Recently I've got a bit fed up with not having a totally level continuous run, so when I've got a really long train that a loco can pull on the level or downhill, it gets really frustrating when the loco slips to a stop going uphill.
So, a few weeks ago I decided to build such a level line, branching off at the top of the hill, and continuing on the level until it reached some convenient location where it turns back by means of a 180 degree curve of 2 ft. 6 in. radius.
I also decided to experiment a bit in the construction of this loop. Firstly, all the track is code 83. Secondly, the junction from the main line was to be the largest radius points I could get my hands on without having to construct them myself. So I ended up buying a couple of Shinohara code 83 No.10 points (frog angle 5 degrees, 43 1/2 minutes) from Scale-Link Ltd (see
One thing I discovered when building this is that Shinohara code 83 and Peco code 83 rail have different cross-sections, and Peco fishplates won't fit on Shinohara rail without opening the fishplates up somewhat (I used a screwdriver for this).
I started re-testing some of my locomotives and so far I've come up with some startling results. In these tests, the wagons I used are made by Dapol, and have had their weights removed and their tension lock couplings replaced by Kadees. These 'weightless' wagons tip the scales at 30 g. as opposed to 35 g. for the unmodified wagons. This might not appear much, but 7 of my wagons weigh the same as 6 of the originals, which helps locomotives pull longer trains.
The Bachmann 56XX class 0-6-2T will pull 70 wagons, and is only limited by the friction caused by the sharpness of the 2 ft 6 in radius curve.
The Bachmann 9F class 2-10-0 will easily pull 100 wagons. I decided not to try it with more wagons, as the weight of the train was causing the couplings on some combinations of wagons to part.