Trains.com

Norfolk Southern Reports Loose Wheels on Additional Railcars

5520 views
38 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    April 2007
  • 4,557 posts
Norfolk Southern Reports Loose Wheels on Additional Railcars
Posted by Convicted One on Thursday, March 9, 2023 11:18 PM

https://www.cbsnews.com/news/east-palestine-toxic-derailment-railcars-loose-wheels-discovered-norfolk-southern-says/

 

What I find interesting is that  (evidently)  there is no procedure in place specifically tuned towards detecting such defects in newly acquired cars.    How long should a new car go before a "fitness" test is performed?

  • Member since
    December 2001
  • From: Northern New York
  • 25,020 posts
Posted by tree68 on Friday, March 10, 2023 7:22 AM

Without knowing how "loose wheels" present, I wouldn't know how such an inspection would be done.  I'm sure there's a procedure in the works, though.

This does sound like a QC issue with the manufacturer, though. Presumably that's being addressed as well.

Rather like checking the torque on your lug nuts after a visit to the mechanic.  Or a recall.

As the old saw goes, if you don't know what you're looking for, you won't know when you've found it."  This may apply here.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    September 2017
  • 5,636 posts
Posted by charlie hebdo on Friday, March 10, 2023 7:33 AM

Sounds like a systems problem where the big picture (operating safe equipment) gets lost in a sea of technical details and excuses.

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Friday, March 10, 2023 7:59 AM

charlie hebdo
Sounds like a systems problem where the big picture (operating safe equipment) gets lost in a sea of technical details and excuses.

I could be wrong - but I suspect that the railcar manufacturers BUY completely assembled wheel sets, if not wheel sets AND truck parts from suppliers.

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Friday, March 10, 2023 7:59 AM
That news report makes it sound as though the loose wheels may have caused the East Palestine wreck.  From the article:
 
Norfolk Southern — the railway company whose train derailed last month in East Palestine, Ohio, contaminating the surrounding area with toxic chemicals — announced Thursday night that it had determined that some of its railcars, of a specific make and model, had loose wheels.
 
During its cleanup of the derailment site, Norfolk Southern investigators discovered that a "specific model and series of railcars had loose wheels," the company said in a news release Thursday night, calling the discovery "an urgent safety issue."
 
However, other reports say that loose wheels caused the Springfield wreck.  In that derailment, video shows new coil cars initiating a derailment while passing over a grade crossing at speed.  Reports cite the loose wheel problem as affecting the new coil cars.  The report also seems to be saying that it was those coil cars shown in the derailment video that caused the derailment due to having loose wheels. 
 
I wonder what that loose wheel condition actually amounts to.  The wheels are press fitted onto the axles with what is called an “interference fit.”  The holes in the wheels are smaller than the outside diameter of the axles.  That fit can fail manufacturing final inspection if either wheel or axle diameter is out of spec.  If the condition is far enough out of spec, the wheel can either turn or slide on the axle.  This might lead to a visual indication of the problem during an inspection.  But in other cases, the fit may be too “loose,” but still tight enough to keep the axle and wheel bound together and appear to be perfectly okay. 
 
As far as I know, the only way to test the press fit is to measure the force it takes to pull it apart.  I have never heard of railroads performing such a test, but they may.  Ideally, you would have a test fixture that applied a threshold amount of pull on the fit without it slipping, and that would be deemed good enough.  In any case this press fit is a very technical detail that has to be very finely controlled to be not too tight, or not too loose.  Wheel presses are common in shops.  They can press a wheel off of an axle, or press a wheel onto an axle.
 
I have never heard that this press fit can loosen up over time just by normal operation.  If that it deemed possible, then the fit should be inspected by testing on a periodic basis.  In any case, this current loose wheel problem is on brand new rolling stock, so it must be a manufacturing defect.  
 
Pressing railcar wheels onto axles during manufacture:
 
 
  • Member since
    March 2023
  • 11 posts
Posted by Reading467 on Friday, March 10, 2023 8:19 AM

https://www.assemblymag.com/articles/97033-press-fitting-railcar-wheelsets

According to this article, which is a summary of a larger research paper, if the interference fit is too great between wheel and axle, the wheel can rotate on the axle under normal service braking action.  The paper includes graphic info that was cut from the journal article.   

I'm not sure what research has been done by the FRA/AAR on this topic. 

  • Member since
    April 2007
  • 4,557 posts
Posted by Convicted One on Friday, March 10, 2023 9:01 AM

Euclid
However, other reports say that loose wheels caused the Springfield wreck.  In that derailment, video shows new coil cars initiating a derailment while passing over a grade crossing at speed.  Reports cite the loose wheel problem as affecting the new coil cars.  The report also seems to be saying that it was those coil cars shown in the derailment video that caused the derailment due to having loose wheels.    I wonder what that loose wheel condition actually amounts to.  The wheels are press fitted onto the axles with what is called an “interference fit.”  The holes in the wheels are smaller than the outside diameter of the axles.  That fit can fail manufacturing final inspection if either wheel or axle diameter is out of spec.  If the condition is far enough out of spec, the wheel can either turn or slide on the axle.  This might lead to a visual indication of the problem during an inspection.  But in other cases, the fit may be too “loose,” but still tight enough to keep the axle and wheel bound together and appear to be perfectly okay. 

I agree with you that the subject of this thread likely pertains more to Springfield, than New Palestine. 

Still, NS must be making some sort of field determination that the wheels are loose. I doubt if they are using lab equipment to measure "pull-off" forces at the crash site.

The take away here for me  is how the cars can be put into service without adequate inspection in the first place?  I guess one possible good to come out of East Palestine is that it is forcing people to take a closer look at a lot of things?

  • Member since
    April 2007
  • 4,557 posts
Posted by Convicted One on Friday, March 10, 2023 9:13 AM

Other sources are stating that this particular defect can result in "out of guage" derailments, so perhaps that explains how the field determination is being made (in lieu of "pull-off" tests)

https://abcnews.go.com/US/defective-loose-wheels-risk-derailment-discovered-norfolk-southern/story?id=97761704

  • Member since
    April 2007
  • 4,557 posts
Posted by Convicted One on Friday, March 10, 2023 9:22 AM

Complete speculation, but I recall from my own experiences having to replace bearings that had spent 30 years "mated" to a shaft, that were quite stubborn coming off the shaft.

You don't want to risk physical damage to the shaft, so anything beyond a few hammer taps is out of the question.  But what I found effective is the careful application of heat to the siezed bearing, causing it to expand ever so slightly.   Perhaps similar forces are at work against these train wheels? 

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Friday, March 10, 2023 9:46 AM
I assume that out-of-gage wheelsets could be found by detectors, but I do not know if that is a practice or if such detectors have been perfected.  An out of gage defect does show that the press fit is out of spec.  But a wheelset that passes the gage test can still have an out of spec press fit.  And the out of spec press fit could be right on the verge of, and progressing to an out of gage condition, but still pass the out of gage test.  Then with that little of margin, the developing press fit defect to drift into an out of gage problem say a week later.  But with an inspection having just been made one week earlier, anther inspection may not be due in time to catch the progressing out of gage condition.
 

The responsibility is on the wheelset manufacturer who will inspect the technical intricacies of the actual press fit.  I do not know if this fit is expected to loosen over time and use.  The same question would apply to testing for developing cracks in truck frames, bolsters, etc.  But with a certain level of robustness and metaugical perfection, they may just assume the parts are built to last for their lifetime of use.  

I assume the loose wheel condition found by NS was done by just visual inspection in cleaning up the derailment.  The first indication of being out of gage would present the inner wheel hub being too far from the shoulder of the the axle.  There would be a gap between the wheel hub and the axle shoulder.  

The inspection of the wheelsets should have been warranted by the manufacturer.  

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Friday, March 10, 2023 10:05 AM

Euclid
I assume that out-of-gage wheelsets could be found by detectors, but I do not know if that is a practice or if such detectors have been perfected.  An out of gage defect does show that the press fit is out of spec.  But a wheelset that passes the gage test can still have an out of spec press fit.  And the out of spec press fit could be right on the verge of, and progressing to an out of gage condition, but still pass the out of gage test.  Then with that little of margin, the developing press fit defect to drift into an out of gage problem say a week later.  But with an inspection having just been made one week earlier, anther inspection may not be due in time to catch the progressing out of gage condition.
 

The responsibility is on the wheelset manufacturer who will inspect the technical intricacies of the actual press fit.  I do not know if this fit is expected to loosen over time and use.  The same question would apply to testing for developing cracks in truck frames, bolsters, etc.  But with a certain level of robustness and metaugical perfection, they may just assume the parts are built to last for their lifetime of use.  

I assume the loose wheel condtion found by NS was done by just visual inspection in cleaning up the derailment.  The first indication of being out of gage would present the inner wheel hub being too far from the shoulder of the the axle.  There would be a gap between the wheel hub and the axle shoulder.  

The inspection of the wheelsets sound have been warranted by the manufacturer.  

Derailments that I have personally been involved with, either wreck clearance or cause investigation, from my personal observations is that the wheel/axle combination is rarely 'damaged' (bent or broken).  With that being the case, it becomes almost too easy to see when a wheel was not properly 'set' on its corresponding axle - the wheel will have moved and will leave witness marks on the axle detailing its move from its installed position on the axle to its current position on the axle.

Never too old to have a happy childhood!

              

  • Member since
    May 2004
  • From: Valparaiso, In
  • 5,921 posts
Posted by MP173 on Friday, March 10, 2023 11:45 AM

Today's Wall Street Journal (march 10, page 3) has an article indicating that the American Associationof Railroads "issued a rare advisory calling for certain railcars tobe taken out of service and inspected amid concerns that loose wheels might increase the risk of derailments."

If I recall (not going back to check) the Springfield, Oh derailment featured a coil car derailing.

Ed

 

  • Member since
    July 2010
  • 351 posts
Posted by ns145 on Friday, March 10, 2023 12:00 PM
  • Member since
    February 2018
  • 299 posts
Posted by adkrr64 on Friday, March 10, 2023 12:51 PM

You picked a fine time to leave me loose wheel.....

  • Member since
    April 2007
  • From: Bridgman, MI
  • 283 posts
Posted by bogie_engineer on Friday, March 10, 2023 1:08 PM

Wheels are mounted by simply pressing them on to the axle. There is a very specific surface finish, and interference range required along with a measure of bore and seat taper. The pressing operation requires a lubricant meeting AAR specs and the tonnage to apply is recorded on a chart during the pressing operation. The tonnage has to increase at a specified slope as the wheel is further engaging the axle. For locomotive wheels, the final tonnage has to be between 95 and 145 tons IIRC. The shop doing the work has to be inspected and approved by AAR showing they can meet all the requirements including the record keeping needed. Axle and wheels are serialized for traceability.

EMD had a problem with wheels moving on the axles on the first batch of SDP40F's for Amtrak on the Santa Fe. After many cab rides, it was found a specific diamond crossing in the SW had bad alignment and the impact at the higher speeds that Amtrak ran was enough to move the wheels slightly. This resulted in increasing the tonnage required to the 95-145 range for the 10.25" bore loco wheels, prior, I believe, the minimum was about 75 tons.

Whichever shop mounted these wheels will be under a lot of scrutiny, rightly so.

Dave

  • Member since
    September 2013
  • 2,505 posts
Posted by caldreamer on Friday, March 10, 2023 1:48 PM

I am suprised that the wheels are not welded to the axles which would prevent the wheels from moving in either direction.

  • Member since
    April 2016
  • 1,447 posts
Posted by Shadow the Cats owner on Friday, March 10, 2023 2:50 PM

If you weld something like a wheel to the axle when both are both fairly roughly finished forged pieces except for the mating surface which is machined.  You take a risk of getting a weak spot on the part that could cause a failure in the parts.  So you risk a bigger problem.  Interference fitment works fine 99.9 percent of the time.  

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Friday, March 10, 2023 3:39 PM

caldreamer
I am suprised that the wheels are not welded to the axles which would prevent the wheels from moving in either direction.

The heat of welding would create a lot of distortion to both the axle and the wheel and would require the entire assembly be remachined after welding to get it back to true.  It is unknown to me how the heat stresses of welding would affect the forged metalurgy of both the axles and wheels.

Railroads have used interference fits for any number of connections for approaching two centuries.  Driver tires for steam engines where held on by interference fit - thus using locomotive brakes do control train speed was discouraged as the heat from braking could cause a driver tire to heat up and expand beyond its interference fit.

Never too old to have a happy childhood!

              

  • Member since
    December 2001
  • From: Northern New York
  • 25,020 posts
Posted by tree68 on Friday, March 10, 2023 4:06 PM

BaltACD
Driver tires for steam engines where held on by interference fit...

There are videos on-line showing workers heating steam locomotive tires before installing them on the wheels.  I'm sure it was an every day operation at the major steam maintenance shops.

It's still an interference fit, though.

Any reference bringing East Palestine and Springfield together is simply reaching by a news organization, looking for an angle.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    January 2019
  • 1,686 posts
Posted by Erik_Mag on Saturday, March 11, 2023 12:27 AM

tree68
BaltACD
Driver tires for steam engines where held on by interference fit...

There are videos on-line showing workers heating steam locomotive tires before installing them on the wheels.  I'm sure it was an every day operation at the major steam maintenance shops.

It was also a common operation with some diesel and electric shops where the locomotive wheels were equipped with tires to handle tread/flange wear without having to replace the whole wheel.

Conversely the hinge pins for the B-1's variable sweep wings was pre-cooled in liquid nitrogen prior to being placed in the wing hinge.

  • Member since
    December 2001
  • From: Northern New York
  • 25,020 posts
Posted by tree68 on Saturday, March 11, 2023 6:42 AM

Erik_Mag
Conversely the hinge pins for the B-1's variable sweep wings was pre-cooled in liquid nitrogen prior to being placed in the wing hinge.

I think I saw something along that line recently in a railroad context, too.

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Saturday, March 11, 2023 6:56 AM

tree68
 
Erik_Mag
Conversely the hinge pins for the B-1's variable sweep wings was pre-cooled in liquid nitrogen prior to being placed in the wing hinge. 

I think I saw something along that line recently in a railroad context, too.

Heating and Cooling are both used, on a case by case basis, when it comes to creating interference fits.  In some cases both methods are used - the larger part of the fit gets heated and the smaller part of the fit gets cooled.

Fitting wheel bearings on my race car - I put the aluminum hub in the oven at 500 degrees for a period of time and put the bearing in the freezer for a period of time.  Once time 'expires' the bearing slips right into the hub with no force necessary.  Once the temperatures stabilize at room temperature - the bearing isn't going anywhere.

Never too old to have a happy childhood!

              

  • Member since
    July 2021
  • 10 posts
Posted by NYC-Ohio on Saturday, March 11, 2023 9:29 AM

FWIW - I know that Steelcar assen=mbled their trucks from parts. No idea though about the actual wheelsets.

Also have seen loads of wheels already mounted on axles being shipped by Trinity, Trinty has flatbeds with rails embedded in the deck to haul these. I have loaded other parts on these flatbeds.

Pretty sure that those wheels are made in the Pittsburg area.

  • Member since
    December 2001
  • 1,190 posts
Posted by mvlandsw on Saturday, March 11, 2023 8:52 PM

tree68

Crank pins for the UP Big Boy?

 

 
Erik_Mag
Conversely the hinge pins for the B-1's variable sweep wings was pre-cooled in liquid nitrogen prior to being placed in the wing hinge.

 

I think I saw something along that line recently in a railroad context, too.

 

  • Member since
    May 2003
  • From: US
  • 25,292 posts
Posted by BaltACD on Sunday, March 12, 2023 8:23 AM

mvlandsw
 

 

tree68

Crank pins for the UP Big Boy? 

Erik_Mag
Conversely the hinge pins for the B-1's variable sweep wings was pre-cooled in liquid nitrogen prior to being placed in the wing hinge. 

I think I saw something along that line recently in a railroad context, too. 

Interference fits have been used in manufacturing ever since it was determined that things expand when heated and contract when cooled. Steam engines were a prime instrument of interference fit. The tires that contain the flange on locomotive drivers are heated to expand them so that they can create a tight interference fit when the assembly is cooled to normal temperatures. The interference fit is also why locomotive engineers on steam engines use the independent brake sparingly so the don't heat the tires to a temperature where they expand and have the potential to leave proper alignment on the wheel - it has happened.

Like anything manufactured, mistakes happen, try as the manufacturer might to prevent the mistakes.

Suspect the car builder, buys the assembled wheelsets from a supplier with wheels and bearings already installed and ready to be placed in the truck frames.

Never too old to have a happy childhood!

              

  • Member since
    January 2014
  • 8,221 posts
Posted by Euclid on Sunday, March 12, 2023 9:38 AM

Mistakes are always possible, but in this case, they are never acceptable or able to be dismissed by saying "mistakes happen."  I don't know if anybody watched the video I posted showing the railcar wheelset assembly. 

The idea is that those press fits must not ever fail.  The press fit is subject to a bewildering amount specifications and testing at all levels, including the metalurgy, the press tonnage force, the mirror fine finish on the mating surfaces, the technical lubricant, the ambient temperature, and the interference tolerance down to one ten-thousanth of an inch.  In the video, you can see the the entire process which is fully automatic including the loading of the wheels and axle and the pressing. 

You can also see the automatic calipers measuring the interfering inside and outside diameters before each wheelset pressing.  I suspect there is even more testing that perodically pulls apart an assembly and measures the tonnage of pull resistance.  

But the chance of failure is still an interesting question because anyting is possible.  It would be interesting to know the history of failure due to flawed assmebly of materials as related to railcar wheels, axles, truck frames, bolsters, etc.  I recall that recently, there have been quality problems with wheels that causes them to break.  

  • Member since
    September 2017
  • 5,636 posts
Posted by charlie hebdo on Sunday, March 12, 2023 10:44 AM

tree68
Any reference bringing East Palestine and Springfield together is simply reaching by a news organization, looking for an angle.

The connection is the big picture - rail safety or lack thereof.  That is the angle. As Balt put it, not much of asafety cultre on the NS.

  • Member since
    December 2001
  • From: Northern New York
  • 25,020 posts
Posted by tree68 on Sunday, March 12, 2023 4:15 PM

charlie hebdo
The connection is the big picture - rail safety or lack thereof.  That is the angle. As Balt put it, not much of a safety culture on the NS.

I get that part.  I'm pretty sure, though, I saw it suggested that loose wheels had something to do with East Palestine...

LarryWhistling
Resident Microferroequinologist (at least at my house) 
Everyone goes home; Safety begins with you
My Opinion. Standard Disclaimers Apply. No Expiration Date
Come ride the rails with me!
There's one thing about humility - the moment you think you've got it, you've lost it...

  • Member since
    October 2014
  • 1,139 posts
Posted by Gramp on Sunday, March 12, 2023 5:26 PM

Here's derailment today near Tulsa.  From the video, looks like bulkhead and coil cars were off track going into crossing?

https://www.msn.com/en-us/weather/topstories/that-just-happened-train-derails-and-crashes-near-tulsa/vi-AA18lzsR?ocid=msedgdhp&pc=U531&cvid=edef6e5cc0e14fdab99b0641cb833515&ei=60

  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 11,919 posts
Posted by blue streak 1 on Sunday, March 12, 2023 10:43 PM

Gramp

Here's derailment today near Tulsa.  From the video, looks like bulkhead and coil cars were off track going into crossing?

https://www.msn.com/en-us/weather/topstories/that-just-happened-train-derails-and-crashes-near-tulsa/vi-AA18lzsR?ocid=msedgdhp&pc=U531&cvid=edef6e5cc0e14fdab99b0641cb833515&ei=60

 

 
Was it the bulk head car rear truck jumped up at the grade crossing?  Does that mean the bulk head car was off the rails?
 
Are these press fits on same diameter each for axels and wheels or slightly tapered ?  That difference would have indicated press Kns different during press oerations ? Would it be possible to make common freight wheel sets have a larger diameter inside solid collar that would prevent wheels to go less than inside gauge specs ?
 
Also are the roller bearings fitted so the wheels cannot go gauge + ?

Join our Community!

Our community is FREE to join. To participate you must either login or register for an account.

Search the Community

Newsletter Sign-Up

By signing up you may also receive occasional reader surveys and special offers from Trains magazine.Please view our privacy policy