When it comes to the truss tube inserts and the adjustment of collimation on a hexapod, I'm participating in some some relatively new experimentation. There are several ways the ends of the tubes can be fastened, and it appears the most popular method so far on this forum is to use heim (rod end) joints. On the hexapod thread Mark Cowan has shown he'll be using heim joints, but with two hold down knobs and bolts per pair for a robust way to keep the spheres of the heim joints seated firmly. Other promising methods shown on that thread are variations on using spheres, and a couple of options Carl has suggested would eliminate Jonathan's concerns about the spheres themselves turning, but may require more custom machining than most can accomplish at home. As Jonathan demonstrated with Merope, looseness and slop must be pretty much eliminated, but adjustments must be possible with moderate torques on the poles so that collimation can be tweaked by hand.
I'm going to attempt, at least on a first pass, to use simple eye bolts salvaged from turnbuckles in place of the heim joints. I'll be trapping the eye bolt heads in the radial direction of the optical tube assembly with angle aluminum "v-blocks", and I'm experimenting with a couple of methods to trap them consistently in the perpendicular or tangential direction around the base. I need to be careful to use a method that, like Merope, allows for the tubes to be bundled when the scope is broken down.
Regardless of the actual ends of the trusses--be they heim joints, spheres, or turnbuckle eyes--the threaded sections of the assembly must be tight, but not too tight in the tube ends. Too loose and collimation may shift dramatically as trusses go from compression to tension, whether from altitude adjustments or from the torque induced by the user in aiming and tracking the scope. Jonathan offered one solution on the first page of the hexapod thread, in which Nyloc nuts are potted with epoxy into the tubes. This is a very reasonable way of eliminating the slop, however it does require using both left hand and right hand Nyloc nuts. These appear to be much more available in metric rather than inch sizes. Of course, coarse and fine right hand threads are another option to keep everything right handed, although for me I think the adjustment would be way too fine. In this case the effective pitch is the difference of the thread pitches rather than the sum. In my case of using left and right handed threads, 1/4"-20 threads would each give 0.050" per turn, or a tenth of an inch total per revolution. Across the diameter of the scope I believe that will be plenty fine enough for small angular adjustments.
I experimented with making tightly threaded inserts with my Red Stuff by partially tapping through with a taper tap so that the threaded eye bolts would cut their own threads somewhat. I discovered that there is a very fine line on tap depth between producing inserts that are either too loose or too tight. Because of that, I'm concerned that with the slightest wear a formerly tight fit may loosen fairly quickly over time. This Red Stuff tooling board is similar to Delrin for hardness, but I don't have any Delrin to experiment with to see if it would do any better. Nylon compresses much more, and I can see why it is used in Nyloc nuts for such a purpose. I fear that if nylon were used for the entire insert, though, that it may be a bit too flexible and yield too much (elastically) under the varying loads in use.
I'm choosing to make inserts out of Red Stuff that is tapped completely through with 1/4"-20 threads, and I have obtained left handed taps for the six inserts that will be left handed. Surprisingly, I got a set of three taps, taper, plug and bottom for less than $15 delivered. Cool! To make the bolts tight, and to give me the ability to maintain this tightness over time I'm cross-drilling my inserts and installing some nylon set screws into them for the bolt threads to cut into their tips, essentially making my own Nyloc nuts. I have a bag of 100 1/4"-20 nylon set screws that I bought cheap online a few years ago as hardware for a possible lightweight string telescope. If these wear significantly they could be either turned in deeper against the bolts or simply replaced.
I've already turned my inserts using woodworking equipment as shown here:
The only critical dimension is the diameter that fits into the tube. This was easy to sneak up on using the outside caliper in the above photo. Note that this minor diameter has angled undercuts similar to, but reversed in direction of, eyepiece undercuts. The purpose of these is to provide an anchor for epoxy to adhere to within the tube. I know from experience that this urethane material epoxies very well, and this may be a significant material advantage compared to Delrin for this use as an insert. Carl also pointed out earlier in this thread that this material is also slightly less dense than Delrin, and less than half as dense as aluminum.
A dry fit of the inserts satisfied me:
The cross hole is through the insert and was made by a number 7 drill in anticipation of being tapped for the nylon set screw. Because it is outside of the aluminum tube I could service it by extracting the set screw if needed and replacing it. I doubt I'd simply tighten it because the eye bolt tends to really mutilate the tip of the screw and it would have to be advanced a full turn for the threads to realign with the eyebolt. A bit of experimentation has me satisfied that if the two nylon set screws 180 degrees apart are threaded just inside the minor thread diameter of the axial hole then the result is just enough friction.
I tapered the ends of the inserts to allow the eye bolts heads to be closer together, better approximating a true truss with triangles on the scope. One disadvantage of using turnbuckle eye bolts as compared to left handed and right handed rods is that I'm limited by their 2" of threaded length. I can't have them extend about 2" out of the inserts or they would unscrew completely. On my scope the included angle between adjacent trusses will be almost exactly 20 degrees, so I tapered the conical section about 10 degrees per side. This allows for screw eyes to be inserted almost completely into the inserts.
The following photo shows my sequence of operations for making these inserts:
From left to right you see I table sawed the board into square stock, then table sawed corners to make them octagonal, then drilled axial holes, then drilled cross holes for the set screws and saw cut the ends for the lathe spur to engage into, and then finally turned them complete with a skew chisel and a parting tool (see first photo on this post).
The two shorter inserts in the top of the above photo have already been installed into a cross brace tube between my altitude bearings (see first photo on this thread). These will have thumbscrews to hold them in place in use, but for now simply have 1/4" bolts.
Here's a photo showing how I drilled the axial holes. They came out nicely close to on center, but the holes were used to center the inserts on the lathe anyway, so the holes became concentric with the exterior profile when turned:
Edited by jtsenghas, 26 January 2019 - 10:13 AM.