunction with that there are distinguished three kinds o f slips: a slip in longitudinal direction (lengthwise the track axle), a crossw ise slip (perpendicular to the track axle) and a drilling slip (connected with the wheelset turn in relation to the axle orthogonal to a track plane as well as with wheel elastic strain).

The perform ed laboratory and operation tests have revealed that slips during starting and braking display the greatest values. By way o f exam ple the slip values are presented in the fig.

1 and 2.

In Japan in high-speed trains o f Shinkansen “STA R 21“ type there had been applied a new method of slip control due to which advantageous results were obtained, especially under bad conditions of the w heel/rail couple cooperation.[4] Usage of the proposed m ethod o f control (m onitoring and adjustm ent) o f wheel slips has contributed to a considerable im provem ent of the wheel/rail system operation and, thus, to a significant prolongation o f applied running times of wheels betw een succeeding rollings.

2. TH ER M A L ASPECTS O F M A RTEN SITE FORM ING

Available reference and experim ental data convinces that ju st slips in the wheel-rail system cause to achieve tem peratures, initiating structural changes in the wheel material. W hen a wheel is locked and is skidding along a track there a flat place is often form ed on the rolling surface of the wheel. At the sam e tim e there are generated high tem peratures adequate for pearlitic austenitizing the wheel steel, i.e., to reach a critical tem perature A l. M any authors have succeded to prove the above-m entined experim entally and theoretically (Jaeger, Archard.

Tanvir)[3], In their w orks there are accepted the follow ing assumptions:

• There-dim ensional problem , sim plified to a two-dim ensional area,

• Effects o f the plastic strain are neglected so that a slip velocity for all contact points is equal with a relative slip velocity (Vs)

Effect of wheel slips 77

• Total friction energy, transform ed into thermal one,

• a- a fraction o f the heat, absorbed by the body, 1-oc - a fraction o f the heat remained in the counterbody.

On the basis o f the above assum ptions there was formulated the follow ing dependence between the velocity V , the friction coefficient p., the thermal properties a and K, the temperature 0 and the slip R for a wheel being braked on a rail (R<1 and V>Vw):

© » = © « = ^ p - j ^ x ( i - v r ^ R )

Examples o f the tem peratures generated in the contact wheel/rail for skidding w heels are presented in the T able 1. The presented results were calculated according to the A rchard’s equation[3].

A (U ,J: B ( U |) - p n r b i c g i p rę d k o ś c i o bw o d o w y c h Vót w funkcji czasu l

F ig . 1. T h e g r a p h o f F f ( t) R y s . 1. W y k r e s F = f ( t)

F ig - 2 . T h e V = f ( t) c u r v e R y s .2 . K r z y w a V = f ( t)

L. Perchuć, M. Sitarz

Effect of wheel slips 79

Table 1 Tem peratures, generated at the wheel/rail contact during wheel skidding for the different

running tim es V and the friction coefficient p[3]

P. = 3200 kg P. = 10 000 kg

a = 5,45 mm a = 8 mm

V (m/s) H = 0 . 1 p = 0.3 |i =0 . 1 ■ p = 0,3

0.5 1 2 2 367 218 653

I 177 531 314 941

1,5 219 657 388 1163

2 254 763 450 1350

3 314 941 554 1663

1 0 581 1744 1025 3075

2.1. F o rm a tio n o f s u rfa c e defects

An increase in the tem perature, caused by the slip and rapid cooling can initiate m artensite transformation and on a given m icraarea o f the skidding wheel m aterial there m artensite is formed. M acroscopically it is easily identified because it is visible in the form o f clear, shiny spots on operated rolling-w heel surfaces (fig. 3) [3,6]. The untem pered m artensite is a brittle phase and has a cristal structure (tetragonal system), leading to an about 0.5% volumetric expansion as com pared with wheel-m aterial pearlitic structure. That change in volum e results in very high com pressive residual stresses into the m artensite structure and respectively high tensile residual stresses into the m aterial, surrounding the above-m entioned m artensite structure.

If a w heelset with flat places has been still w orking as well as the m artensite has not been removed through im proving the profiles then w ithin the area o f m artensite structure there will have been form ed fatigue cracks ow ing to the im pact o f contact stresses during wheel skidding. Those cracks will still extend and relatively large material pieces will detach off the rolling wheel surface. Such spallings are used to leave cavities o f 1 to5 mm deep, surrounded by cracks.

The spallings, caused by flat places, can also create great im pact loads, originating from the skidding wheel. They can contribute to severe dam ages in railw ay track structure, vehicle mounting and bearings as well.

There are som e ways leading to a decrease in the costs by m eans of:

• im provem ent o f anti-slip devices,

• im plem entation o f steel grades, and for railway wheel production, with less pronem ess to generate m artensite structures.

80 L. Pcrchuć. M. Sitarz

F ig .3 . A f r a g m e n t o f r o ll in g w h e e l s u r f a c e w ith d is ti n c tl y v is ib l e p la c e s , e v i d e n c i n g th e in i tia tio n o f m a rtc n s iic t r a n s f o r m a tio n ( c l e a r , s h in y a r e a s ) a n d s p a ll in g s ( d a r k a r e a s ) . B e l o w o n e c a n s e c a n im a g e ( f r o m the s c a n n i n g m i c r o s c o p e ) o f th e s u r f a c e d e f e c t o f th a t ty p e

R y s .3 . F r a g m e n t p o w ie r z c h n i to c z n e j k o la z w y r a ź n ie w id o c z n y m i m ie js c a m i, ś w i a d c z ą c y m i o z a p o c z ą tk o w a n iu p r z e m ia n y m a r te n z y t y c z n e j ( ja s n e , b ły s z c z ą c e p o la ) i w y k r u s z e n ia m i ( c i e m n e p o la ) . P o n iż e j o b r a z z m i k r o s k o p u s k a n in g o w e g o t e g o ty p u d e f e k t u p o w ie r z c h n io w e g o

Effect o f wheel slips 81

2.2. S lip c o n tro llin g in th e s y s te m w h e e l- r a il

To solve the problem o f wheel slipping o f electric railway vehicles, first of all, one should study closely this phenom enon during starting /braking. An axle velocity should be determined precisely to this end. In this case one should be careful at choosing and installing of sensors. At the first place eccentricity between the axle, where the sensor is installed, and the axle of steel sensing elem ent must be very' low (small), because it causes fluctuations in an instantaneous velocity. A conventional m ethod for determ ining the wheel slip is based on calculation of im pulses o f the steel sensing elem ent, fixed at one end o f the wheel axle (of the wheelset) or at the tractive m otor shaft. It is very difficult to choose a sensor (a sensing element) with so many im pulses per rotation as an encoder has in industrial use. T herefore in the work [4] there was applied a m ethod for calculation o f the mean im pulse width. Usage of this method is essential to detect small slips and adhesion loss. A practical method, controlling slips for vehicles with electric three-phase motors, is based on the use o f the least velocity o f all the m otor axles as a value o f the reference frequency o f an inverter controlling device and feeding asynchronous traction engines. It is proposed to this end to detect small pseudo-slips too. There is also a possibility to predict adhesion "on line” so that to cope with slip-readhesion reiteration under bad conditions of thé wheel/rail system operation, for example, icy rail, steep gradient, etc. Test results have evidenced the effectiveness o f the proposed control o f adhesion loss[4].

3. CONCLUSIONS

Considering a w orld-w ide interest in the problem o f martensitc forming around flat places on the rolling surfaces o f railway wheels, caused by slips it seems to be advisable:

• to carry on investigation, leading to a better understanding of the conditions, prom oting the form ation o f structures o f that kind,

• to im prove anti-slip devices by means o f im plem entation of the m ethod o f controlling slips and readhesion prediction.

4. ACK N OW LEDGEM ENTS

This work was partially supported by the Polish C om m ittee for Scientific Research, KBN, under grant 9 T12C 044 16.

REFERENCES

1. Perchuć L., Sitarz M., W iederm ann J.: K lasyfikacja i przyczyny defektów kolejowych zestawów kołow ych w czasie eksploatacji, K onferencja N aukow o-Techniczna “Postęp i przemiany w P K P -1997” .

82 L. Perchuć, M. Sitarz

2. Jergeus J.: M artensite form ation and dam age around railway w heel flats. 6lh International Heavy Haul C onference, Cape Town, South Africa, 1997.

3. Magel E., Kalousek J.: M artensite and contact fatigue initiated wheel defects, 12lh International W heelset Congress, Safety and High Efficiency, Sem ptem ber 21-25, 1998, Qingdao, China, p. 100-111.

4. W atanabe T., Y am anaka A.: A dhesion phenom ena and optim ization o f readhesion control for high speed trains with wheel-rail adhesion prediction. Railw ay Technical Research Institute, T raction C ontrol. httD://w w w .rtri.or.ip/infoce/w crr97/D 4537/D 4537.htm . 5. D żuła S.: D ynam ika i stateczność układów w ruchu złożonym z uwzględnieniem

elem entów w irujących, cz.2, Pojazdy Szynowe nr 2, 1998, s. 8-14.

6. Perchuć L., Sitarz M ..: B adania laboratoryjne i eksploatacyjne poślizgu kól napędowych pojazdów szynowych. M iędzynarodow a K onferencja “T ransport’97” . Ostrawa-Katowice 1997, s.483-491.

Recenzent: D r hab.inż. M arek Sitarz Prof. Politechniki Śląskiej

S treszczenie

Jednym z w ażniejszych czynników , które dotyczą problem u sprzężenia kól z szynami jest zależność m iędzy siłą na kole napędowym a poślizgiem tego koła po szynie. Z przeprow adzonych badań laboratoryjnych i eksploatacyjnych oraz rozw ażań teoretycznych wynika, że poślizg na obw odzie kól napędowych występuje zaw sze w czasie ruchu pojazdu szynowego. W w arunkach eksploatacyjnych szczególnie niepożądane i szkodliw e jest zjawisko m akropoślizgu i zarzucania, które pow oduje nadm ierne zużycie powierzchni tocznych kól kolejow ych poprzez m.in. powstawanie defektów takich, ja k wykruszenia, wyłuszczenia, płaskie m iejsca. W pracy om ów iono term iczne aspekty tw orzenia się martenzytu.