Trains.com

Wheels too thin for frogs

12181 views
54 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    March 2021
  • 115 posts
Wheels too thin for frogs
Posted by Former Car Maintainer on Friday, May 21, 2021 11:31 PM

Saw this recent article for HART. https://www.staradvertiser.com/2021/05/16/hawaii-news/search-continues-to-fix-oahu-rails-too-narrow-wheels-and-too-wide-tracks/

Can't see what they are talking about in the picture but sure seems like a SNAFU

  • Member since
    January 2014
  • 7,039 posts
Posted by Euclid on Saturday, May 22, 2021 10:38 AM

That is a major screw-up in which the wheelset gage does not match the track gage.  The wheelset gage is too narrow in relation to the track gage, so the backs of the flanges interfere with the guard rail features of the switch frogs.  So, to tiptoe though the interference zones, they will slow to 15-20 mph.  But apparently they will run full speed of 55 mph on the rest of the track. 

In my opinion, they are trying to put a happy face on a disastrous error.  So they are doing this by only slowing down for the track-to-track crossing frogs.  But there is plenty of danger by running on the rest of the track while allowing such excess side to side oscillation due to the mismatched track/flange gage.  

The system should be shut down and completely repaired, rather than to save face by falsely minimizing the problem to just the frog interference.  I find it hard to believe they are doing that. 

  • Member since
    September 2003
  • 17,990 posts
Posted by Overmod on Saturday, May 22, 2021 11:25 AM

I heard Professor Slaby muttering about 'postwar technology' looking at this stupidity.  

Note the trendy high-speed flat-tread wheels, pathetic excuse for flange fillet.  No cone means full excursion to flange contact.  How 'wider wheels' are supposed to fix the stated issue, given the rather clear if distorted-perspective picture "B" provided, is not clear; the prospective effect on self-guarding frogs (a problem for stable hi-rail vehicle design!) is even less clear. 

Prospectively, a little selective taper or 'transition' ground into the engaging wing going into the frogs would (I think) serve to locate and then guide the 'off' flange enough to pull the flat-tread wheelset off what is presumably flange bearing against the point side.  They might have to learn to dress the back side of the flange at wheeltread-turning time.  Such is the price of stupid design and requisition.

Selectively flange-greasing the backs of the flanges is a weird idea, but it would have... I think... relatively little propensity to migrate to tread contact, and would simplify the issues with almost-certain screeching and shuddering as these kludged cars go through the switches...  

I'll be highly interested to read FCM's thoughts on this, as I think he'll have had extensive professional familiarity with some of these issues.

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Saturday, May 22, 2021 4:35 PM

Is the flange too thin or the wheel depth (wheel face) wrong? Couldnt the pointed frog part be widened? How was a non standard frog purchased for a standard track gauge? Are the wheels non standard? Need a lot of dimensional drawings to see the big picture... https://www.researchgate.net/figure/Schematic-representation-of-the-wheel-set-crossing-over-the-frog-view-from-above-2_fig1_319898593

  • Member since
    January 2014
  • 7,039 posts
Posted by Euclid on Saturday, May 22, 2021 5:51 PM

Former Car Maintainer

Is the flange too thin or the wheel depth (wheel face) wrong? Couldnt the pointed frog part be widened? How was a non standard frog purchased for a standard track gauge? Are the wheels non standard? Need a lot of dimensional drawings to see the big picture... https://www.researchgate.net/figure/Schematic-representation-of-the-wheel-set-crossing-over-the-frog-view-from-above-2_fig1_319898593

 

I would not conclude that either the flange or the entire wheel is too narrow.  The gap between the outside of the flange and the inside of the rail appears too wide, especially if the wheelset is centered on the track in the photo.  Also the outer edge of the wheel running on the rail head is abnormal; even if the wheelset is shifted as far left as it will go with this wheelset on this track.  Even in that circumstance, the outer edge of the wheel should be overhanging the outside of the rail. 

So, with what is shown in the photo, I assume that the track gage is as it was intended to be.  I also assume that the wheel profile is as intended, including tread and flange. 

I think the problem is a mistake in specifying the wheel set.  There is a dimensional error on the axles locating the axle shoulders against which the wheel hubs bottom on when being pressed onto the axle. Basically the shoulders are too near the center of the axle.  So the wheelset ends up pressed too far onto the axle, thus resulting in a wheelset gage that is narrower than the track gage, when the two should match. 

One solution would be to replace the axles with the correct shoulder to shoulder dimension, and then remove the wheels from the incorrect axles and reinstall them on the correct axles.  

It also may be possible to remove the wheels from the existing axles, and press on a new spacer bushing on each side that bottoms out on the existing shoulders, and has a width that will properly space the wheels to gage when they are pressed back on and bottom out against the new spacer bushing. 

This overall problem would be the most likely mistake to have been made.  That is to wrongly specify the wheelset gage.  This is because there is no hard line feature on the wheel that represents the gage datum.  So it is the most likely measurement to be misstated or misunderstood. 

Even the way this is being reported is ambiguous and symptomatic of the same ambiguity that probably led to the error in the first place.  The news says the wheels are too thin and the tracks are two wide.  This is gobbledygook.  Furthermore, they seem to only see an issue with the wheels striking the guard rails of the frogs.  That is a problem, but it is not the only problem.  The entire problem also includes the danger of derailment over the entire line because the incorrect wheelset would allow excessive truck hunting, resulting in instability that could cause derailments. 

It would be like the old “Compromise Wheels” used in the 1800s to allow trains to run on multiple gage tracks by using an extra wide wheel tread.  They proved very unstable because when running on the widest gage track, their flanges were too far inward from the rails, so the wheels could wander side to side.  The wandering could gain momentum and force the flange to climb the rail when it suddenly bottomed out hard on the side of the rail. 

  • Member since
    December 2001
  • From: Denver / La Junta
  • 10,297 posts
Posted by mudchicken on Saturday, May 22, 2021 11:36 PM

Not that surprising. Transit vs. common carrier freight railroading.

(FRA rules everywhere and FTA almost none; AREMA vs ASCE ; Federal rules vs State Regulation; proven technology & testing vs. seat of the pants new....and so on)... The "anything goes" transit guys got bitten this time and fortunately nobody got hurt.

Mudchicken Nothing is worth taking the risk of losing a life over. Come home tonight in the same condition that you left home this morning in. Safety begins with ME.... cinscocom-west
  • Member since
    March 2011
  • 59 posts
Posted by dpeltier on Sunday, May 23, 2021 9:31 AM

Folks, in the time you spend writing wild and crazy speculation on the Trains forum, you could have used Google to find the following article:

https://www.civilbeat.org/2021/03/hart-hitachi-at-odds-over-wheel-defect-as-rail-costs-soar/

And you would have seen that the problem is that the wheel flanges are a half inch "too thin" for the flange-bearing frog design, and that the observed symptom was excessive frog wear. Also, the issue only affects trains taking the diverging route.

This still isn't a full explanation, but it rules out most of the malarkey generated so far on this thread.

In the freight world the tolerances for allowable wheel profile - and the reality that out-of-spec wheels can travel some ways before getting caught - mean that you would never design a system that relied heavily on the wheels having a particular flange width. So I can't really envision exactly what the problem is. It's not really clear to me from the article whether this is actually a safety issue (trains could fall off the tracks) or an economic issue (frogs have an unacceptably short life). 

Dan

  • Member since
    May 2003
  • From: US
  • 21,506 posts
Posted by BaltACD on Sunday, May 23, 2021 9:51 AM

dpeltier
Folks, in the time you spend writing wild and crazy speculation on the Trains forum, you could have used Google to find the following article:

https://www.civilbeat.org/2021/03/hart-hitachi-at-odds-over-wheel-defect-as-rail-costs-soar/

And you would have seen that the problem is that the wheel flanges are a half inch "too thin" for the flange-bearing frog design, and that the observed symptom was excessive frog wear. Also, the issue only affects trains taking the diverging route.

This still isn't a full explanation, but it rules out most of the malarkey generated so far on this thread.

In the freight world the tolerances for allowable wheel profile - and the reality that out-of-spec wheels can travel some ways before getting caught - mean that you would never design a system that relied heavily on the wheels having a particular flange width. So I can't really envision exactly what the problem is. It's not really clear to me from the article whether this is actually a safety issue (trains could fall off the tracks) or an economic issue (frogs have an unacceptably short life). 

Dan

My understanding that where freight railroads have frogs bearing the weight of the wheel flange in railroad crossings at grade - that route is done at a slow speed.

Trying to use the frog bearing the weight in a relatively high speed crossover is like presenting a stick of butter to a hot knife.  The impact pressure of the flange against the frog acts like a powered knife over time.

The originally posted 'article' left much to one's imagination.

  • Member since
    March 2011
  • 59 posts
Posted by dpeltier on Sunday, May 23, 2021 10:23 AM

BaltACD

My understanding that where freight railroads have frogs bearing the weight of the wheel flange in railroad crossings at grade - that route is done at a slow speed.

Trying to use the frog bearing the weight in a relatively high speed crossover is like presenting a stick of butter to a hot knife.  The impact pressure of the flange against the frog acts like a powered knife over time.

Probably the most common application of a flange-bearing frog is the "jump frog" or "lift frog", used on a low-speed, low-tonnage route that either diverges from or crosses over a heavier-tonnage route. Any disadvantages or maintenance issues associated with flange bearing are dwarfed by the benefits on the heavier line.

Another usage I know of is in yards, where "hybrid flange- and tread-bearing frogs" are used in some pretty high-tonnage situations. These are basically regular cast frogs with a shallower-than-normal flangeway, so that worn wheels with tall flanges run on the flanges while newer wheels run on the treads. (The worn wheels are the ones that cause the biggest impacts and most damage in a tread-bearing frog.)

Then finally there are some crossing diamonds that are fully flange-bearing in both directions. Development of these was hindered for a long time by FRA regulations, and I don't think they have caught on very widely. The flange-bearing sections due tend to wear quickly. And you do still have some impact load where the wheel transitions from tread-bearing to flange-bearing or vice versa.

But a feature of the hybrid frogs and the fully flange-bearing diamonds is that, if the flangeway wears too deep, the whole thing just reverts to a traditional tread-bearing frog. So it doesn't result in any safety problems.

Dan

  • Member since
    September 2003
  • 17,990 posts
Posted by Overmod on Sunday, May 23, 2021 10:33 AM

dpeltier
the problem is that the wheel flanges are a half inch "too thin" for the flange-bearing frog design, and that the observed symptom was excessive frog wear. Also, the issue only affects trains taking the diverging route.

The problem is that this amounts to more newsworker malarkey, masquerading (poorly) as engineering insight ... and having it delivered in a mocking tone only makes the relative lack of substance worse.

"Half an inch too thin" as a description is one of those things like the old joke in "Meet the Applegates" -- the admonition 'You can't have too much water in a nuclear reactor'.  Any attempt to figure out what Bob Good actually intended (and I am reasonably sure that he does know what the situation is, and understands the physics involved) is lost in wack semantics and, apparently, the usual news desire to find blame for when the "experts" screw the pooch.

Still undocumented is a critical issue involving the "silicone coating" supposedly banged or stripped off key portions of the frog architecture -- it still having been left unexplained why such material was necessary in the first place -- and any indication whether it is continuous or oscillating force causing the damage.

I suggest gently that the next person wishing to debunk this kerfuffle actually find a transcript of Good's actual language -- or actually try to contact him, now that he's been identified, or Prof. Prevedouros -- and find out in actual engineering terms what the heck is actually going on.

And then what they and Hitachi might plan to do about it.

  • Member since
    January 2014
  • 7,039 posts
Posted by Euclid on Sunday, May 23, 2021 11:08 AM

I have googled several articles on this wheel problem, and have yet to find any explanation that does not convey a multiple, conflicting meaning of the problem.  They refer to wheels width being wrong, wheel flanges being too thin, wheel flanges being too narrow, and the track being too wide. One article in the OP, it says that the flanges hit the guardrails of the frogs. 

If that photo showing ½” gap is taken with the wheelset centered on the track, that means that there is 1” of free play side-to-side for the trucks.  As Overmod points out, the flanges seem to be made quite sharp, and lack a fillet radius at their base, and the wheel seems to lack taper; and such taper would allow the wheel to rely less on the flange for guidance.   

This lack of a flange fillet radius is equivalent to worn flanges in railroad practice.  In that case, the entire side of the flange comes to bear on the side of the rail when the wheel shift is arrested by the flange.  This larger contact of flange to rail causes the flange to exert lift on the wheel.  If the side force is great enough, the excess flange contact can lift the wheel off of the rail, and allow it to climb over the rail and derail.   

Note the photo in the article in the first post shows the flange contact wear extending over most of the flange face on the gage side.  This is the excessive flange-to-rail contact that you get when most of the fillet radius between the wheel tread and the flange face is omitted as it is in this design. 

So there are two problems:

  1. The wheelset gage is smaller than the track gage.

  2. The wheel profile is dangerous because the tread is too narrow, the tread lacks taper, the flange lacks a sufficient fillet radius, and the flange face is parallel with the side of the rail head. 

 

Also, the two problems work perfectly together to combine their danger.  The gage discrepancy between the track and the wheelsets provides a lot of space for oscillation, and the poor wheel profile lacks the features that combat oscillation.  This is a perfect recipe for wheels climbing over the rails. 

It is the exact fundamental flaw that was discovered with the infamous “Compromise Cars” that were used in the late 1800s to allow wheels to run on two or different gages (4’-8.5” and 4’-10”) that were in use at that time.  A major train wreck known as the Angola Horror occurred in 12/18/1867.  It killed 49 people.  The use of Compromise Cars played a role of adding instability to one of the cars. 

https://en.wikipedia.org/wiki/Angola_Horror 

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Sunday, May 23, 2021 3:32 PM

Many great observations. First I might comment...HART is a design, build, maintain project...so premature frog wear will be on the builder. Second, HART is a two track mainline operation, usually with each track going in an opposite direction, with crossovers ocasionally used instead of routinely used. The train control system could be easily programmed for a speed reduction through the frogs. Third, the HART management has politics written all over it. When completed, HART will appear as a blight to the city skyline, fail to allieve roadway congestion, and only serve to line the pockets of residential developers wishing to build mega communities in the H2 corridor Farmland.

  • Member since
    January 2014
  • 7,039 posts
Posted by Euclid on Sunday, May 23, 2021 7:43 PM

Is it in operation now, or just under construction and testing?  How do they test it?  Is this autonomous operation?  Will the operational speed limit be 55 mph?  Has the method of fixing this wheel problem been adopted, or is it still under consideration?

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Sunday, May 23, 2021 7:53 PM

Euclid

Is it in operation now, or just under construction and testing?  How do they test it?  Is this autonomous operation?  Will the operational speed limit be 55 mph?  Has the method of fixing this wheel problem been adopted, or is it still under consideration?

 

The system is still being built with a projected opening of 2 to 10 years. The wheel solution is still being pondered. It is an autonomous system and unsure how they are testing it.

On another note, nothing has been mentioned about the frog size numbers and how that plays into it... http://armytransportation.tpub.com/Tr06711/Table-I-Permissible-Speeds-Through-Various-Turnouts-52.htm

Perhaps a movable point frog is the solution...

  • Member since
    December 2001
  • From: Denver / La Junta
  • 10,297 posts
Posted by mudchicken on Sunday, May 23, 2021 10:52 PM

Turnout size (including the speed) is a simple physics determination based on the turnout curve between the heel of the switch and the toe of the frog. Movable point frogs do more to protect wheel damage to the frog and smooth the ride than anything else. 

Mudchicken Nothing is worth taking the risk of losing a life over. Come home tonight in the same condition that you left home this morning in. Safety begins with ME.... cinscocom-west
  • Member since
    December 2007
  • From: Georgia USA SW of Atlanta
  • 10,534 posts
Posted by blue streak 1 on Monday, May 24, 2021 3:53 PM

It may be the problem is that the wheel gauge is less than it shoud be ?  I would really like to know the track gauge and wheel guage. Maybe all that is needed is to remove wheel from axels and place a sim on each end of the axel ?

Not being an expert the mention that wheel rims are climbing the guard rails seems to indicate that the wheels are less than track gauge.

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Monday, May 24, 2021 5:34 PM

mudchicken

Turnout size (including the speed) is a simple physics determination based on the turnout curve between the heel of the switch and the toe of the frog. Movable point frogs do more to protect wheel damage to the frog and smooth the ride than anything else. 

 

As I understood it, The problem was first discovered with excessive wear on the frogs, because of the frog guard could not steer the narrow flanges into the correct crossing position. A moveable point frog may have eliminated this. Upon further reading, the vehicle wheels were manufactured by the car builder (someone other than Hitachi who acquired it). Also the 7 or so crossovers were ordered and installed from different builders and installing contractors. Also, since the system is  a "first" for autonomous, the FTA is putting it through rigorous testing...

  • Member since
    December 2001
  • From: Denver / La Junta
  • 10,297 posts
Posted by mudchicken on Monday, May 24, 2021 6:44 PM

I'd be looking for where FTA has issued an exemption for both the wheel design and the flange bearing frog. (FRA over on the common carrier side is adamant about the rail/wheel interactions around frogs (guard face gauge, guard check gauge and depth/width of the flangeway) -there is an exemption that comes with some pretty steep requirements and plenty of testing/evaluating where there are fbf's)

The wheel tread, speed and the lesser flange width just sound like a recipe for truck hunting troubles from somebody on the heavy haul/ common carrier side. FTA is still a chicken-outfit with serious integrity issues for a reason. They need to be more than just a pa$$-through for federal fund$ to local short sighted non-railroad transportation agencies who don't want to or know how to railroad safely. Stuff they allow would never pass muster at FRA and the failings seem to appear on a regular basis.  

Mudchicken Nothing is worth taking the risk of losing a life over. Come home tonight in the same condition that you left home this morning in. Safety begins with ME.... cinscocom-west
  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Wednesday, July 28, 2021 2:24 PM

Latest spin for the over budget HART:

The door malfunction follows revelations earlier this year that HART’s trains each were running on 32 wheels that were too narrow for track “frogs” that are half an inch too wide at 12 junctions where the tracks cross one another.

Too thin or too wide?

  • Member since
    September 2003
  • 17,990 posts
Posted by Overmod on Wednesday, July 28, 2021 4:46 PM

Former Car Maintainer
Too thin or too wide?

Nobody said it couldn't be both simultaneously, which I hadn't even thought about as it's just too stupid even for design-build where vehicle design is divorced from track structure design.

Both being mutually out of spec would account for the issues.  Interesting to see if both have to be remediated... and at whose expense... to fix the situation safely.

  • Member since
    April 2011
  • 81 posts
Posted by Samuel Johnston on Tuesday, August 3, 2021 8:02 AM

Compromise Wheels??  As in The Angola Horror (1867)??  Congress made Compromise Wheels illegal in Interstate Commerce (the Constitution remember states as an "enumerated power" that Congress may regulate Interstate Commerce!) very quickly.  This very weekend I will be saying two days in Angola NY which has a display in Downtown Angola on the Angola Horror.

  • Member since
    January 2014
  • 7,039 posts
Posted by Euclid on Wednesday, August 4, 2021 7:16 AM

Samuel Johnston

Compromise Wheels??  As in The Angola Horror (1867)??  Congress made Compromise Wheels illegal in Interstate Commerce (the Constitution remember states as an "enumerated power" that Congress may regulate Interstate Commerce!) very quickly.  This very weekend I will be saying two days in Angola NY which has a display in Downtown Angola on the Angola Horror.

 

Yes, I was referring to the "Angola Horror" which was caused by the use of "comromise wheels."  As I understand, these wheels allowed running on two slightly different gages by widening the wheel tread.  But for the wider gage, the flanges were closer together than what is ideal for any given gage with properly matched wheels.  So, while the compromise wheels did have treads wide enough for both gages, they could excessively "hunt" on the wider gage track.  If the hunting gets violent, the rotating flanges can grab and climb the rail.  So the "compromise" compromised safety.

I thought this may be similar to the effect that is occurring with the railroad defect that is the subject of this thread, but I have not been able to interpret this and other references to that problem.

  • Member since
    September 2003
  • 17,990 posts
Posted by Overmod on Wednesday, August 4, 2021 9:41 AM

Keep in mind that the track gauge involved with the 'compromise' wheels was 4'10", so the wider tread was only ¾" per side.  As Ron noted this could cause effects of oscillation to develop under some circumstances.  What I recall about Angola was that the problem started with derailment caused by a bridge guard rail set what turned out to be too close to the flange face... an interference effect that may well apply to the system in Hawaii if the frogs are too far out and the flanges too far in, as I understand the reports to be saying.

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Tuesday, October 5, 2021 2:14 PM
  • Member since
    December 2001
  • From: Denver / La Junta
  • 10,297 posts
Posted by mudchicken on Tuesday, October 5, 2021 6:56 PM

Former Car Maintainer

FRA check gage violation (Guard and Face) ... Don't go there.

The only thing that saves them is FTA jurisdiction and the fact that FTA is deaf/dumb/blind when it comes to track safety. (FTA has no track safety rules to speak of, compared to FRA)

Hope nobody gets hurt until they figure out this is a bad idea. 

Mudchicken Nothing is worth taking the risk of losing a life over. Come home tonight in the same condition that you left home this morning in. Safety begins with ME.... cinscocom-west
  • Member since
    May 2003
  • From: US
  • 21,506 posts
Posted by BaltACD on Tuesday, October 5, 2021 7:41 PM

mudchicken
 
Former Car Maintainer

FRA check gage violation (Guard and Face) ... Don't go there. 

The only thing that saves them is FTA jurisdiction and the fact that FTA is deaf/dumb/blind when it comes to track safety. (FTA has no track safety rules to speak of, compared to FRA)

Hope nobody gets hurt until they figure out this is a bad idea. 

Are FRA and FTA specs for wheels and track that different to create a 1/2 inch difference when wheels go over frogs?

  • Member since
    December 2001
  • From: Denver / La Junta
  • 10,297 posts
Posted by mudchicken on Tuesday, October 5, 2021 10:10 PM

FTA has no specs .... Anything goes. The local transit operations create their own rules.(the problem with FTA in a nutshell is transit can largely ignore FRA rule if it suits them - the issue gets worse when you introduce historic transit into the mix (not an issue here)

Mudchicken Nothing is worth taking the risk of losing a life over. Come home tonight in the same condition that you left home this morning in. Safety begins with ME.... cinscocom-west
  • Member since
    May 2019
  • 465 posts
Posted by BEAUSABRE on Tuesday, October 5, 2021 11:38 PM

 I remember a story about Adm Rickover, who insisted he be aboard the sea trials of every new nuclear submarine. On one occasion, the lead ship of a new class reported a hot bearing on the prop shaft. He went aft, studied the bearing, listened via a stethoscope, studied the drawings and pronounced, "There is nothing wrong with this bearing. It's performing exactly as designed. It was just designed wrong"

  • Member since
    March 2021
  • 115 posts
Posted by Former Car Maintainer on Wednesday, October 6, 2021 4:05 AM

mudchicken

FTA has no specs .... Anything goes. The local transit operations create their own rules.(the problem with FTA in a nutshell is transit can largely ignore FRA rule if it suits them - the issue gets worse when you introduce historic transit into the mix (not an issue here)

 

it seems per the news article that HART was more interested in establishing a "Island Based rail welding contractor" for long range use than for an immediate fix for the problem regardless of oversight or lack of it..

 

  • Member since
    May 2019
  • 465 posts
Posted by BEAUSABRE on Wednesday, October 6, 2021 5:43 AM

Former Car Maintainer
it seems per the news article that HART was more interested in establishing a "Island Based rail welding contractor"

Can you say "Pork Barrel", kiddies?

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