Updated - 9 September 2020
I have bought a 3D printer as described here. One of my first trials was fishplates using the drawings in The Lynton & Barnstaple Railway Measured and Drawn. These fit Peco SM32 rail and are shown below.
However the gap created by the spacer together with an additional gap for expansion does not look right so I modified my design to produce more conventional slide on fishplates as shown below:
and in situ:
I have been creating 30ft track panels from Peco yard lengths as shown here:
There are 12 sleepers per panel, which is correct. The rail ends are 16mm from the centre of the end sleepers which is also correct. However the rail length is only 448mm instead of 480mm, as it should be, and the sleepers are at Peco spacing. I think this is an acceptable compromise allowing 2 panels to be created from each yard of Peco track.
Having laid a few panels and tried ballasting I am now cutting off the webs between the sleepers of the Peco track. This makes it easier to curve the track and lay the ballast with a nice gap under the rails. I have made some templates to help align the sleepers and maintain the track curvature as shown below. The track was drawn in Templot, the template stuck to some plastic sheet and then cut along the inside edge of the rails. The tracks forming the loop on the curve at Chelfham are spaced at 6 1/4" centres so I created templates for 9', 8' 5 3/4" and 6' radius, and straight.
The track is fixed with a blob of Evo-Stik Strong Stuff Waterproof Adhesive under the end of each sleeper.
Finding something for ballast has been a problem. I used bird grit in 7mm but that is too fine for 16mm scale. I have several packs of pond grit but they don't have enough brown in them and they are a bit too shiny. I think it was road stone that Peter Mann kindly gave me. He had already sieved it for 2mm and below for 7mm scale ballast. I tried various sieves and settled on the following 5mm mesh from eBay taped to the bottom of a plastic cheese biscuit box. This may be a bit course but it was a compromise between volume recovered and looks. Some pictures of the L&B show the ballast to be a bit course and sparse, stopping at the sleeper ends.
I am now experimenting with the cess and between the tracks in the loop. I have some fine sieved coal dust but it is a bit glossy. I hope to try some ash next time someone has a bonfire.
The ballast and cess are laid dry, sprinkled with water, then Ronseal outdoor clear matt varnish followed by more water all using a pipette. I added a bit of washing up liquid to wet the coal dust as otherwise it gathered into blobs.
I cut some foam from packing material to level the gap between the concrete bases of the loop at Chelfham. I don't know if it will be good enough to stand the test of time and weeds. We will see.
The results can be seen here.
The ballast described above was a bit coarse and did not stay stuck very well. I liked the look of the ballast on the Forest Green Railway described in 16mm Today no.174 and 175 by David Dawson. David kindly advised me of the product he uses which I have now bought here:
https://www.themodelcentre.com/jfg7-javis-javis-7lb-fine-ballast-chips
Here is the result on the main platform road entering Lynton:
You may notice the cess in the middle of the picture above. That was ash kindly given to me by Rob Wigley from his log burner but I don' think it is dark enough. I have some ash rescued from a barbeque found next to Rutland Water but that needs crushing before I try it.