Bad Wheels

[lechee]

I was north of Bakersfield over the weekend and saw a box car being pulled at 65 miles per hour, the front set of wheels were oscillating back and forth at a very rapid rate, shaking the whole car. I have never seen this before and was wondering if this happens very often and how dangerous is this, can this be picked up by a detector?  

[Modelcar]

Sounds very much like they describe "truck hunting".....If it's violent enough to be shaking the whole car, one would think it's more than what they would want it to be doing.

We'll let the experts comment on any specifics.  One would think it could be causing some track damage and or even potential truck damage and or worse.....

[kolechovski]

That's never a good thing.  Much more, and it will derail.

[wabash1]

Ive seen cars bad enough that both ends was dancing but never derailed. I dont think they will derail. and yes engines do it to.

[lechee]

Thanks, now that I know what it is called, I see many articles written on truck hunting and there are several different types of detectors. There is even a video showing a cement car truck hunting, I watched it several times but could never see the trucks moving back and forth. When you actually see one that is very bad it sure looks like it could climb out of the track and be a big problem for the railroad. . 

[ericsp]

I once saw an acetone tankcar with a hunting truck. 

[Paul_D_North_Jr]

I think there have been some derailments caused by excessive "truck hunting" at speed, esp. where the ties and track fasteners - typically spikes and tieplates, etc. - weren't in the best of condition.  That could lead to the rail twisting from the banging forces being applied at the top, and so the rail would rotate and partially turn over towards the outside, leading to a transient wide-gauge condition and hence causing a derailment.  It would be similar to a rail "roll-over" type derailment, except those typically occur in curves with high lateral forces involved; but truck hunting typically occurs in tangent track (only) because there is no curve to force the truck against one rail or the other for an extended distance or period of time.  I suppose it's possible that hunting and such problems could also be caused by poor "detail" alignment of the track - little wiggles in the rails as you look down it - but that would be really rare and related to some oddball condition, such as curve-worn and bent rail being relaid into a tangent track or similar.

I believe an Amtrak derailment on CSX (ex-NYC) in upper New York State about 5 years ago was determined to have been caused by excessive hunting of one of the trucks of the baggage car.  Also, Amtrak's E60 electric locomotives and SDP40 diesels - both from the 1970s - had tracking issues.  I'm not sure what the technical difference is between "hunting" and "yawing", but I beleive that latter term was used in connection with the E60's problems.  The SDP40's were more complicated, but recently I read (and posted here) something about a determination being made that their derailments were somehow caused by the following baggage car.

I'm not aware of a detector for truck hunting, but in theory there's no reason why one couldn't be developed and installed along the same lines as a wheel impact load detector ("WILD").  Perhaps that's already been done in lab situations - the FAST test track in Colorado comes to mind - but I can't say for sure.  Also, for all of the above, I haven't researched or checked it yet, so don't bet the ranch on it until I get back here with futher details and links, etc.

- Paul North.

[Paul_D_North_Jr]

No luck yet on finding that Amtrak / CSX derailment - maybe I mis-remembered it.

Link to FRA's Truck Hunting Detector Demonstration (note that it refers to the WILD as well):

http://www.fra.dot.gov/us/content/1447

- PDN.

[Paul_D_North_Jr]

Link to video of "UP Train Cement Car Truck Hunting", supposedly at about 10 cars from the end of the train, or about 1:38 into this 1:51 minute video:

http://www.youtube.com/watch?v=n-EfKFxGJc4 

[Paul_D_North_Jr]

"According to Federal Railroad Administration (FRA) statistics, the average annual cost of hunting-induced derailments over the last five years was $2,440,000." [Section 1.0 - EXECUTIVE SUMMARY, Report page 1 (PDF Page 6 of 37), 1st para.; emphasis added - PDN.]  Average number of such derailments per year (from 2000 to 2004) was 6.8 - the range was from 5 to 10 per year [from "TABLE 1 - Truck Hunting Cause Derailments" report page 3 (PDF Page 8 of 37) ].

From:

SYSTEM TO DETECT TRUCK HUNTING ON FREIGHT RAILROADS

Final Report - Safety IDEA Project SAFETY-06

Prepared for Safety IDEA Program

Transportation Research Board - National Research Council

Prepared by: Richard Morgan - Sam Chapman - Joe Brosseau

Transportation Technology Center, Inc., a Subsidiary of the Association of American Railroads

Pueblo, Colorado - February 2006

EDIT - add link:  

 http://www.trb.org/studies/idea/finalreports/safety/S-06_Final_Report.pdf 

This probably has more than you'd ever want to know (technically) about this topic.

- PDN.

[Murphy Siding]

    I saw a similar thing yesterday.  A train had 2 empty gondola cars between blocks of ethanol tank cars.  I figured the tank cars were full, as they rode very steady.  The gondolas were bouncing like a pickup truck on a washboard gravel road.

[Mookie]

Is it always a track problem as opposed to car problem that makes them hunt?

[gopherstate]

Truck huntting is a very serious problem and is being addressed by the AAR and railroads. The current remedy of choice is installing constant contact side bearings on the trucks to help control this violent movement.  It is believed that huntting may be caused by the relationship between wheels and track, but is more of a car problem.  When a truck starts to move back and forth it can reach a point of harmonic resonance which sets up an uncontrollable and  dangerous situation.  There seems to be a relationship between speed, track and car that is not totally understood. 

[Paul_D_North_Jr]

Mookie

Is it always a track problem as opposed to car problem that makes them hunt?

No, it's almost always a car problem (only) - particularly, in the trucks, and more specifically, in the wheels, or their relationship with the truck, and sometimes the rails, as gopherstate noted above.  When hunting occurs, it's usually only 1 or at most only a couple of the trucks in a train.  If it was a track problem, then they'd all be hunting (or most of them).  Also, certain individual cars can be observed and known to be prone to hunting = poor tracking characteristics, when their wheels and trucks become worn to a certain point and configuration - again, not well understood.  Further, in the TRB Report that I referenced previously above, NS used a test train that included 5 cars that were known to be "hunters" - they didn't go look for a piece of track that was known to cause hunting.  See:

 http://www.trb.org/studies/idea/finalreports/safety/S-06_Final_Report.pdf 

at Report page 18 (PDF Page 23 of 37), as follows: [emphasis added - PDN.]

"5.1.2 Test Car Selection

Ten rail vehicles comprised the test train consist: one NS locomotive, one NS research car, one NS box car, two NS covered hopper cars, one NS flat car, three NS gondola cars, and one TTX autorack.  Of these 10 rail vehicles, five were instrumented at each end with accelerometers to measure carbody end lateral accelerations: the NS box car, one NS covered hopper car, the NS flat car, one NS gondola car, and the TTX auto-rack. These five cars were selected as known “hunters” and remained instrumented for the entire test. Data from the cars was collected using the NS research car."

Typically, hunting seems to be caused or affcted the shape of the contour of the tread of the wheel up to the flange, and it's interaction with the rail head.  Other factors are how worn the mating surfaces of the truck bolster and sideframe are - how much "slop" is in them there - and how easily the truck can pivot or turn slightly from side-to-side on the center plate in the middle of the bolster and how worn each of those are, plus such other aspects such as whether the axles are truly parallel, the wheels are truly square to the axle, etc. 

If you've ever had an experience of driving or even looking at an automobile or truck tire that is badly worn from poor alignment - camber, caster, and toe-in/ out, etc. and/ or balancing (which is kind of like the "flat spot" problem in a way) - which show up as scuffing, cupping, bald spots or rings around the tire, etc., then you'll have seen what can happen from these kinds of things.

Hope this is informative.  You're stretching my knowledge envelope pretty well here, so don't be surprised if someone adds to or corrects what I've said, but this is as good as I can do.

- Paul North.

[Mookie]

Paul - I peeked in your envelope and you are doing just fine.  I need work a little harder and come up with more questions that need answers.  The forum has always been very kind to me with explanations.

Thank you!

[CShaveRR]

We were pacing a BNSF freight train outside Galesburg on U.S. 34 a couple of weeks ago. Two PSPX covered hoppers were hunting rather violently, and a couple of other cars to a lesser degree. Kind of alarmed my sisters and brother-in-law. (Trains don't travel that fast where they live.)

Freight cars equipped with constant-contact side bearings were permitted to travel at higher speeds by UP operating rules. I am assuming that empty cars tend to hunt more than the loads, but I may be off base on that one. Anyway, empty cars with constant-contact bearings presented problems in our yard--they tended to bind a little more. In my tower, I keep an ear open for cars that make a distinctive sort of noise when hitting a curve (as every car does, in one place or another, when it goes by me): if I hear it, I let the operators in the lower towers know, and I don't retard the cars--they'll probably stop short anyway.

[Paul_D_North_Jr]

Thanks, Mookie !

Here's a link to a Hunting Truck Detector manufacturer's webpage for it:

http://www.salientsystems.com/prod_hunting.html 

And for the "Product Description" brochure:

http://www.salientsystems.com/htd.pdf

See also the description and photo of various Truck Performance Detectors at 3.3.1 (pages 19 and 20, PDF Pages 25 and 26 of 62) of the July 2007 Canadian Railway Safety Technologies report at:

http://www.tc.gc.ca/tcss/RSA_Review-Examen_LSF/Research/Technologies.pdf 

- PDN.

[edblysard]

Mookie,

In theory, if the wheel profile and tread were in perfect shape, and the rail head was also perfect, along with the elevation and such, the wheels would not even need flanges, all the proper forces would guide the wheel and truck where it was suppoded to go...

Now, nothing in railroading is ever perfect, and if a wheel wears badly enough, or a side bearing wears enough, then that wheel will track differently than the others, as in Pauls example of a badly aligned automobile, it pulls to one side.

This "pull" increases the wear and the problem.

Lots of causes for this, but the result is one side of the truck tries to go a little to the inside or the outside, while the other side tries to follow the rail properly, and the wheel flanges keep nudging it back , then it hunts again, gets corrected...so forth and so on.

Left alone, it will derail the car when the wear cause becomes great enough, either at a switch point or a frog.

It can reach the harmonic that was pointed out, and the entire car will "sway" side to side, adding to the problem...rare, but this harmonic sway can cause a car to go over, although most of the time, it just looks scary.