Derailment problem solved... for now

[Paulus]

Some time ago I posted a thread on derailment problems that I assumed were caused by metal wheel.
It was already mentioned that the track was also causing problems on a certain spot.
When running a train today I had again derailments on this same spot and I noticed something suspicious. There was a gap between the tracks and the concrete base seems cracked...



Just around the corner of this curve the track continues straight and runs over a border (garden one side, small patway other side. About 50cm high at the patway side). But I noticed the straight part shows cracks in the concrete "ballast" (see right side of track):



So.. I am afraid the raised border tilts to one side, like this:



The tilt is than causing the tracks in the curve to move. The concrete road is than working like a lever point, causing the curve to widen up as it presses against the outer rail of the curve (hope it is clear waht I try to explain...). It is causing slight but noticeable tension/deformation of the tracks. The deformation is most likely the cause of the derailments on this spot.

I removed the concrete road piece and loosened the track in this part by removing the track-nails and pulling the tracks gently up from the ballast (the ' conrete - ballast mix is only preventing the ballast to flow away during rains, not very strong so the track loosens easy). Leveled the track and placed some gravel/ballast underneath.





The train runs perfectly for now. The tilting border does not push the track away anymore.
But I am afraid when time goes the tilting can get worse. Have to see how I can prevent that.

 

[idlemarvel]

Nice piece of detective work. I don't think you can expect to escape occasional re-ballasting on an outdoor layout. Just like the prototype!

 

[Greg Elmassian]

Cross level is often missed at first, and it causes strange problems. In the US, the evolution of garden railroads started with trying to bolt the track down every few inches, the "wisdom" being keeping it from moving and keeping track joints tight.

Soon, the "common wisdom" realized that wood warped, and grew and shrunk and rotted.

Next pouring concrete was tried, and just concrete alone cracked and came apart as it settled, and also with freeze and thaw cycles.

At this point, half the people went to free floating track in ballast like the prototype, the other went to steel reinforcement of the concrete roadbed, which has led to large chunks of concrete that may or may not stay in place.

As you have found, movement of the concrete roadbed can result in different effects on the track.


Greg

 

[Paulus]

Thanks for your answers.
I try to let the track float free indeed, at least for this part and see how it goes with the tilt of the concrete. If it stops it is oke for me but if it does not it have to be reinforced or replaced. It will be a kind of a challenge than but that is what makes it interesting as well ;-)

 

[phils2um]

One other thing I noticed in your photo just to the right of the red arrow. The tie plate on the sleeper has its tabs that hold the rail either missing or the rail is not properly seated. A similar problem happened near a joint in a piece of curved flex track in my tunnel about a year ago. Certain 4 axel wagons always seem to derail at the location. I replaced the flex track with some sectional pieces which cleared up the problem. I probably did not need to though. If i had just reseated the rail into the sleeper tie plates the problem would also most likely have been solved. I have since reused the "problem" piece of flex track on my trestle after replacing the questionable section of ties and have had no problems with it.

 

[Greg Elmassian]

If you put loose ballast under the track, and you have a cross level problem, it make a week or so for the track to "level out" again, the heating and cooling will help relax the rails.

So ballast as best (level both ways) as possible, then then wait a day or two, check it again and adjust as necessary.

Greg

 

[Gavin Sowry]

That joint is out of line..... trained eye of a track man.

 

[Gavin Sowry]

That joint is out of line..... trained eye of a track man.

Allow me to elaborate. When joining two bits of track, they need to butt up properly, and have a smooth 'line' from one piece to the other.
In this case, from the look of the photo, the adjoining tracks are not smooth... both ends are pointing into the centre of the 'circle', thus creating a kink in the track.
This is further exaserbated by the loose rail joiner seen in one photo at the right hand of the joint, this has allowed the kink on one rail to push even further inyo the centre, and also not line up with the other rail. There is a high likelyhood of the wheel flange 'picking' the misaligned rail, and derailing. Get the pliers out, and squeeze up the joiner so you get a smooth joint.

In the track game, there are various 'parameters' that are taken individually, and in total, to measure track quality. These are 1) Line, are the rails straight, and the curve constantly smooth, 2) Top, no dips in the track, particularly at joints, 3) Gauge, dictated by standards, with tolerances, 4) Cant, for model purposes, don't use it, rthen you won't have problems with the associated Rate of Run Off, and finally, the biggest derailer of all 5) Twist. That's where you have a bit of cant followed quickly by a bit of negative cant (or even the same). For Cant, in North America, read Superelevation.

Hope this helps.

 

[PhilP]

If your track is not fastened down, and the border concrete moves a little more...
You can just adjust the ballast to give you a level track.

What sort of 'ground' do you have? - Sandy, clay, etc.

I am on clay, and there is considerable movement, with the ground temperature, and amount of moisture in the ground. - I can get a couple of inches (change in 'ground-level' from warm and dry, to a sudden wet spell.

 

[Madman]

When I initially built my garden railway, I used the misguided wisdom Greg posted out, concrete and screwing the track to it. After a few short years, I adopted the floating track method. Haven't looked back since.

 

[Greg Elmassian]

I'd like to underscore a distinction: superelevation is done on purpose for higher speed trains on curves, the outer rail is raised, to give (usually) a 2 to 4 degree tilt. Also superelevation REQUIRES a vertical "easement" or transition from level track to and from the superelevated track. This is an effect not normally used in model layouts and requires excellent trackwork and equalized trucks, etc.

Cross-level is the measurement to check whether your track is level across the rails. It should ALWAYS be so (except in the case of deliberate superelevation of course). Sometimes this is also called warp. Excessive cross-level will wreak havoc and is often missed by the casual observer. The "suspensions" in our models are nowhere near how a real train operates, and many locomotives cannot handle cross-level issues.

The symptoms are often "mystery" derailments, where gauge and wheels look ok.

Best, Greg

 

[Rhinochugger]

I've probably mentioned this before, but in the UK, Building regulations now require a minimum foundation depth of 2 metres to prevent ground heave - and that's for a house.

Given that no sensible person is going to dig 2 metre deep foundations for their garden railway, then there is no way of constructing an intermediate type of foundation that will guarantee no movement.

Therefore, you've only got to go down to a 'firm bottom' (technical term ) and then lay the track on some form of ballast with very occasional fixings to keep it in the right field.

Yes, the formation will move over time, and there will need to be a bit of re-alignment every now and then.

The alternative is to dig down 2 metres. I have about 300 ft of track ( 'scuse the mixed measurements) and I certainly haven't dug anything like that, nor have I ever wished I had

 

[Paulus]

Thank you all for your replies and advise gentlemen! And sorry for my late reply... work got in the way...
Indeed I had make the mistake Greg pointed out; fixing the tracks on the concrete. I also fixed the ballast with a concrete/gravel mixture (to prevent it from heavy rain and children hands...). But I guess floating rails, loose gravel and regular cross-level inspection are better indeed.

The ground I placed the concrete on is just the sandy-soil that was present in the garden. The beds were raised by former house owner and it is mostly soil. Underneath there is a layer of yellow sand, probably from an old terrace (or left over from house construction?).

Gavin and Phil: you have good eyes!
The joint is indeed not smooth anymore. I don't know if it was in this condition when I laid the tracks as I had squezed all the joints with a pair of pliers to fit tight. The rails is used on my previous layout and I aquired it second hand (could even be fourth hand) in the first place. As I do not use track power this was not that of a problem. This also explains the missing / incomplete tie plate on the sleeper. This piece of track missed a sleeper and I had slide a loose sleeper from the side in it after I laid the track (it is a pain to remove the joints and slide a good sleeper on the rail).

 

[Gavin Sowry]

Gavin ........you have good eyes!

Once, (upon a time, as the saying goes, but 45 years ago to be precise)I was crew on one of our Track Recording Cars, similar to you blokes Neptune. Middle of winter, absolutely hosing down, and blowing a gale. I was analysing the trace, and momenterily looked up ahead, and let out an expletive..... then we hit the twist in the track, sent one bloke flying, and the pen line way off the end of the scale. It was a 'Stop all trains, and repair immediately' job. The track gang had dug a trench through the crib to release ponded water from one side of the track, only trouble was, water did what water does, and scoured out even more ballast. District Engineer ordered the Inspector to 'go fix, now'. Poor old fella, month off retirement, and didn't look a day under 80, took a hike.

 

[Sean.]

We have some who will install posts every 4' or so down past the frost line for there raised layouts ..

 

[Paulus]

Once, (upon a time, as the saying goes, but 45 years ago to be precise)I was crew on one of our Track Recording Cars, similar to you blokes Neptune. Middle of winter, absolutely hosing down, and blowing a gale. I was analysing the trace, and momenterily looked up ahead, and let out an expletive..... then we hit the twist in the track, sent one bloke flying, and the pen line way off the end of the scale. It was a 'Stop all trains, and repair immediately' job. The track gang had dug a trench through the crib to release ponded water from one side of the track, only trouble was, water did what water does, and scoured out even more ballast. District Engineer ordered the Inspector to 'go fix, now'. Poor old fella, month off retirement, and didn't look a day under 80, took a hike.

Auwks. not a good day to be track inspector... by take a hike you mean he ran of? Or did he go do the job?

Ah, Neptune system; Exeter and Teign Valley Railway - Christow Station (I had to look it up) ;-)

 

[Gavin Sowry]

Auwks. not a good day to be track inspector... by take a hike you mean he ran of? Or did he go do the job?

Ah, Neptune system; Exeter and Teign Valley Railway - Christow Station (I had to look it up) ;-)

He got out, and fixed the track.
Thanks for that link.... our two cars were basically the same, gauge excepted. Was always a Sunday job in the Wellington suburban area..... paid at overtime rates (thus a sort after assignment). We never ran at night, though.