Peter pc163 Show full post »
Pad-Ply-Pen
Thank you 
re the scarf joint in model form if manufactured would be when cutting a short length you would not have joint replicated 
Quote 0 0
Julian
Continuous welded rail, on the full size Railway, uses a similar type of sliding expansion joint. 🛤️

I am familiar with the scarf joint, from wooden narrow boats and barges. 👩🏼‍🔧

I have always liked the shape of a proper scarf joint...😊

The top diagram, above, would not be very good at maintaining the track gauge, whereas the bottom diagram seems to be ok from most points.

🙋🏼‍♀️🏴󠁧󠁢󠁷󠁬󠁳󠁿

I had the same feeling about the wedge shape, but on reflection the maximum expansion is in the region of a couple of mm, which might well enable the joint to work within a fishplate.  I will go back to the AnyRail and do a large scale diagram to see what the actual measurements might look like.  Could be fun, finding different slope angle results.

J
Quote 1 0
Julian
I had a look at drawing 3 expansion joints, to see what might be revealed.  I also looked at a few pictures, online, to see what they might reveal.

The AnyRail Ap was used in 1:1 scale and the Snap to grid set to 0.05in, but what you see in the picture is a magnified version, once complete, for clarity.  The grey rhombus at the top of the diagram is 1" wide, for guidance.
There are some points to note;
I set the 3 expansion gaps at 1/10" {2.5mm}, which might cope with a reasonable length of rail.
The top version has a short diagonal scarf, which would fit into a Peco fishplate [grey rectangles].
The middle vs will also fit into a Peco fishplate.
The lower vs has a 1" diagonal scarf and will not fit completely into a fishplate, {of this design}.
sliding expansion joint.jpg  Observations:
The top vs would need the fishplate soldering to one of the rail ends, to prevent it walking and keep the fishplate from twisting, too.
It looks fairly easy to construct, particularly with a small jig.  It may well be quieter than the middle vs.
The middle vs would be less likely to walk and little risk of twisting within the fishplate, but would also benefit from a touch of solder to one rail.
It looks very easy, particularly with a jig, perhaps just dropping a router to mill off 3/10" {8mm}, to halfway through the rail end.  It would need testing to see whether the inside gap might be large enough to drop a properly tapered rail wheel into it, sufficiently to cause a large enough bounce to derail it.  It certainly looks large enough to be very noisy.
The lower vs had a very small gap along the 1" scarf, as expected and could only just be detected at 5:1 magnification.  I later found that "some" pictures of full size UK sliding expansion joints were of similar proportions.  It should stay within the fish plate, as the gap is so small, but would require soldering it to one of the rails, to prevent walking. 
It may be that suitable jigs are already available from companies which sell materials for track building enthusiasts.

The 2 brown rectangles are placed where the sleepers were on the UK 1:1 scarf joints. Of further note was that the 1:1 versions had check rails inside their joints.   It was also clear that the last 3 to 4in of the scarf was ground // with the main part of the rail.  The Chairs on the sleepers [above] were much larger that the other ones and there was bracing, across 4 sleepers inbetween the tracks.  It would seem to be a prudent arrangement to have a few sliding expansion joints, suitably placed on a layout which has significant lengths of track soldered in the manner of the OP.

Regards
J

Edit:- It's possible to place 2 x fishplates across the longer scarf, in the lower rail.  They would cover the whole scarf joint, possibly best both soldered to one rail end.
Quote 1 0
AHJAY
*** My thoughts...

* Personally I think an interesting thing to try but perhaps not a practical thing on a model railway overall... Lots of work for almost no gain and- with potential pitfalls.
More specifically, Model flatbottom rail, especially the heavier code 100, has a natural tendency to try to twist when bent because the foot is wider than the head and the rail is already pre-stressed with this bias when it is rolled. A halved joint would be less controlled because you are filing away half of the foot on each rail and therefore lessening the ability of the fishplate to control it. Once the radii gets to be less than 24" this will become an important factor. However, give it a go - its all part of the fun.

* PPP What part of soldering do you want help with? I can certainly fill in any gaps for you.

* BTW sliding joints are less and less on the real thing. Currently most current continuous welded rail does not use sliding joints anywhere - it would not help as the portential expansion is huge now the lengths ate so long. Expansion is mostly controlled to force the expansion to be largely lateral by using high mass/ deeper than normal sleepers, stronger rail fixings and specifically tamped deep ballast.

The expansion possibilites are too big to take care of any other way - for example on Australia recent completion of the line from Alice Springs to Darwin, the potential expansion spread from coldest to hottest is more than 2 kilometres!

regards, Ahjay
Quote 2 0
Stuart Birks StuartB
I'm not an expert or even a rocket scientist but I can assure you that the sleeper spacing on my track is correct even on curves. It just takes a bit of care to adjust it. I find it strange that you are so worried about this being correct but then compleatly ignore prototype practice of having the track joins in line. Are you trying to match the protype or not?

Stu
Quote 0 0
newbryford
I'm not an expert or even a rocket scientist but I can assure you that the sleeper spacing on my track is correct even on curves. It just takes a bit of care to adjust it. I find it strange that you are so worried about this being correct but then compleatly ignore prototype practice of having the track joins in line. Are you trying to match the protype or not?

Stu


Staggered joints are prototype practice in the US.
And how many UK layouts have joints at 240mm - or 120m (4mm scale)?

If anyone is using Peco track for British stuff, then the sleeper spacing is too close anyway.
And the rail joiners do not look like the real thing either - so does it really matter if the joints are staggered?
Quote 2 0
Julian
* AHJAY  Absolutely agree, particularly for my piddly small layouts.  It simply caught my interest to see how it might be applied, should someone, with a much larger layout, wish to adopt soldered joints to simulate continuous welded rails and solutions to the expansion.

Re Alice Springs to Darwin, I saw that in a couple of the search references, it made me giggle thinking about the possibility of mobile station platforms to keep pace with the expanding rails. Commuter's timetable nightmare and could they run fast enough to keep up with the coaches on the expanding rails, in the rising morning sun and evening sunset!  More seriously, I also noted the references to pre-expansion techniques before laying the welded rails {and re-directing of forces from length to width within the rails, physics employed is new to me, but interesting and clearly practical - for the full size!!}.  

* Stu  I am sorry if I gave the impression that I was making comment about your sleeper spacing on curves, or otherwise, I wasn't.  My reference to sleepers was only an observation on where the 1:1 relation of sliding expansion joints were to their sleepers.  Likewise, my observation of a practical modelling method to enable easier smooth joins on flexitrack curves, with the use of staggered joints.  Again, I apologise if I gave any impression that I was ignoring full size practice.

Regards
J
Quote 1 0
AHJAY
Often the simplest things work well. The thread shows a good technique that will help many modellers.

Please note I have modified the thread title to make it a more positive search in future for those looking for help in this area - and to be a little less controversial. (I hope to OP dies not mind)

Its a god, tried and tested approach used by many, but given the soldering of fishplates, it may have drawbacks with later expansion if the modelling room has direct sunlight or significant heat changes.

An added thing that helps is to lay the track as demonstrated, then at regular intervals, secure the rail at a convenient place by driving in a couple of our gold plated track pins within a sleeper gap. Solder the pin heads to the rail (easy because of the gold plating that solders beautifully) then cut an expansion gap with a thin dremel disc between them. The pins can flex a little so rails will stay in alignment and expansion will then be catered for.

regards, Ahjay
Quote 1 0
Terry Metcalfe Terry Metcalfe
Unfortunately, I used Peco code 55 track and it is next to impossible to slide the rail into the sleepers.
Or has someone found a way.  If so please share.
Quote 0 0
AHJAY
Understood. Much of this specific discussion is really only for HO or OO use and larger.

N scale Code 55 is indeed difficult as the rail is in fact much heavier than C55 (the C55 is really just the visible bit), with much more metal out of sight and moulded into the sleepering. Unfortunately that is a problem thats not really solvable with the techniques here.

Ahjay
Quote 0 0