You may have noticed I’ve made a lot of posts on the blog recently. This wasn’t planned but I tend to post when I’m doing something I really enjoy. As doing wiring and benchwork are two of my favourite things I’m like a pig in mud at the moment, hence the posts 🙂
Many years ago a modelling friend of mine asked me how I managed to avoid bumps in my sub-roadbed. I replied “the finger gauge”. If you can feel a bump with the end of your index finger between one section of sub-roadbed and another then your trains will feel it too. As is the way with these things you will only discover the bump after the track is laid, scenery is in and you have some friends round to display your newly completed section of layout. Have you ever said to friends visiting your layout: “the trains ran perfectly over that section of track yesterday”? If you have (and I defy anyone to tell me they’ve never said something like this) then you haven’t paid sufficient attention to the finger gauge.
One of my absolute, non-negotiable rules is that when two pieces of ply (acting as sub-roadbed) butt up against each other there must be no discernible difference in the level of their top surface. If I can run my finger over the joint and not feel even the slightest bump then I’m happy. If I can’t do this then I bloody well keep fiddling and working till I can’t feel a bump. These bumps always, always, always transfer themselves up into the track and will affect running no matter what scale you’re working in. A layer of cork and flex track will not make the bump go away! This is true for both home layouts and portable ones.
In my experience you can’t rely on the thickness of ply to remain consistent between batches or even on a single sheet. A designated thickness is not an absolute, it’s merely an average. 12mm ply (1/2″ for those of us still in the imperial mind-set) is likely to sometimes be 12mm thick but the next sheet of 12mm ply may be 12.2mm thick, or 11.8mm thick. You can’t be sure and you should never, ever assume that all the sheets will be the same thickness or even that one sheet will be 12mm thick around its whole perimeter. A sheet might be 12mm thick at one end and 12.3mm thick at the other. That .3mm will cause you no end of problems if you ignore it and rely on the manufacturing tolerances to be perfect because sub-roadbed is secured down to a surface, thus making the underside the datum with the top surface left to reflect the differences in the nominal thickness of the material. This is ply I’m talking about, not MDF, but then I’ll never again use MDF a sub-roadbed after what happened on Queens Wharf.
If you look at the short length of curved ply I posted a photo of the other day you’ll see that it crosses a join between two sections of layout. I have spent the last couple of days getting the tools and materials I need to turn this length of ply into a section of sub-roadbed that I could cut and with no discernible bump across the joint. The problem with this part of the project was that it already had a ply scenery base down (the surface of what will become the water of the Hunter River and the stream that joins it at this point) and I didn’t want to go cutting holes into these pieces of ply. So this presented me with the situation where I couldn’t allow for slight differences in sub-roadbed thickness by adjusting a riser’s height (as per L girder or box frame benchwork). This short length of critical sub-roadbed would be held up by three small blocks of 3×1″ pine and to avoid bumps at the join these would need to be cut extremely accurately. The best way I know of to make such cuts is with a compound mitre saw with a nice, sharp blade. I happen to own just such a machine but it was 1 1/2 hours drive away in the garage at my partner’s home. So yesterday morning I made the trek over to pay her a visit, listen to her complain about work, say hello to the kids and headed back home after loading the saw into the boot of my car. Without telling me my 17-year-old step daughter appeared in the kitchen minus her beautiful, long blonde hair, having had it chopped to shoulder length. She ignored my devastated expression and complained that people were telling her the new do made her look younger. I thought “just wait a few years, that won’t be something you’ll complain about!” 🙂

This shot shows the new work in place but prior to being secured permanently in place. I’ve added in the two sections of MDF fascia I describe in the main text and the cut line and notch can be see where I’ve labelled them.
Today I got stuck into the good stuff after a trip to Bunnings in the morning to pick up some supplies. Because I’d installed the MDF fascia in this spot prior to final arrangements being settled I had cut the profile of the fascia in a way that didn’t suit the situation I’ve now settled on for the track crossing the layout joint. I wanted to fascia to rise up to just below the sub-roadbed and so the first item on the agenda was to cut two sections out of the original fascia and install two new sections that raised the profile. In doing my final checks I also realised that I’d cut things a little too fine on the inside of the curve and it hung over the edge of the layout right at the point of where the sections meet. I didn’t want to re-cut the curved ply at a larger radius and I couldn’t shift the river module over a centimeter easily so I decided to simply notch the sub-roadbed and disguise it with scenery. I can install a little fence or something here if I think this is a problem but I don’t mind the trains running that close to the edge: live dangerously I say. It would only be a problem if there was a bump in the sub-roadbed and I was going to make absolutely sure there wasn’t going to be one! 🙂 I also discovered that my original cut line for the join was out by a considerable margin (you can see the original line in the photo. It’s the light black line in front of the dashed line labelled “cut line”). I drew a new line and cut the curved ply into two sections so that the edges aligned with the layout edges.

I’ve included this photo without the roadbed to illustrate the pine blocks I cut today to hold the sub-roadbed in place. These will eventually be secured with short lengths of 1×1″ pine screwed into the ply surface of the river.
After a bit of measuring and fiddling about I managed to get three short pine blocks to sit in the spots they will eventually be placed in permanently to hold up the ply sub-roadbed. After I was happy with their length I cut the sub-roadbed and immediately the length of ply sitting on the plain river module dropped below the piece on the module with the scenery. Not only was it lower, it was ok at the far side and was dropping at the point nearest the fascia. So I’d discovered my ply river’s surface wasn’t absolutely flat and that there was a slight drop at one corner. This translated into a bump which would have eventually shown up as a difference in rail height on a curve, at the very edge of the layout and in possibly the most prominent spot in on the entire layout. I could foresee a loco plunging to a concrete floor at some point in the future. A cut another piece of pine, added about 2mm to its length and made the cut 1 degree from square. To my surprise this fixed the problem first go. These blocks are too short to safely adjust by cutting small slices off their length salami style so each adjustment to length required me to cut a new block.
The point I’m trying to make with this post is that I wasn’t surprised by this drop in the roadbed, in fact I’d have been really surprised if there hadn’t been a difference in height. You have to start with the assumption that things will not match up at critical spots like this and then plan with this in mind to get the level of your sub-roadbed as perfect as possible. If you assume perfection on the part of the raw materials that go into your layout the end result will be poor running.
Will things shift and move later and cause problems? Possibly, but I can say from experience that if you have a problem in the sub-roadbed that any shifting will only make this worse. Ignore these bumps at your peril. The idea is not to build layout modules that are so heavy and robust that nothing will move: the plan should be to build a layout where the inevitable shifts and changes are accounted for and accommodated. This is just good design practice, not rocket science. My layout sections are made from at least 4 different materials and these all have different rates of expansion and contraction with possible dire consequences. However in my experience the only time I’ve ever had real problems with track going out of alignment is when the rail couldn’t move a little along its length (on soldered points) or when I ignored bumps in the sub-roadbed. You need to design in and plan for the inevitable movement of your track, not try to build things so that you force the track to remain in place. It will move no matter how robust your framework.
My track is hand laid, I live in a sub tropical climate and the ends of the rail are not held in place with anything other than the track spikes I use to hold all my rail in place. No soldered pads, now brass screws and no built in provision for adjustment. I’ve had this layout for 5 years and there hasn’t been any movement in the rails that have adversely affected running. Don’t ignore the finger gauge.