An update with something to show. No, the fixtures aren't done yet but chip will be flying this Thursday.
To start, there are 32 rail gripping/ positioning fixtures for the range of Code 40 and Code 55 turnouts, 9 each from #4 thru #14, ignoring #11 and #13. Then 3 Frog soldering fixtures with 3 sizes per fixture, and a precision rail bending brake with adjustments for precise angle and precise position. There will be others made in my manual machines for assembly, hinge soldering, etc, for all of the little detail stuff and assembly rigs. I'm getting anxious.
Meanwhile, I made a rail grip fixture set to test the design approach for the tiny code 40 rails and also on the Code 55. This will be used for frog and especially closure rail preps where the angles get very shallow and a thin point is most desirable but difficult to accomplish with the rail head and foot overhanging the web. It's one of those things that has huge effect on the rest of the design and if not executed pretty near perfect results in gauging fluctuations, the need to relieve the stock rail head for the fatter point as is common for most commercially produced turnouts, or jockeying the nose with a taper inside and out, trying to get the nose slim enough, but then ending up with a gauge fluctuation by whatever the inside face taper is.
What I've measured on Atlas turnouts is .015" point at the closure rail point and on some hand builts I have that were made using a point filing tool they are .010" but also had a bit of an inside taper at the very point, showing up as a slight taper at the very end, getting a little fat within the first 1/8" of the stock rail contact point. Meanwhile, that closure rail from the point all of the way to the heel block (hinge) is supposed to be perfectly straight, holding gauge with the opposing straight stock rail. Any bends in either one at this point show up in gauge fluctuations. With some tweeking and work with the gauge this can be adjusted somewhat and kept within a generous NMRA tolerance for this position in the turnout. But it's still fluctuating wider or narrower. This is where the wide treads on our models are saving the day, taking up the non-prototype fluctuations. For all of the turnouts that work well, the real tight tolerances are at the frog and guard rail positions but the more accurate the turnout is throughout its many dimensions, the smoother it will run. Those who have hand laid and with some practice and care have already experienced this.
With that said, my point rails will be/ are all machined, accurately positioned in the machine and maintaining perfectly straight cuts, not filed with all of the inaccuracy that introduces, even if tolerable, and with a full, straight, vertical point with the finest point that can withstand use. Also, while the stock rail foot will need to be relieved there is no filing of the mating stock rail head to tuck a fat point in to get a functional dimension. (There's more reason for this too but enough for now.)
The machined test piece pictured below is for a #4 turnout. The point is .002", about as fine as it can be while still being strong enough.
The plan is .003"/ .002" as the standard point thickness dimension. It is aligned perfectly straight and parallel with the opposing stock rail, holding extremely accurate gauge throughout the assembly, and fully supported as the web is
not removed. The rail starts the process by precisely bending the point end of the rail outward to move the face of the thin web outward in the direction of the gauge face of the rail, and then the now bent rail head and foot are precisely machined back to the original straight plane but now with the web fully supporting the head vertically. With the rail head now back to perfectly straight on the gauge face and with the web fully supporting the rail head, the rail is placed in a second fixture where the angle is precisely machined from the back side at the proper angle and to the extra fine point that the solid web can support.
This process will be the same for the frog points which allows them to be very fine points. With that fine frog point it will allow for less unsupported gap in the rail transition across the gap left where the flanges pass on the opposite route. That gap is an unavoidable turnout design problem (unless you make a movable frog point)
and it gets worse as the frog angle increases and is further aggravated by a less than sharp frog point. By prepping the end so that the web is supporting all of the way to the point and then precisely machining the components this gap can be kept to the minimum possible for each size.
Here are the pics of the Code 40 machined test part. The ones being held were photographed through a magnifier light to get a view of the .002" point.
[ Guests cannot view attachments ]
[ Guests cannot view attachments ]
