where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Nondestructive testing of railroad wheels at the German Bahn AG
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

25408 views
00:46 Jun-06-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Nondestructive testing of railroad wheels at the German Bahn AG



Dear Forum participant

In sad coincidence with this month's railway testing issue, Germany just
last week suffered its worst railway accident in more than half a century.
It is very strange that I (Rolf Diederichs) came back from the ECNDT '98
conference Copenhagen via the same route and on the same train, with topical
railway material in my suitcase, only 5 days before the Eschede tragedy.

The accident occurred on June 3 - just as we were putting the final
touches on this month's issue.
We went ahead and released the new issue, with a few additions pertinent to
the accident, although it is not our intention to focus on catastrophic news
(That is fully covered on the CNN webpage).

We'd rather not get into speculations (with the clarity of hindsight)
on what may have been the exact reason for the derailment, but we would like
to see some dialogue regarding the technical background, particularly now that
it has come to light that a broken wheel which was tested ultrasonically has
been implicated. Let us first focus on the following two articles:

1. The ECNDT '98 presentation of
E. Schneider and R. Herzer - Ultrasonic Evaluation of Stresses in the Rims
of Railroad Wheels. You can also hear it in RealAudio.
http://www.ndt.net/article/ecndt98/rail/306/306.htm

2. Archive 1997 No. 6 - June:
Nondestructive testing of railroad wheels at the German Bahn AG
H. Hintze, Deutsche Bahn AG, Kirchmöser (English summary by Rolf Diederichs)
New: The complete German Version: Zerstörungsfreie Überwachung an Radsätzen
der Deutschen Bahn AG. We'll have an English translation of the full version soon.
This article asserts that the applied ultrasonic testing method does not
effectively recognize all possible defects of the wheel. For that reason a task
optimized the technique which showed first good results and plans for the
necessary modifications are underway for a long term investigation at the service
station inHamburg. Our question mark: Are those investigations completed and
modifications applied??
http://www.ndt.net/article/report/df97/hintze/hintze.htm

It seems that there is a great degree of uncertainty regarding applied
techniques for ICE wheels - Not that the technology is not available - it is!
However, in-service inspection budget issues can compromise safety.
Wheels must be tested economically without disassembling, but how can the
dual needs of safety and economy be met?
More reliable Ultrasonic techniques are used in manufacturing and refurbishing
shops than in-service inspection.

Mr. Hintze Deutsche Bahn AG, in his article:
'The experience of two units (since 1992 Hamburg and 1994 in Munich) showed
that ultrasonic testing with this configuration could not be used.
After a relatively short operation time the wheel material characteristics
changed by hardening in such a way that sound was already absorbed and no
evaluation could be performed by futureinspection. For that reason a task force
together with the FhG IzfP optimized the technique ....'

This was stated at the German-French Workshop (DGZfP/Cofrend)
in Aachen in April '97. There were plans to investigate the optimization in Hamburg.
But what happened?
Was it successful?
Was it applied to Munich as well?

Who knows more about this subject?
We will be looking into this in the next few days.

We hope you will all join us in discussing the details of the wheel testing
issue as well as related topics such as 'Risk based NDT management' and
Probability of detection.

Personally, it is troubling to me that as usual, we see economic issues
take precedence over safety.

Rolf


 
01:43 Jun-07-1998

H. Wuestenberg, BAM - Berlin

R & D
BAM Berlin,
Germany,
Joined Nov 1998
26
Re: Nondestructive testing of railroad wheels at the German Bahn AG Remark concerning Mr. Hintze's article

The ultrasonic inspection of the rim of the railrod wheel is based on a Surface wave travelling around the rim. Mr. Hintze gives some hints concerning the basic problem of the method: It's crack detection potential seems to be limited at aged wheels where the risc of cracks is increased. This is due to the increased scattering of the Rayleigh waves at shattered surfaces. The special contactless excitation and reception of the ultrasonic waves at that application is surely facilitating the practical use of the UT inspection, but it is not essential for the basic function of the method. Rayleigh wave probes with liquid couplants - e.g with a wedge or a comb structure - can do the same job. If there are limitations and problems with the crack detectability of the method at wheels with a high "mileage" (e.g. greater then 200.000 km), then we have to admit that this is due to the Rayleigh wave diffraction at the shattered rim surface. Therefore it seems to me astonishingif the modification of the probe shape or the application of a more magnetostriction based wave excitation would improve the situation. Therefore we have to rise the basic question about the principal limitation of this approach for aged wheel rim's. Can we expect that even at somehow shattered rim's a high crack detection probability (that means almost 100%) will be assured by this "single shot" inspection? There are other informations that about 40% of the inspected wheels are of that kind.
Or must we look for methods which are scanning the surface? E.g. with TOFD probe arrangements, with creeping wave probes or others.
Further commentaries are requested.

H. Wuestenberg
BAM - Berlin




 
07:46 Jun-08-1998
C.M. Fortunko
Re: Nondestructive testing of railroad wheels at the German Bahn AG : Remark concerning Mr. Hintze's article

: The ultrasonic inspection of the rim of the railrod wheel is based on a Surface wave travelling around the rim. Mr. Hintze gives some hints concerning the basic problem of the method: It's crack detection potential seems to be limited at aged wheels where the risc of cracks is increased. This is due to the increased scattering of the Rayleigh waves at shattered surfaces. The special contactless excitation and reception of the ultrasonic waves at that application is surely facilitating the practical use of the UT inspection, but it is not essential for the basic function of the method. Rayleigh wave probes with liquid couplants - e.g with a wedge or a comb structure - can do the same job. If there are limitations and problems with the crack detectability of the method at wheels with a high 'mileage' (e.g. greater then 200.000 km), then we have to admit that this is due to the Rayleigh wave diffraction at the shattered rim surface. Therefore it seems to me astonishing if the modification of the probe shape or the application of a more magnetostriction based wave excitation would improve the situation. Therefore we have to rise the basic question about the principal limitation of this approach for aged wheel rim's. Can we expect that even at somehow shattered rim's a high crack detection probability (that means almost 100%) will be assured by this 'single shot' inspection? There are other informations that about 40% of the inspected wheels are of that kind.
: Or must we look for methods which are scanning the surface? E.g. with TOFD probe arrangements, with creeping wave probes or others.
: Further commentaries are requested.

: H. Wuestenberg
: BAM - Berlin


I agree with Dr. Wuestenberg's comments. The subject of wheel inspection
should be revisited. However, I think that we must be careful about using
magnetostrictive ultrasonic transducers, a version of the EMAT, which
operates on Lorentz/Ampere forces.

Chris


 
08:19 Jun-08-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG s accordingly to Mr. Wuestenberg's message)

In our report [1] you can read the transcript of my questions and concerns which I presented to Mr. Hintze's after his speech in Aachen at the German French workshop. At that time concerns about the reliability of the method were already open discussed. We know that other people in Aachen had the same doubt.

In my opinion the DB was somewhat remiss all these years, in that they took no action toward the purchase of a better system, preferably one based on scanning testing methods such as the pulse echo technique in combination with TOFD (time of flight diffraction). Similar systems are working already in refurbishing shops at the DB in Darmstadt and Neumuenster. [2,3]

Should we travel on an ICE before such a system is installed?

Rolf Diederichs


1. http://www.ndt.net/article/report/df97/hintze/hintze_d.htm#6
2. http://www.irt.co.il/products/as220pr/as220pr.htm
3. http://www.krautkramer.com



 
01:06 Jun-08-1998

Christian Grosse

Teacher,
Technische Universität München,
Germany,
Joined Nov 2000
11
Re: Nondestructive testing of railroad wheels at the German Bahn AG >
>Based on the information available to us, we know that in Hamburg and Munich UT systems have worked since 1992.
>

Anyone have some details about the UT systems the Deutsche Bahn AG is
using for inspection? What type of measuring device and transducers are
used?

Regards

Chr. Grosse

--------------
Dr. Christian Grosse
Institut fuer Werkstoffe im Bauwesen (Construction Materials)
Universitaet Stuttgart (University of Stuttgart)
+++ phone/fax/address at:
http://www.uni-stuttgart.de/UNIuser/fmpa/grosse/grosse.html


 
02:44 Jun-08-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Update Since 1992 the Deutsche Bahn AG used a system from the company
Hegenscheidt (Erkelenz Germany).
The system works with EMAT probes generating surface waves (see Hintze's article [1]).
The method was licensed from the FhG IZfP Saarbrücken 10 Years ago.

Today we received also some officially details from
Christine Geißler-Schild
Sprecherin Forschung und Technologie
Tel. 069 / 265 68 87

The Inspectomat was applied in the 2 days inspection (approx. 3.500),
However, when discovered the problems of this method (details see 1.)
the system was switched off (the exact date we will receive later,
we assume this happened in 1997).

From this time an ultrasonic inspection was ONLY applied within the
big inspection in the refurbishing plant in Nürnberg
(350.000 - 400.000 km approx.. in a period of 6-7 month).

Rolf Diederichs

BTW: Today you may see inforamtion in the TV ZDF Heute Journal at 21:45.

1. http://www.ndt.net/article/report/df97/hintze/hintze_d.htm



 
02:54 Jun-08-1998
chris gartside
Re: Nondestructive testing of railroad wheels at the German Bahn AG : : Remark concerning Mr. Hintze's article

: : The ultrasonic inspection of the rim of the railrod wheel is based on a Surface wave travelling around the rim. Mr. Hintze gives some hints concerning the basic problem of the method: It's crack detection potential seems to be limited at aged wheels where the risc of cracks is increased. This is due to the increased scattering of the Rayleigh waves at shattered surfaces. The special contactless excitation and reception of the ultrasonic waves at that application is surely facilitating the practical use of the UT inspection, but it is not essential for the basic function of the method. Rayleigh wave probes with liquid couplants - e.g with a wedge or a comb structure - can do the same job. If there are limitations and problems with the crack detectability of the method at wheels with a high 'mileage' (e.g. greater then 200.000 km), then we have to admit that this is due to the Rayleigh wave diffraction at the shattered rim surface. Therefore it seems to me astonishing if the modification of the probe shape or the application of a more magnetostriction based wave excitation would improve the situation. Therefore we have to rise the basic question about the principal limitation of this approach for aged wheel rim's. Can we expect that even at somehow shattered rim's a high crack detection probability (that means almost 100%) will be assured by this 'single shot' inspection? There are other informations that about 40% of the inspected wheels are of that kind.
: : Or must we look for methods which are scanning the surface? E.g. with TOFD probe arrangements, with creeping wave probes or others.
: : Further commentaries are requested.

: : H. Wuestenberg
: : BAM - Berlin

:
: I agree with Dr. Wuestenberg's comments. The subject of wheel inspection
: should be revisited. However, I think that we must be careful about using
: magnetostrictive ultrasonic transducers, a version of the EMAT, which
: operates on Lorentz/Ampere forces.

: Chris

Asfar as I know, the techniques applied in Germany have used
non-conventional methods of generating ultrasound. Should we
be looking at more traditional methods, or have these been discarded
in the past for any reason.
There is a US patent referenced on ndt.net from German company
Hegensheidt MFD - my company has also been involved with Hegensheidt
in producing an ultrasonic testing system which is an accessory for
underfloor wheel lathes. These lathes are used by all Maintenance workshops
to remachine the profile on worn wheels - we have fitted an ultrasonic probe
head onto the toolpost, together with a multi-channel ultrasonic system.

This uses a conventional pulse echo technique with an irrigated probe
scanning over the wheel tread surface, using the motion which normally
drives the cutting tool. The inspection is designed to achieve the same
sensitivity as specified in the ISO5948 standard, really intended for
inspection of new wheels.

Is there any experience of applying these standards for in-service tests?





 
07:21 Jun-08-1998

Godfrey Hands

Engineering,
PRI Nadcap,
United Kingdom,
Joined Nov 1998
281
Re: Nondestructive testing of railroad wheels at the German Bahn AG
: Dear Forum participant

: In sad coincidence with this month's railway testing issue, Germany just
: last week suffered its worst railway accident in more than half a century.
: It is very strange that I (Rolf Diederichs) came back from the ECNDT '98
: conference Copenhagen via the same route and on the same train, with topical
: railway material in my suitcase, only 5 days before the Eschede tragedy.

: The accident occurred on June 3 - just as we were putting the final
: touches on this month's issue.
: We went ahead and released the new issue, with a few additions pertinent to
: the accident, although it is not our intention to focus on catastrophic news
: (That is fully covered on the CNN webpage).

: We'd rather not get into speculations (with the clarity of hindsight)
: on what may have been the exact reason for the derailment, but we would like
: to see some dialogue regarding the technical background, particularly now that
: it has come to lightthat a broken wheel which was tested ultrasonically has
: been implicated. Let us first focus on the following two articles:

: 1. The ECNDT '98 presentation of
: E. Schneider and R. Herzer - Ultrasonic Evaluation of Stresses in the Rims
: of Railroad Wheels. You can also hear it in RealAudio.
: http://www.ndt.net/article/ecndt98/rail/306/306.htm

: 2. Archive 1997 No. 6 - June:
: Nondestructive testing of railroad wheels at the German Bahn AG
: H. Hintze, Deutsche Bahn AG, Kirchmöser (English summary by Rolf Diederichs)
: New: The complete German Version: Zerstörungsfreie Überwachung an Radsätzen
: der Deutschen Bahn AG. We'll have an English translation of the full version soon.
: This article asserts that the applied ultrasonic testing method does not
: effectively recognize all possible defects of the wheel. For that reason a task
: optimized the technique which showed first good results and plans for the
: necessary modifications are underway for a long term investigation at the service
: station in Hamburg. Our question mark: Are those investigations completed and
: modifications applied??
: http://www.ndt.net/article/report/df97/hintze/hintze.htm

: It seems that there is a great degree of uncertainty regarding applied
: techniques for ICE wheels - Not that the technology is not available - it is!
: However, in-service inspection budget issues can compromise safety.
: Wheels must be tested economically without disassembling, but how can the
: dual needs of safety and economy be met?
: More reliable Ultrasonic techniques are used in manufacturing and refurbishing
: shops than in-service inspection.

: Mr. Hintze Deutsche Bahn AG, in his article:
: 'The experience of two units (since 1992 Hamburg and 1994 in Munich) showed
: that ultrasonic testing with this configuration could not be used.
: After a relatively short operation time the wheel material characteristics
: changed by hardening in such a way that sound was already absorbed and no
: evaluation could be performed by future inspection. For that reason a task force
: together with the FhG IzfP optimized the technique ....'

: This was stated at the German-French Workshop (DGZfP/Cofrend)
: in Aachen in April '97. There were plans to investigate the optimization in Hamburg.
: But what happened?
: Was it successful?
: Was it applied to Munich as well?

: Who knows more about this subject?
: We will be looking into this in the next few days.
:
: We hope you will all join us in discussing the details of the wheel testing
: issue as well as related topics such as 'Risk based NDT management' and
: Probability of detection.

: Personally, it is troubling to me that as usual, we see economic issues
: take precedence over safety.

: Rolf

It seems to me that some technique that can be simply and cheaply applied, capable of detecting the larger cracks could be an option for testing here.
Thius makes me think back 30 or 40 years to the old art of WHEEL TAPPING or hitting the wheel with a hammer and listening to the ring of it. It was originally used to detect cracks in the wheels, but modern and better techniques appear to have taken this task over.
I gather that it is still used in Austria for testing that brakes have released from Freight Cars.

Are these modern techniques viable and sufficiently reliable for periodic crack testing of wheels for high speed applications ?
The suggestion is that they are not !!

There is a company in Germany that currently manufactures and markets a high tech version of the WHEEL TAPPING technique. This is RTE from Pfinztal/Karlsruhe .
They market an instrument ProCon 9000. This should be able to simply detect larger cracks (about 2 seconds test time) in wheels.
I am also the UK rep for this company.

Regards,

Godfrey Hands


 
06:24 Jun-09-1998
Tom
Re: Nondestructive testing of railroad wheels at the German Bahn AG Hello all,

I think that maybe 4 years ago there was a request for design suggestions on a
wheel rim inspection system for rail car wheels in the USA. I believe the
specification required that the system be capable of performing the inspection
on a slow "roll by" of the rail car (2-3 miles per hour). I don't remember if
this was in Mat'l Eval or in an NTIAC newsletter.

Casually, at the time, I assumed that an electromagnetic test would be the
likely method for this application. Apparently UT is more common, at least in
Germany. Does anyone remeber the request for quotation that was published a
few years ago? If so, do they know who secured the contract and what method
they utilized for the inspection? I have no vested interest in this, I am
merely curious.

Tom




 
07:52 Jun-09-1998
Ingolf Hertlin
Re: Nondestructive testing of railroad wheels at the German Bahn AG Ultrasonic inspection seems to be talk of the town following the terrible railway accident last week. This forum is an ideal place to continue the discussion of ultrasonic inspection technology (and limitations – see the contribution of Rolf Diederichs and the article from H. Hintze) under experts, but remains in the US area. The following questions and additional remarks may be provocative in this area and on this forum, but may be essential in the general NDT field:
1. Does anybody know whether resonant inspection testing technology (in sound and / or ultrasound range) was considered and tested as an alternative acoustic - and very effective - NDT technology for inspection purposes?
We have reached very good results in the last two years with sound resonant technology in industrial applications for acoustic material testing for:
- crack detection (internal and external cracks),
- structure / texture modifications (in casting),
- hardness,
- inclusions,
- composite material

The principle is simple: hit (or stimulate for resonant testing in the US range) the probe definitely, measure the resonant sound, calculate the characteristics in real time (in fractions of a second with digital signal processors) and compare them with characteristics from ok (reference) probes (this is a simple learning procedure).

The advantages are:
a) Small material changes cause changes in the resonant behaviour of the object: You get an answer from the whole body and not from local fractions. It is his language!
b) Easy to perform and fast! 100 % testing is no problem.
c) Objective and reliable if the reference pieces are defined (e. g. by other methods like US, x-ray and so on)
d) Inexpensive in comparison to other technologies.

The disadvantages are:
a) small surface defects can not be recognized,
b) the reference data must be confirmed by other methods,
c) it is true that it is older than US technology but has up to now limited dissemination because it is considered as a human subjective assessment than as an objective procedure,
d) characteristics can not be gained automatically but only in conjunction with a (human based) signal analysis. But we have developed robust procedures which allow a reliable air-borne signal analysis in an extremely noisy industrial environment (more than 90 dBA)
2. Are there made any investigations or are there latest findings how fast a beginning damage of railroad wheels under continuous stress may lead to dangerous conditions? Or to say it with other words: is it really sufficient to inspect the wheels all some thousand kilometers or is it (in addition) not necessary to install an online system for fast monitoring and decision?
I could imagine that a small defect at the surface can rapidly work out to a severe damage under continuous and impulsive stress conditions. This would mean that offline inspection methods are useful but must be combined with online systems.
RTE Akustik + Prueftechnik GmbH in Pfinztal near Karlsruhe has recently developed a small DSP based box that could be used in vehicles like coaches, cars, lorries and industrial machines to perform online evaluation of sound and vibration signal analysis. What is needed is only the software and characteristics for the special application for online wheel inspection (not limited to trains!).
3. Has anybody heard of investigations and research work on the correlation between the overheating of an iron wheel and the noise it generates under these conditions? Or again in other words: isn’t it worth to extend the discussion to fast online material testing? We should not forget some very terrible accidents with tankers and other lorries on the motorways and in cities!
All comments are welcome!
Ingolf Hertlin



 
00:18 Jun-11-1998
Rainer Link
Re: Nondestructive testing of railroad wheels at the German Bahn AG Some of the contributions to the discussion within this forum on the tragic ICE accident in Eschede to my opinion are not appropriate to scientific technical discussion.

Wordings like
"Personally, it is troubling to me that as usual, we see economic issues take precedence over safety"
"Should we travel on an ICE before such a system is installed"
"Why was safety not accorded a higher priority"

I am really disappointed on such statements.

At a time when it is not clear, what the final cause of the rupture of the wheels was, at a time when the type of defect which might have caused it, is not known, at a time when it is not known, if it was a crack, and how it developed. Was it already during the manufacturing process. What was the time to develop to a critical defect? There are a lot of questions to be claryfied before adress might be given to the NDT methods and techniques to be employed.

In the moment, nobody really can tell, if there is a NDT method, which could in time be used to detect a defect, which still has to be defined.

In the moment a direct cause of the rupture of the defect leading to this tragical accident and the safety concept of German Rail (DB) cannot be deducted.

There is no question, that we have to discuss the technical problems of detecting cracks in railway wheels of the kind of the ICE. And there is no question, that, after clear definition of the cause of the accident, NDT techniques to be employed to prevent it have to be evaluated.

But it is not ethical nor helpful to discuss it in a way as described above.

This is below my understanding of scientific or technical statements or discussions.

Rainer Link
German Society for Nondestructive Testing



 
03:39 Jun-11-1998
Manuel Haces
Re: Nondestructive testing of railroad wheels at the German Bahn AG I do agree complety with Mr. Rainer Link, of the German
Society for Nondestructive Testing. I belive that it is
necessary to complete the investigation of why did fail
the wheel, and after that to discuss in a technical way.
If there was economic reason or other reasons, probably
is competence of other kind of forum. This is my point
of view.




 
01:21 Jun-12-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG First of all thanks to Rainer Link from the German Society for Nondestructive
Testing for using this Forum, by means of holding public discussions by this
new way of communications.

Somehow this kind of communication is a Chat (with a delay).
Often we write down thoughts which may not really be like usually
written statements, we need to get use to it.
Of course we talk seriously here, but like in a panel discussion
words can be said much faster - and that is good in it.

It seems that most of Rainer Link's concerns have been addressed to me.
Still I stand behind my opinion - even now when I have more confidence after
some facts came up during the last days.

In principle Rainer Link's message criticized that our discussion take place
to early, based on doubts that the reason of the broken wheel would yet not
be clear enough.

Fact is that even the involved experts from the Fraunhofer IZFP and
Deutsche Bahn pointed out in some publications on this server that
'UT applied in short inspection intervals on the wheel is absolutely
needed' also 'visual methods are not reliable enough'.

If final investigations would blame other causes than a bad applied NDT
concept, obviously we would return to daily business as usually,
again ignoring the knowledge of the high risk.
NDT concepts should be discussed before an accident happens and not afterwards.

The kind of media relations of the Deutsch Bahn is also a very important reason
why we need an early discussion. DB seemed happy being blamed with another
guiltiness on Wednesday (sensor to detect a lost wheel). They were happy to
bring the public and journalists some 'food to eat', successfully in covering
the real serious mismanagement, also the world football games coming at the
right time. I am also very seriously concerned about the power of the DB in
controlling many media channels and other institutions here in Germany.

When everybody of us stays at home, we may see TV, then we'll criticize often
the corruption and lobbyism of this society.
But we must watch our chance not to do the same thing we blame other for.

Anyway, thanks again for bringing up the issue of ethics in NDT.

Rolf Diederichs
NDTnet

BTW: Our RealAudio speech is ready.
It runs 3 min about Deutsche Bahn and NDT responsibilities.
On Friday we'll add some slides running synchronized.
(Currently only in German)
I hope that this will avoid TV channels in my house.




 
03:12 Jun-14-1998

Dr. Ulrich Mletzko

R & D, Retired
Germany,
Joined Nov 1998
89
Re: Nondestructive testing of railroad wheels at the German Bahn AG I'm seconding very strongly the contribution and
opinion of Rainer Ling from DGZfP.

Since several days, I'm following with more and more
irritation the discussion on this difficult subject.

On one side it's clear that the facts of such a
terrible crash have to be discussed by the experts
from the point of view of technique and science.

On the other side it is evident that the loss of 100
persons will lead to court cases. And, without doubt,
some parts of the German NDT scene will be participants
of that cases, either as experts to the court or as
experts for the plaintiffs or (hopefully not) as
defendants. Therefore, I'm very unhappy on the fact
that the subject is discussed in such a way by the
experts at a time, were we really don't know all the
facts.

Especially, we should avoid to go beyound the facts
in our discussions.

Best regards

Uli Mletzko
NDT Group,
State Materials Testing Institute (MPA),
University of Stuttgart, Germany

P.S. This is my private opinion and not, in any case,
the opinion of my employer.


 
08:45 Jun-15-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG : I'm seconding very strongly the contribution and
: opinion of Rainer Ling from DGZfP.

: Since several days, I'm following with more and more
: irritation the discussion on this difficult subject.

: On one side it's clear that the facts of such a
: terrible crash have to be discussed by the experts
: from the point of view of technique and science.

: On the other side it is evident that the loss of 100
: persons will lead to court cases. And, without doubt,
: some parts of the German NDT scene will be participants
: of that cases, either as experts to the court or as
: experts for the plaintiffs or (hopefully not) as
: defendants. Therefore, I'm very unhappy on the fact
: that the subject is discussed in such a way by the
: experts at a time, were we really don't know all the
: facts.

: Especially, we should avoid to go beyound the facts
: in our discussions.

: Best regards

: Uli Mletzko
: NDT Group,
: State Materials Testing Institute (MPA),
: University of Stuttgart, Germany

: P.S. This is my private opinion and not, in any case,
: the opinion of my employer.


Our main discussion should not be on WHO shall use this discussion forum and WHAT in particular is allowed to discuss here.

The forum is open and we shall not control this - anybody's opinion is welcome (But no advertising please).
Also it may happen that some people cannot find a publisher for bringing up their topics to public/colleague awareness - Internet is information democracy - and this forum is just a part of it.

If a court matter coming up here, that should not be done to judge or hunt guilty people of this particular accident. However, for responsible NDT manager it can be an important matter which may need further discussions to proceed with a learning process.

Experts who are directly involved in the investigations of the broken wheel should choose carefully what they may say here. But it does not means that they shall not participate at all. Readers of this forum which are part of the NDT community, would like to receive information, learn how to choose a NDT method
for safety parts, how to consider POD, and how to apply NDT on safety parts.

Rolf Diederichs


 
07:25 Jun-18-1998
Ed Blessman
Re: Nondestructive testing of railroad wheels at the German Bahn AG : Hello all,

: I think that maybe 4 years ago there was a request for design suggestions on a
: wheel rim inspection system for rail car wheels in the USA. I believe the
: specification required that the system be capable of performing the inspection
: on a slow "roll by" of the rail car (2-3 miles per hour). I don't remember if
: this was in Mat'l Eval or in an NTIAC newsletter.

: Casually, at the time, I assumed that an electromagnetic test would be the
: likely method for this application. Apparently UT is more common, at least in
: Germany. Does anyone remeber the request for quotation that was published a
: few years ago? If so, do they know who secured the contract and what method
: they utilized for the inspection? I have no vested interest in this, I am
: merely curious.

: Tom


There used to be an annual conference on magnetics at IIT in Chicago. One
of the sessions delt with NDT and there were many presentations on rail
wheel testing. Some of the systems have been field tested. I don't know
the results. I have been out of the field for many years, but perhaps there
is still activity in this area at IIT. Ed B.

--------
Cross Information from the NDT Newsgroup
To: nde@coqui.ccf.swri.edu
Date: Wed, 17 Jun 1998 22:10:11 -0500
From: "Ed Blessman"
Organization: ioNET Inc.
Subject: Re: Nondestructive testing of railroad wheels at the German Bahn AG
Sender: owner-nde@coqui.ccf.swri.edu



 
00:18 Jun-24-1998

Todd Snyder

R & D, Metallurgy
Union Pacific Railroad,
USA,
Joined Nov 1998
2
Re: Nondestructive testing of railroad wheels at the German Bahn AG I am a metallurgist/mechanical engineer and I work
for Union Pacific Railroad. I have spent the last
six months analyzing cracks in railroad wheels using
ultrasonic,magnetic, acoustic, and destructive
metallurgical techniques.

At the present time, most of my efforts have been
spent dealing with what we call shattered rims.

This is where a subsurface, internal fatigue crack
grows until a big chunk of the wheel flies off.
This sounds very similar to the failures experienced
by the high speed trains in the UK and Germany,
but due to the lack of metallurgical technical
information published, I am unable to discern what
type of failure has occurred. Is this information
available? Most news agencies say "it broke".

We have had about twelve (12) of these failures which
have resulted in derailments within the last year. We
currently have a manual ultrasonic inspection
procedure in place for wheels in service made
by certain manufacturers because of their high
incidence of this type of failure. Its a very simple
inspection.

As information, I'd like to explain that there are many
types of cracking which occur in these railraod wheels.
THe thermo-mechanical shelling and spalling (from
martensite formation) are relatively harmless. The
thermal cracks(caused by a change in the residual
stresses) and shattered rims(internal fatigue cracks)
are the catastrophe makers. Thermal cracks are usually
oriented transverse to the rolling direction on the
tread of the wheel and extend toward the axis.
Internal cracks of the shattered rim type failures are
oriented parallel to the tread surface and grow
diametrically until the crack penetrates the rim face.



 
04:22 Jun-24-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG : I am a metallurgist/mechanical engineer and I work
: for Union Pacific Railroad. I have spent the last
: six months analyzing cracks in railroad wheels using
: ultrasonic,magnetic, acoustic, and destructive
: metallurgical techniques.

: At the present time, most of my efforts have been
: spent dealing with what we call shattered rims.

: This is where a subsurface, internal fatigue crack
: grows until a big chunk of the wheel flies off.
: This sounds very similar to the failures experienced
: by the high speed trains in the UK and Germany,
: but due to the lack of metallurgical technical
: information published, I am unable to discern what
: type of failure has occurred. Is this information
: available? Most news agencies say "it broke".

: We have had about twelve (12) of these failures which
: have resulted in derailments within the last year. We
: currently have a manual ultrasonic inspection
: procedure in place for wheels in service made
: by certain manufacturers because of their high
: incidence of this type of failure. Its a very simple
: inspection.

: As information, I'd like to explain that there are many
: types of cracking which occur in these railraod wheels.
: THe thermo-mechanical shelling and spalling (from
: martensite formation) are relatively harmless. The
: thermal cracks(caused by a change in the residual
: stresses) and shattered rims(internal fatigue cracks)
: are the catastrophe makers. Thermal cracks are usually
: oriented transverse to the rolling direction on the
: tread of the wheel and extend toward the axis.
: Internal cracks of the shattered rim type failures are
: oriented parallel to the tread surface and grow
: diametrically until the crack penetrates the rim face.
-----------

Thanks for the contribution by Todd Snyder of Union Pacific Railroad.
It seems that Todd needs more room to explain his expertise of railroad
wheel inspection than we can provide herewith a more or less plain text.
That means if we would get the full information, obviously including pictures,
we could provide it as an article. All experts, especially the
German railroad scene could access to this material easily.

Anyway, meanwhile let us here continue the discussion.
Todd Snyder is right, yet it's not clearly published what kind of failure
caused the breaking of the wheel ring.
The media reports that a crack from inside the wheel caused it,
however we should not relay on it since it could be just media politics
from the Deutsche Bahn as well as strategies on justice matter.

There exist only one officially press release of the transport minister
and the German railroad authority - Eisenbahnbundesamt (EBA), see it at
http://www.ndt.net/news/1998/bahn192.htm (in German).

This press release contains no statement on the type of crack which may caused
the breaking. 'Only the Deutsch Bahn spread the speculation of an inner crack.
But so far they can have their knowledge only from photos'
said Mr. Wille from the EBA. The broken material was locked and only
the Fraunhofer Institut in Darmstadt (address below) received the material
for crack evaluations a few days ago.

I have some questions to Todd Snyder.
Did your work based on cracks of mono-block wheels?
Thermal cracks usually occur strong on those type of wheels.
What was the principle UT set up? Is it now a frequently applied test?
As I mentioned already, it may be better to publish more details
in an article on NDT.net - can it be in the next issue already?

Find below some contacts which may be interested in shearing your results.

Rolf Diederichs

-----------------
Dietrich Flade
Press and Public Relations
Fraunhofer Institute for Strength of Structure under Operational Condition
(Fraunhofer Institut für Betriebsfestigkeit in Darmstadt)
Phone +49 (0) 61 51 / 7 05 - 2 67
Fax +49 (0) 61 51 / 7 05 - 2 14
http://www.lbf.fhg.de/home.htm

Hartmut Hintze
Head of NDT at Deutsche Bahn in Kirchmöser
Phone +49 (0) 33 81 / 8 12 - 3 12

Mark Wille
Press Speaker of the Eisenbahn Bundesamt in Bonn (EBA)
Phone +49 (0) 22 8 / 98 26 - 1 86
FAX +49 (0) 22 8 / 98 26 - 1 99

Eckhardt Schneider
Expert in Ultrasonic Evaluation of Stresses in the Rims of
Railroad Wheels.
Fraunhofer Institut Zerstörungsfreie Prüfverfahren (IZFP)
Saarbrücken, Germany
University Bldg 37, D-66123 Saarbrücken
Phone: +49 681 302 3840
Fax: +49 681 39580
Email: schneider@izfp.fhg.de
http://mm.fhg.de/depts/izfp-e.html
http://www.ndt.net/article/ecndt98/rail/306/306.htm

Massimo GORI
Expert in Ultrasonic Evaluation of Stresses in the Rims of
Railroad Wheels.
CISE SpA - Segrate (MI) - Italy
Email: 0852gori@s1.cise.it
(His ECNDT article coming next month on NDT.net)



 
01:13 Jun-24-1998

Todd Snyder

R & D, Metallurgy
Union Pacific Railroad,
USA,
Joined Nov 1998
2
Re: Nondestructive testing of railroad wheels at the German Bahn AG All work that I have done has been with what you call
mono-block wheels, cast steel and forged steel. Yes,
thermal cracks are very detrimental to these types of
wheels. Thermal cracks are also fairly common in the
freight cars I deal with because the brake shoes are
applied directly to the tread surface. On high speed
trains the brakes are applied to other components
which are designed to have brakes applied to them- like
brake disks. Hence, thermal cracking should not be a
problem on high speed trains, unless chain sliding
occurs. Chain sliding is a term used to describe
what happens when the brakes are sticky and the wheels
turn at a diffenerent speed with respect to the rail.
This also happens when the wheels are skewed or offset
from the center of the track.

BUT, we are NOT talking about thermal cracks.
We are talking about shattered rims or internal
cracks or subsurface fatigue cracks. From some
e-mail responses I got, it is clear that Idid
not explain this type of crack well enough.

A shattered rim starts from some internal defect
such as an inclusion whose depth is usually between
1/4-inch and 1-inch from the tread surface. In the
earliest stage of growth, the crack consists of
two waves-one growing in the rolling direction
and one growing opposite. Once the crack is larger
than the inclusion, the two waves combine at the tails
to form an elliptically shaped crack. At this point
you have a crack which is parallel to the tread
surface. This crack will grow like a ripple in a pond
and will always be parallel to the tread surface. This
makes it very easy to detect !!!!! Using a single or
dual element pulse/echo transducer coupled to the tread
surface, it is very easy to detect cracks with a diam-
eter less than 1.0 mm. The fatigue crack grows and
leaves a bulls-eye pattern as evidence of its ripple-
like growth. This bulls-eye pattern becomes elongated
forward and back inthe direction of rolling. When
the crack grows large enough, it breaks through the
surface either on the rim face or it grows right
through the tread surface. Either way, the crack is
so large now that the centrifugal force may be enough
to cause the wheel to rupture and all the mass above
the bulls-eye crack will fly off.

This problem is of course related to the cleanliness
of the steel (the number of inclusions present).
New forged wheels which have several detectable
inclusions pose the highest risks. We scrap
wheels which contain a defect 1/8-inch in diameter
or a wheel with four detectable inclusions between
1/4 and 1-1/4 inches from the tread surface. On a
train carrying people, I'd not allow any detectable
imperfections.

Our UT technique consists of inspecting 100% of the
tread surface using a zig-zag motion from inner to
outer rim face and traversing around the wheel.
So far the test has been used on 2500 wheels. 20%
failed the test. There are 30000 of these wheels
that we consider high risk due to their uncleanliness.
Most wheels are scrapped without testing, but wheel
supply is low in the US and companies with thousands
of these suspect wheels must UT them or quit shipping
goods by train.




 
01:07 Jun-26-1998

Rolf D.

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG id at 5 p.m. Thursday.
Four cars overturned and one was leaning, spilling about 500 tons of coal, the
spokesman said. .........there is about $300,000 in damages.
----
Also this server reports from more accidents:
http://bst-tsb.gc.ca/raillist.html
Railway Occurrence Reports by the Transportation Safety Board.
For example: .....inadequate inspection practices/equipment to detect slid-flat
wheels; and the lack of emphasis placed on safety inspections of passenger
train brakes. ........
----
Do we have Railway Occurrence Reports in Germany?

There might be arguments like 'just freight trains', but this shall not be
excusable. An accident of freight trains can lead to also to a
catastrophe. We must consider that dangerous goods are loaded.
Another problem is when personal trains and freight trains using the same railroads.

Rolf Diederichs




 
03:00 Jul-01-1998
Jürgen Rohmann
Re: Nondestructive testing of railroad wheels at the German Bahn AG We are a small company (50 employees in Germany) and have been working with eddy-current technology for more than 20 years, mainly in aircraft maintenance.

I was not the only one who was shocked to find out that the ICE-disaster was caused by a broken tire. According to the information available at this time, the tire broke because of a fatigue crack which originated on the inside and had reached approximately 60 % of the cross section.

Between 1993 and 1996 we worked on a project which used the magnetic inductive method to detect deep cracks(more than 4 mm below the running surface) which originated at the running surface, but were not connected to the running surface any more. Finally, with the means available to us, we were able to achieve a depth of penetration (drilled holes with a diameter of 2 mm and filled with ferrous bolts) of up to 20 mm below the running surface. For more details, please visit our home page at http://www.rohmann.de.

Unfortunately, we did not get a chance to prove this principle on natural fatigue cracks. However, now that things look different here in Germany, we hope to receive wheels with fatigue cracks to continue this project.

Therefore, I would like to pose the following question: Which type of "magnetic technique" was used at Union Pacific Railroad? We would appreciate a detailed account which could help us with continuing our project.

Thank you very much in advance.
Yours sincerely
Jürgen Rohmann
Rohmann GmbH


 
02:47 Sep-03-1998
Shefford P. Baker
Re: Nondestructive testing of railroad wheels at the German Bahn AG TO the ndt commnunity

Greetings. I am Professor in the Department of Materials Science and
Engineering at Cornell University. I am currently preparing lecture
notes for an undergraduate course on mechancial properties of materials
and would like to include technical information regarding the recent ICE
disaster as motivational material. I have followed commentary in this panel
discussion. If anyone knows where I could find sketches of the
design of the broken wheel and a technical failure analysis of the cause
of the accident, I would be most appreciative. I realize that the failure
analysis may still be under way. Still if some facts are known, it would
be helpful. I am reasonably fluent in German, particularly
technical German and will be happy for any information or connections
that you can send.

Sincerely,

Shefford Baker



 
06:33 Sep-04-1998

Rolf Diederichs

Director, Editor, Publisher, Internet, PHP MySQL
NDT.net,
Germany,
Joined Nov 1998
600
Re: Nondestructive testing of railroad wheels at the German Bahn AG
zoom image[zoom] crack hypotheses

According to news (Bahn AG Press releases) it seems that a crack starting
from inside of thewheel rim caused the breaking of the wheel. But even the
German Bahn Authority (EBA) says that this is still speculative until a formal
investigation from neutral experts confirms this.
Those ICE wheels running under extreme load conditions. We read that a wheel
of 90 cm diameter turns 25 times per sequent,
that means in 24 hours 2160000 load cycles.

A sketch of the wheel together with a crack hypotheses see on top of this page
which shows cracking caused by fatigue and weariness.
(Email list users please access this message on Forum Web)

It must mentioned that for an in-service inspection at a safety relevant component
all kind of possible risks from cracking damages have to be assumed no matter where
and how they are generated.

Please contact me if you like to get Phone/addresses of EBA, Deutsche Bahn etc.

Rolf Diederichs



 
04:07 Aug-03-2002
esmaeil
ut defects : : : hi sir/mis
we have an important problem which there is so many UT DEFECTS in our railroad wheels by B2 mark in UIC STANDARD THAT is producing in E.A.F & V.O.D.C konvertor(12ton).we will be thanks if you can help us in this problem.





 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|