Trains.com

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Derail issues

2985 views
25 replies
1 rating 2 rating 3 rating 4 rating 5 rating
  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Wednesday, August 17, 2022 10:38 AM

Thanks again for the more recent replies. I had some unexpected downtime yesterday, so being my usual impatient self, the hidden turnouts are out, track re-laid, and all is running smoothly. 

jpg, just to keep your blood pressure in check LOL, single tracks running into single tunnel portals. The original intent of the hidden turnout was to have additional hidden staging area for operational purposes, but I will find another solution.

 

And Simon, I have access, awkward but doable, from behind, but the PITA factor is directly proportional to my age, so out they went.

 

 

  • Member since
    November 2013
  • 2,775 posts
Posted by snjroy on Tuesday, August 16, 2022 9:41 AM

Coaster

I found some time to do a bit of troubleshooting Friday and Saturday. The answer is.....basically all of the above. The turnout was slightly out of level were the train enters, and the loco had a slight hitch just after the turnout, which resulted in the 2nd or 3rd car hitting the uneven track at the same time as the hitch. I mentioned above I had not connected all the feeders yet, so I connected that section to the bus and took care of the hitch, then added a tiny shim to level out the turnout. Now all but very light cars seem to get through OK. However............this whole experience has me thinking I don't want the curve/incline/turnout combination in an area of the layout that will have limited accessability once I get the mountain and scenery in place. So, I am thinking of reworking the track arrangement to remove the turnout and have 2 individual tracks through this section. I still have to have the curve and incline, but that is not an issue in other areas of the layout. I won't get back to working on it until late September, so I'll mull it over in the meantime.

Thanks to all who responded.

 

A turnout in a tunnel is high risk for sure. We have one at the club (not my idea!), and it's a PITA when derailments occur. Maintenance can also be highly problematic. Of course, there are ways to make track in tunnels more accessible. Rear-access is probably the best option, if that can work for you.

Good luck!

Simon

jpg
  • Member since
    July 2019
  • 32 posts
Posted by jpg on Sunday, August 14, 2022 3:16 PM

Another two cents - I agree if you can avoid a hidden/covered turnout that's good.

I don't know your plan, or the possible change, but one thing I sometimes see in model railroads that bugs me no end is a single line that splits into parallel lines just before a tunnel or bridge, so as to make an unnecessary double-track tunnel/bridge that leads onto a single-line. Completely backwards from the prototype which would save money on bridges and tunnels by single tracking them (or using a gauntlet track) even if the adjacent lines were double.

Ok, that's just one of my rants... it's your railroad.

  • Member since
    October 2020
  • 3,604 posts
Posted by NorthBrit on Sunday, August 14, 2022 1:05 PM

Excellent answers.

Just to add my My 2 Cents

In real life  at curves etc.  there are speed limits.  Most model railroading exceeds the speed.  Off course the trackwork has to be good, but speed limit signs add atmosphere.

 

David

To the world you are someone.    To someone you are the world

I cannot afford the luxury of a negative thought

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Sunday, August 14, 2022 11:34 AM

I found some time to do a bit of troubleshooting Friday and Saturday. The answer is.....basically all of the above. The turnout was slightly out of level were the train enters, and the loco had a slight hitch just after the turnout, which resulted in the 2nd or 3rd car hitting the uneven track at the same time as the hitch. I mentioned above I had not connected all the feeders yet, so I connected that section to the bus and took care of the hitch, then added a tiny shim to level out the turnout. Now all but very light cars seem to get through OK. However............this whole experience has me thinking I don't want the curve/incline/turnout combination in an area of the layout that will have limited accessability once I get the mountain and scenery in place. So, I am thinking of reworking the track arrangement to remove the turnout and have 2 individual tracks through this section. I still have to have the curve and incline, but that is not an issue in other areas of the layout. I won't get back to working on it until late September, so I'll mull it over in the meantime.

Thanks to all who responded.

jpg
  • Member since
    July 2019
  • 32 posts
Posted by jpg on Saturday, August 13, 2022 9:41 PM

My suggestion: get a loose truck from somewhere, just an ordinary two-axle 4-wheel truck. Check the gauge on the wheels. Then roll it through the turnout with your finger on top of the truck, back and forth, a number of times. Use your finger to press the truck toward the outside, simulating lateral force. If you feel any bumps at all, then either file more or replace the turnout. 

I'm thinking most likely the filing has made the problem worse, and more filing won't help. You might be able to rescue the situation by doing a lot of filing, to thin the rail head where that bump is so it gradually tapers (holding the file vertical and go up and down). But then you'll have a problem coming the other way (from the left) with wheels picking the points when the turnout is thrown the way you show it. In that case, you'll have to file on the point also to make it sharper, again vertically filing up and down.

If a single truck goes through the turnout in all directions with you applying side pressure that would tend to derail it, and you don't feel any bumps at all, then the turnout is good and you need to look at some of the other stuff people are talking about like coupler slack or whatever. But the truck test will eliminate a lot of variables and let you focus on just the turnout.

  • Member since
    December 2001
  • From: AU
  • 713 posts
Posted by xdford on Saturday, August 13, 2022 7:59 PM

selector

Have you checked the wheel gauge of both the axles that seem to be troublesome AND the diverging/through rails.  Some points lose gauge quite substantially after the actual tip and indent...

Thanks Crandall,  You reminded me about a Peco trait with their more recent turnouts in that the curved point rail can go under gauge by virtue (I think) of its elastic memory.  I have had an issue with 1 turnout in a yard ladder when pushing cars into a siding and it has often been the second or third car through the turnout.  

The cure for me was to very gently bend it with thumbs so that my NMRA gauge goes through it OK.  It was only a couple of freight cars presenting a problem so that very slight loosening of the trucks mentioned earlier also helped.

Hope this and others have helped,

Regards from Australia

Trevor

  • Member since
    February 2005
  • From: Vancouver Island, BC
  • 23,330 posts
Posted by selector on Friday, August 12, 2022 1:29 PM

Have you checked the wheel gauge of both the axles that seem to be troublesome AND the diverging/through rails.  Some points lose gauge quite substantially after the actual tip and indent. I'm also guessing that weight is a problem, maybe also with trucks that aren't as freewheeling as they might be. 

  • Member since
    November 2013
  • 2,775 posts
Posted by snjroy on Thursday, August 11, 2022 9:56 PM

Hi there. Make sure that your turnout and track are level laterally (across), otherwise some wheels may lift and derail. Also make sure that the turnout is flat - use a file to remove excess height at frogs, etc.  

Simon

  • Member since
    December 2001
  • From: AU
  • 713 posts
Posted by xdford on Thursday, August 11, 2022 6:11 PM

Hi there,

I did have a thought about the voltage/jerk bit so bear with me please.  Bearing in mind the 2-3 cars behind the loco,  perhaps if you check the track where the loco is over when that 2nd or 3rd car is on the turnout. 

Most locos hve flywheels and that is enough the keep a loco moving but even an instant of no contact with voltage will slow it for that micro instant. The flywheel carries it but it becomes a microjerk on a 1:87 size vehicle.

Plan of action? 

Double check your track level...

1. The track may require packing if the rail does drop away from the wheel.  You may be able to detect a dead spot by taking your slowest moving loco and testing it over the section and seeing if there is a slight stall. A shim out of a plastic lid may be enough.

2.  The track in that instance may need simple cleaning. It may look clean but a slight build up may cause the interruption to power.  If there is a thin film of white glue, matte medium etc for ballasting on the rail top, you have an interference in the supply to the track and consequently the loco.  I have had an issue similarly.

3.  If you have spiked to track down tightly,  your track at that point may be tightly gauged and momentarily putting the squeeze on the flanges, slowing it slightly.

I wrote an article in Rail Modeller Australia about "Towards more reliable track" which may help.  It is a freebie to download in issue number 3  (https://railmodelleraustralia.com/RMA%20Mag%20Web%20Oct%20Nov%2020.pdf) which may help also.

Good Luck

Trevor

 

 

 

 

 

 

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Thursday, August 11, 2022 1:43 PM

Hey Matt, probably my reading more than your wording....interestingly enough, I do not derail on the inner loop, only on the outer loop. More testing to come. Thanks

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Thursday, August 11, 2022 1:41 PM

OldEngineman

Does the grade begin after the switch, or before it? Could the weight of "the train behind" be pushing off the cars up front? Have you tried different train lengths for purposes of experimentation?

Curve, turnout and subsequent straight section are all on the same down grade. I will try different train lengths when I get a chance to do more testing. Thanks

 

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Thursday, August 11, 2022 1:39 PM

Hello jjdamnit, The layout cannot be reworked, it was designed for a specific area, and the incline is needed to get clearance for another section. This particular area will actually be hidden under a mountain scene and I will have a somewhat awkward access area to service it if something goes wrong, thus my testing over and over again before I proceed. Thanks for the comments, and as mentioned above, I will post an update when I get back to working on it.

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Thursday, August 11, 2022 1:32 PM

Thanks Trevor for the comments. The curve, turnout and straight section after the turnout are all 1.5% down grade. I will have to look into your comment re voltage/jerking, I have not connected all my feeders to the bus yet, as I wanted to be sure the trackwork was good before making the connections. The couplers are all body mounted. As you suggest, I am going to take a step back and try to be a bit more methodical in my problem solving, starting with 1 or 2 cars that I know are correct weight, etc, then adding one at a time. My schedule is such that I won't get much time to investigate further until late Sept, but I will post an update, and hopefully solution, when I get back to it.

  • Member since
    February 2021
  • 1,110 posts
Posted by crossthedog on Wednesday, August 10, 2022 11:32 PM

Coaster

Thanks Matt, I'm derailing coming downhill into the frog end, runs fine uphill coming into the points.

Maybe I said it upside-wrongwards. I did mean, yes, when the cars approach the frog end on the inside curve. This is probably overkill in light of the many other more likely culprits that are being suggested, but I'll clarify just in case.

If you click on this photo, then close it, and then click it open again, you'll get the max resolution and maybe you can see where I've filed down the corner indicated by the yellow arrow. The cars that derail here do so when the train is coming from the divergent path at upper right, id est, the "inside curve". I think some flanges fall into the space between the end of the point and that corner. Then they hit the corner and ride up. Oddly, they don't derail coming from the "straight" or "outside" path, even though you'd think the corner would present the same issue there. Maybe the slight centrifugal force of the tighter curve is enough to push the wheel further into the hole.

Anyway, if all else fails it might be something to consider. Good luck.

-Matt

 

 

 

 

Returning to model railroading after 40 years and taking unconscionable liberties with the SP&S, Northern Pacific and Great Northern roads in the '40s and '50s.

  • Member since
    February 2018
  • From: Danbury Freight Yard
  • 459 posts
Posted by OldEngineman on Wednesday, August 10, 2022 10:27 PM

Does the grade begin after the switch, or before it? Could the weight of "the train behind" be pushing off the cars up front? Have you tried different train lengths for purposes of experimentation?

  • Member since
    September 2014
  • From: 10,430’ (3,179 m)
  • 2,311 posts
Posted by jjdamnit on Wednesday, August 10, 2022 9:37 PM

Hello All,

xdford
If the track is level after the turnout the locos are perhaps pulling at one end while the rest of the train is pushing and that second or third car is effectively the meat in the sandwich. (Emphasis added for clarity.)

I'm thinking in a similar line...

If the turnout is positioned on a slope- -rather than on level ground- -entering the diverging side from the upslope, could cause the accordion effect so eloquently describe...

xdford
...the meat in the sandwich.

Another possibility is the transition between the slope and the flat is too abrupt upslope of the turnout, presuming the turnout is on level ground.

There have been many threads written about transitions- -their difficulties, and some solutions.

One possible solution is to...if possible- -relay the track- -adjust the track plan (Gulp!).

If the parallel tracks can be extended to level ground, the turnout can be installed on level ground.

That might solve the derailing issue on the slope.

However, if the track plan cannot be modified, rather than having the turnout on a slope, create a flat area to place the turnout.

A #6 turnout might need to be used due to space constraints.

If the cross-over trackage is done on the level, and the transitions in and out of the area are smooth, I suspect derailments will be reduced even with a tighter turnout.

A station scene could be used to justify the area of level ground needed for the turnout.

Other factors/suggestions that have been posted here can also lead to the failure you describe and possible solutions.

Please keep us informed of your progress, questions, and/or solution(s).

Hope this helps.

"Uhh...I didn’t know it was 'impossible' I just made it work...sorry"

  • Member since
    February 2008
  • 2,360 posts
Posted by kasskaboose on Wednesday, August 10, 2022 8:31 PM

Could speed (too fast/slow) be an issue also?  I've sometimes sped up or decreased the throttle to address it.  Hope no one goes into the physics behind my crazy idea.

  • Member since
    December 2001
  • From: AU
  • 713 posts
Posted by xdford on Wednesday, August 10, 2022 7:28 PM

Hi Coaster,

 

You are talking about two sets of dynamics here. Going uphill,  all the couplers are taut.  Coming downhill, the train behind is pushing the cars even on a 1.5% grade. If the track is level after the turnout the locos are perhaps pulling at one end while the rest of the train is pushing and that second or third car is effectively the meat in the sandwich.  The voltage at that point may also be enough that your locos are in effect jolting the train almost imperceptibly.  I would also be checking level of the track up to about 6 inches before the frog.  A little superelevation packing on the outside rail may also help.

If the couplers of the car are truck mounted, that would present an issue in itself.  Have your trucks got enough swing?  You may need to loosen the truck screws minimally like an eighth of a turn and no more than that. 

Perhaps testing various weights and lengths of trains might define this problem for you as well...

 

Hope this helps

Trevor

Editor  Railway Modeller Australia

 

 

  • Member since
    February 2018
  • From: Flyover Country
  • 5,557 posts
Posted by York1 on Wednesday, August 10, 2022 6:07 PM

Is this the same set of cars each time?

I had something like this (in N scale, though) and I narrowed the problem to certain cars and their couplers.

The closest thing I came up with was that going downhill, there was some slack, and the curve allowed the couplers to move just enough to separate.

I might be completely wrong about your issue, though.

Edit:  I reread this and realized you are talking about derailing, not uncoupling.  Sorry about that.  My excuse is that I'm old.

 

York1 John       

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Wednesday, August 10, 2022 5:43 PM

Thanks Matt, I'm derailing coming downhill into the frog end, runs fine uphill coming into the points. 

 

  • Member since
    February 2021
  • 1,110 posts
Posted by crossthedog on Wednesday, August 10, 2022 5:32 PM

I have some rolling stock (metal wheels) that hop a little or even derail at a particular point on a turnout (brand new Atlas), where the stock rail is carved out just a little to allow the point to nestle in when it's switched to that side. The place where the carve-out stops makes an abrupt little bump and the wheels sometimes catch on it -- I think the flange is actually riding up on it. It shouldn't happen -- that's the whole point of the carved out piece of stock rail, but the edge of it is too sharp. I filed the corner of it down a little, seemed to help. This may not be analogous to your situation, though, because this happens when the train is moving through a trailing point turnout -- if I said that right. I mean the train is coming into the turnout from the divergent path.

-Matt

Returning to model railroading after 40 years and taking unconscionable liberties with the SP&S, Northern Pacific and Great Northern roads in the '40s and '50s.

  • Member since
    December 2020
  • 17 posts
Posted by Coaster on Wednesday, August 10, 2022 5:09 PM

Metal wheels. I will check the weight of all the cars in question. I was also thinking of adding a little elevation on the outside rail, but I'll check the weights first as you suggest. Thanks

  • Member since
    March 2011
  • 1,950 posts
Posted by NVSRR on Wednesday, August 10, 2022 4:59 PM

Metal wheels or plastic wheels?

A pessimist sees a dark tunnel

An optimist sees the light at the end of the tunnel

A realist sees a frieght train

An engineer sees three idiots standing on the tracks stairing blankly in space

  • Member since
    March 2011
  • 1,950 posts
Posted by NVSRR on Wednesday, August 10, 2022 4:59 PM

ok. derails the second or third cr on a curved downgrade at a turnout.  What part of the turn?  sounds like the cars have a wieght issue. the crs are to light to keep from being pushed in a straight line.  so it gets to the where the closure and wing rail meet and gets pushed off because of the changes there.   So are your cars weighted to the standards?  might help.   suggest taking one train that derails and weighting the cars to the standard and see if that changes anything. 

Might be a subroad b ed issu there too that changes the rail angle and elevation to each other just enough to give you a headache.    try the weight thing first.

 

Shane

A pessimist sees a dark tunnel

An optimist sees the light at the end of the tunnel

A realist sees a frieght train

An engineer sees three idiots standing on the tracks stairing blankly in space

  • Member since
    December 2020
  • 17 posts
Derail issues
Posted by Coaster on Wednesday, August 10, 2022 4:15 PM

I have a section on the layout that has 2 parallel tracks on a 1.5 % downgrade, 33 inch radius curve. The tracks come out of the curve to #7 turnout then proceed downhill on a straight section. I derail almost everytime as the train enters the turnout, not the loco but the 2nd or 3rd car. I have used a gauge to check the track width and wheels on the train, all are good. The there are no large gaps in the track, and it doesn't derail anywhere else, including other sections with curves leading into turnouts, nor on tighter curves. The cars are all 60 feet or less. The same train will run on the same section in the opposite, uphill direction with no issues. I thought maybe it was too tight from the curve onto the turnout so I replaced the turnout with a #7 curved turnout, same problem. Any suggestions?? This is HO gauge, Peco flextrack and Peco turnouts.

Subscriber & Member Login

Login, or register today to interact in our online community, comment on articles, receive our newsletter, manage your account online and more!

Users Online

There are no community member online

Search the Community

ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT
Model Railroader Newsletter See all
Sign up for our FREE e-newsletter and get model railroad news in your inbox!