I have been doing some work on reverse engineering the RA rear brake based on a few original pieces, field observation and measurements, and lots of photographs of the few surviving examples. I have most of the details worked out, though there are a few questions remaining. Considering the rarity of the subject matter they may never be answered. But nothing ventured, nothing gained!
The following are pictures of the replica Dixon outfit RA build by Bob Thomas.
One of the problems is even with the few examples extent it is clear components evolved over the course of the few years the RA was in production. Most of the changes seem to have occurred between the 1923 season when they were essentially available to the Works teams and the 1924-25 models available to the general public.
One of those changes appears to be the operating arm. The presumably earlier version is shown on the left in the next image. I say presumably because it they type fitted to the one machine known definitely to date to 1923. The one on the right is yet another original example, and matches what is shown in the RA handbook. This then is presumably the 1924-25 version.
One glitch is the eccentric pin that I have is threaded only on one end, the other end being plain (B). And the threaded end is not long enough to pass through the frame lug (A). Reference the next two images. Two other examples of eccentric pin that I have seen are headed and nutted on both ends; the end passing through the link must not have been done up tight so the linkage would still be able to pivot. The eccentricity allowed some adjustment to the contact angle of the shoe to the friction disk. It is possible the eccentric pin I have is for an OB model.
Since the above was originally written for the NCR, I have had some field dimensions sent in for the rear operating arm that attaches to the taper square. This has been added to the preceeding image. The change in background is due to a software upgrade during the interval. The example copied is bent backward to presumably give a better angle of pull for the brake rod. The very few period photos where you can see the brake arm shows it to be straight, as in the following image of the 'bare' operating arm. The wing nut has a pair of nubs that engage slots in the operating arm every 180 degrees of rotation, to prevent the adjustment from slacking off on its own accord. The slots for same can bee seen in the spherical cup end.
The OB uses a similar parallel linkage, but it was a more conventional brake shoe block working in a dummy belt rim, as did the CW model. Also the operating arm is on the inside of the rear stay tube.
A frame repair has to be carried out to replace the brake anchorage lug. If anyone has a junk CW frame that a lug (see next image) could be salvaged from, it would save a fair amount of work.
The lug is not identical, as it sits on top of the frame where as the RA lug wraps fully around. But I think it easier to add the missing portion verses carving the whole from a solid billet. See next.
Other than where it wraps around the frame they appear to be identical. Better yet would be to find a donor OB frame, which does appear to use the same lug as the RA. But OB frames are nearly as rare as RAs, and finding one no longer fit for service to cut up is unlikely compared to finding a rotten CW fit for cannibalization!
–Doug
