9.10.15

7" f/12 iStar refractor 20: Focuser backplate improvements.

*
The previous back plate was the base of a saucepan with most of the body cut away and pressed inside the main tube. It was a very good fit but left the raw, cut edge on the steel tube looking very unfinished.

A search of numerous charity shops provided three more saucepans which all fitted nicely on the exterior of the main tube.

The first task was to bore the chosen saucepan base for the big, Vixen focuser. The aluminium from which these saucepans are made is so soft that it makes very hard work of machining. The tip of the tool rapidly builds up with aluminium making further cutting impossible. Which required frequent cleaning before another cut could be taken. Nevertheless I succeeded eventually. Ideally I should have chain drilled inside the cutting circle and broken the disk out. Finishing off to size in the lathe. Except that I had no guide to to make the circle concentric unless I centered the pan in the lathe first.

The next problem was cutting off the saucepan rim to make a shallower end cap/cum focuser back plate. Turning such a wobbly item proved impossible so I had to hacksaw a strip from the rim of the pot. Then turn the ragged edge straight.

All very noisy because the pan rang like a bell producing chatter marks where it cut. Still, beggars can't be choosers and it was I who chose to use these pans for their long term, structural integrity and neat appearance. Though the latter is, admittedly, very much a matter of taste I find the look easily suits my ATM needs. Had I used the router to produce a solid end cap out of birch plywood it would have taken just as long and would still have needed painting for protection. 

The first image above is of the complete pan, with handles removed, after boring for the 2" Vixen focuser. The second image shows the back plate fitted in place on the OTA after cutting down the height and smoothing the finish. Scotch-Brite abrasive fiber is good stuff for this task. Abrasive paper tends to scratch badly, scores the very soft aluminium and dulls rapidly from the build up of powdered metal.

I may use an angle grinder on the tube seam where the end cap fits to avoid a gap either side of the seam. This gap might collect dew running down the tube during normal observations and cause problems over time. Storing the OTA on its nose, for minimum footprint, would tend to cause any trapped moisture to run down towards the objective. Removing the seam will allow a tight seal to be more easily achieved between the main tube and the back plate. I shall have to align the back plate carefully to achieve collimation before fixing it in place. I thought I might fix a disk of birch plywood inside the main tube and then use push-pull screws for back-plate/focuser collimation.


I decided to bolt the longer dewshield over the stumpy original one to allow their separation in future. This will also prevent the long one from sliding over the other in storage. The two nested dewshields will help to reinforce each other when supporting the OTA. I then painted the new dewshield and back plate insides with matt black. I have fitted the objective cell collimation push screws with Nyloc nuts to stop them sinking into the soft, dewshield-countercell metal. The nuts will greatly increase the surface area of the screw tips and help to reduce distortion of the thin metal. Failing that I shall reinforce the dewshield base ring with metal strips.

Rather than screw the smart new back plate directly to the main tube I routed and turned two 18mm plywood rings. They are dimensioned to fit tightly inside the main tube and have been bored to match the focuser. With both rings screwed firmly together they were pushed into the main tube to just clear the back plate. The second, exposed ring, has been radiused in the lathe to match the inner curve of the back plate.

I also drilled the plywood rings for the three baffle studs [screwed rods] to stop the baffles slipping down inside the main tube over time. The back plate will be fitted with push-pull screws working against the plywood rings for precise collimation. The slight increase in weight of the plywood rings will further aid tube balance.

The two images just above show the focuser back plate held by long screws with springs for collimation. The metal of the back plate is far too thin to take an M6 thread. Once the plywood rings were fixed into the main tube it became a nightmare to fit the back plate. The springs made it very difficult to seat the back plate far enough in to start the screws in the T-nuts. So I removed the springs and screwed the back plate firmly to the plywood rings for the moment.

I will now try to find shorter and stronger springs. I couldn't obtain any M6, pan-head screws longer than 60mm. I also fitted Nyloc nuts to these screws to ensure they were not pushed back out during back plate fitting. All three screws had to go in squarely and simultaneously to ensure the back plate was rotated correctly on the plywood rings. The problem was that I could no longer see the screws when the back plate was offered up to the plywood rings. Had the screws been longer I could have guided them into the pre-drilled holes.


The weather continues to be continuously cloudy so I have yet to test the new telescope's performance.

I became dissatisfied with the occasional stickiness of the rotating focuser. The saucepan base which I used for the back plate was not very true and accentuated itself by sticking on the radial screw heads. I have now turned a ring in 12mm birch plywood. This fits snugly on the Vixen focuser base and presses evenly on the inside of the back plate at a much larger radius. The focuser is now free to turn without the previous rocking. Though it did require three holes to be drilled and tapped M4 in the Vixen back plate adapter to allow fixing screws to secure the counter-bored ring. The image above shows the general idea before I shortened the securing screws. The matt black paint has suffered from all the rebuilding and dismantling. The plywood ring will need to be painted matt black too. Though it should lie well outside the light cone from the objective.

I also found a plain, brass porthole in my scrap brass collection which suits the 8" main tube diameter perfectly. I have already removed the glass retaining ring and bored the housing slightly larger to suit the Vixen 2" focuser base. There was a surprising similarity in the focuser base and the aperture in the brass casting. I haven't decided yet how best to use the brass disk. It can easily be screwed to the existing plywood disks but ideally needs some sort of edge trim. Which might spoil the appearance if it is not reasonably traditional.

The heavy 8" diameter, brass disk would further aid tube balance at the cost of more weight to lift and carry. It had been knocked about in its long life so badly needed to be smoothed and polished on the visible face. It is all but impossible to chuck such an object to take a facing cut in the lathe. I tried repeatedly using both the 3 and 4-jaw chucks but the disk still had a slight wobble when spun in the lathe. No point in thinning it further just to make it pretty. So I relied on coarse abrasive paper followed up with fine wire wool. Such "found" objects are often useful in telescope building when they actually fit one's needs.

This image shows the final backplate assembly with a quick coat of matt black paint.[Not yet dry and it will never be seen so the finish is irrelevant.] The focuser now turns smoothly without any shake. I could still add springs to the fixing screws if it proves worthwhile for achieving consistent collimation. Thanks to the Nyloc nuts there is a degree of push pull in these long screws. Always provided I don't push the T-nuts out of the plywood fixing rings in the main tube! A drop of hot glue on the edges of the T-nuts is often used by speaker builders to avoid this same problem.

I have deliberately collected brass shell casings and tubing in various sizes when they were dirt cheap in flea markets. Tech colleges sometimes scrap useful lengths of solid brass bar up to quite considerable sizes if one is lucky enough to find them at a scrap yard. I suppose the tutors don't have to pay for the brass out of their own pockets so place no great value on it. Even if my brass and alloy scrap eventually ends up as dewshields or objective cells they cost next to nothing compared with buying new materials or commercially made machined items. I even bought a scrapped, fiberglass street light pole [for small change] for my 5" refractor. Though it proved to be too heavy and much more difficult to mount straight on the saddle without very different sized tube rings. It's all a matter of seeing the potential without becoming an obsessive hoarder.

Click on any image for an enlargement.
*

No comments: