65 replies to this topic

[Dave Cook]

Whew!  Did all the tube polishing this afternoon.  I went through the #34, 36 and 38 polishes with the matching foam pads on the Bosch orbital sander.  It's REALLY glossy now - check out the reflected things in the photo (click to zoom in).  This process took up some of the first plus most of the 2nd playoff game (yeah I'm the one who doesn't watch much football around here!). Next up will be putting Black 2.0 on the inside of the tube, and then onto installing the inner rotating rings.

 

 

On a slightly related note, I designed and printed a rack for imperial T-handle hex keys.  When you're a home machinist with Chinese equipment everything is metric, but the Astrola project has caused sudden high use of imperial hex wrenches.  All I had were loose L-shaped ones, so I finally got a proper 11 piece Wiha T-handle set.  Throwing them in a drawer is like playing Barrel of Monkeys.  Turns out, Wiha doesn't have a rack for them and you can't find one for an 11 piece set on Thingiverse.  My design has optional magnets in the base or snap-in legs that fit in the recesses for the magnets.

 

Now somebody has to explain to me why Wiha has one orientation of the flats on the hex for 10 of the wrenches, and a 30 degree different orientation for the 3/8" wrench. <*fumes*>. Discovered this too late, so there will be a reprint.  Cad file here:  https://cad.onshape....03f52dd8bfca454

 

Edited by Dave Cook, 24 January 2021 - 11:15 PM.

[Thomas_M44]

 

Today the polishing goodies showed up.  I got sampler packs of the Chemical Guys polishing compounds and their foam polishing pads.  They actually have 7 different levels of firmness on the pads (overkill anybody??) but I figure the 3 pads in the smaller sampler pack should be fine.  The pads have standard hook-and-loop on the back.

 

Man, I hope you're wearing a *really good* respirator that fits well.

 

Metal polishing can be a real mess, and pose a serious inhalation toxicity danger.

 

The polishing compounds can be hazardous themselves, and many older aluminum alloys actually can contain a substantial amount of lead, and/or other toxic metals you *definitely* don't want to breathe

 

I'm likely "preaching to the Choir" about the above, but I simply care, and thought it worth saying, if even only for the benefit of others reading the thread.

 

But yes, on another note: your work is looking great Dave

Edited by Thomas_M44, 24 January 2021 - 11:29 PM.

[Dave Cook]

Actually this stuff is all for polishing the paint, which produces next to no residue compared to the sanding (everything done outdoors btw with my 3M cartridge respirator).  The Chemical Guys auto paint polish is probably too fine to be very effective on metal.  And yeah, the aluminum in Cave's castings is not the modern stuff, no telling what's in it.  I've been pretty careful about not doing anything that would make a lot of fine aluminum dust.  Rocketry folks know that very finely ground aluminum (we're not talking machining chips here) is actually a viable propellant additive, and having a lot of it floating around is an *explosion* hazard.  Powdered aluminum below a certain size is actually a regulated product because of that.

 

Another thing people should know about older telescopes is that lead based paint was not banned until 1978, near the very end of Cave Optical production.  It's a fair guess that the original paint on many older mounts (Cave, Parks, Criterion, etc) would have lead content.  A good reason to prefer the Citristrip, which doesn't throw dust into the air, vs wirebrush or sandblasting.

[Dave Cook]

Staring into the void...inside of the tube painted with Black 2.0 late last night.  It took just about exactly one bottle and one hour with a 1.25" brush to cover the 12.1 ft**2 surface.  I've got a few places to touch up; you do have to brush around in different directions to get the really rough places fully coated.  The surface doesn't seem to be super fragile, there was no obnoxious odor, and water cleanup is easy.  After the 2.0 touchups it will get over-coated with the BLCK 3.0, mostly just for fun and bragging rights - the 2.0 is already very much darker than it was originally.

 

I can't see much if any difference between the blackness of the rougher parts vs the relatively smooth fiberglass patches, so I would say that if you're using this paint, you don't really need to bother with sawdust flocking or anything similar.

 

 

[Dave Cook]

Prototyping a hardware improvement - here's my first batch of tube backing nut holders, intended for attachment of the inner rotating rings to the tube.  I thought about various ways of reducing the stress on the brittle tube, including enlarging the screws from 6-32 to 8-32 (can't do, interference on ring rim, also minimal benefit), doubling the number of screws (2x load reduction per screw, OK but not great), and going to thru bolts with nuts and load spreaders on the inside of the tube (WIN).  The last option is by far the best since it changes most of the high lateral pressure on the edges of the 0.138 hole in the tube (minimum 5 lb per pin) to compression force between the backer and the ID of the ring, spread over the area of the tube backer, which is about 1 square inch.

 

The 3D nut holder design was pretty straightforward and follows the construction pattern of many 3D printers.  If you size the hex recess exactly for the dimensions of the hex nut, you get a nice light press fit that can be done by using a screw and washer from the other side, and screwing it down until the nut is pulled into the recess.  You can just about see that the nut holders have a 9.375 diameter curve on one side, so they’ll lie down perfectly on the inside of the tube.  They are only 0.225" thick so they won't get into the optical path at all.

 

The only slightly tricky part is reaching them into the middle of the tube for installation.  My plan is to attach a magnet to the end of a piece of aluminum bar or angle that will hang onto the nut until you get the screw installed, and then detach easily and cleanly.  The actual installation is going to use black oxide stainless nuts n bolts, and the nut holders are going to get painted über-black.

 

To use these, you have to bore out the threaded #6-32 holes in the ring to the #6 clearance diameter with a #25 drill.  This will get done on the mill/drill and will let me straighten out the one that is rather skewed.

 

There will be other versions of the nut holders for 1/4-20, M5 and any other screw size that is used to bolt things to the tube.  I've already got everything parameterized in the CAD file so this ought to be super easy.  CAD file here: https://cad.onshape....413d73b9dce4f6d .  To print these you need to stand them on one edge and use 100% infill for strength.

 

 

And here's the final debugged version of the Wiha imperial T-handle hex wrench holder.  The snap-on legs work great.  I've got a metric version printing overnight tonight.

 

 

[Dave Cook]

Overcoated the inside of the tube with BLK3.0; it's noticeably more absorptive than the first coat of 2.0.  With one of my garage spotlights slanting in the top of the tube, you can't see anything in a 2nd reflection.

 

 

And tonight's big push...rotating ring mounting success! Positioning them on the tube left a bunch of black marks but no real scratches - everything polished off easily.  I ended up drilling out the holes #27 for a closer clearance fit.  The nut holders work perfectly and it actually wasn't hard to get everything aligned for smooth operation.  I did need to get my daughter to wield the screwdriver while I reached down the tube with the nut holders.  The magnetic holder idea was a fail because I forgot stainless nuts aren't very magnetic.

 

I think the outer rings look great finished in polished aluminum instead of painted grey like they were originally.

 

No idea what I'm going to work on tomorrow.  I need to polish the tube end rings, shop for a longer travel focuser, make nut holders for the finder and tube counterweight brackets, figure out where to put the Vixen bar, work on the mirror cell, dismantle the dec axis...it's a target-rich environment.

 

 

[Dave Cook]

Some progress made on things I mentioned in the last post.  I got one tube end ring polished, worked on the dec shaft some with a Scotch-brite pad, did some focuser shopping (no decisions yet), modified the nut holder design to have another parameter for tilt on screws that aren't normal to the tube, printed and read up on the OnStep control system.

 

My focus for the next few days is going to be on getting the OTA reassembled, and to take it out at night a few times to enjoy the optics before fully demolishing the EQ head.

 

Link to the improved nut holder in OnShape:  https://cad.onshape....952ab4f69cde398

 

It's fully parameterized now, you can generate a new one just by plugging in the various dimensions.  For #hexsize you should use the exact dimension of the next nut across parallel sides; that will give you a mild press fit.  The depth should be 5-10 mils more than the actual thickness of the nut.

 

 

[starman876]

You will love the images through this scope.  Well worth the time and effort you have put into the restoration.  

[Dave Cook]

Well, I'm going to have to wait a bit for the quasi-first-light view thru the OTA - realized I have to take the outer rotating rings apart again for some measurements and modification to add the Losmandy rail on one side and a dual-width saddle on the other.  Along the way I did get the finder rings and old William focuser mounted with freshly printed nut holders; #10 and #12 sizes with 5˚ toe-in on the screws.  Between the nut holders and machining the feet, the finder brackets are non-wobbly for the first time ever.

 

On a really good note, I just scored an in-stock AT 80ED from our CN sponsor.  That was on my shortlist for the guide/aux scope.  I've been checking for moderately priced 80 APO doublets every few days for a month and a half with no luck until now....pulled the trigger in about 60 seconds.

[Dave Cook]

Working on the saddle to dec shaft interface since I have a blank cut and am ready to start some machining on it.

 

This photo shows the bottom of the original Cave saddle where the dec shaft plugs in.  There is quite a bit to take note of:

• The non-concentricity is real.  The 1.5" saddle bore is laterally off-center by about 0.10" in the casting.  The walls of the rim are about 0.587" on one side and about 0.684" on the other.  This can also be measured on the other side by looking at centering of the bolt hole.
• There is a smaller ~1.23" bore that extends a little beyond the main 1.5" bore.  The shelf between the 1.5" and 1.23" bores forms the bearing surface for the end of the shaft.  Due to the sloppy fit of the bore, this little surface completely defines the relationship between the dec shaft and the tube, and has to be clamped very tightly with the 1/2" bolt.  This surface is not even remotely flat, so it was not done with a boring bar.  It is also non-concentric with the main bore, for which my best guess is that a regular 1.5" drill was used.  The good news is that the slope of the shelf looks to be uniform.
• On the other side (not shown), the clearance hole for the 1/2-13 bolt is counterbored only about 0.30".  The standard head height of a 1/2" hex bolt is 0.50", so the bolt protrudes about 0.20".  That works OK because the saddle arms reach upward, so there is clearance between the top of the bolt and the tube.
• There's no good way at all to statically measure on the casting whether the tube is on the plane perpendicular to the dec axis, because there is no machined reference surface defining the tube position on the saddle.  The curved saddle support surfaces are wildly inaccurate so you can't use them as a reference.
• There is just about exactly 1.00" of shaft engagement in the saddle, and the shaft is 3.5 to 4 thousandths undersize.  I think I mentioned before that that's a pretty coarse fit in the machinery world and is not considered a good location fit, which would have the shaft somewhere between a half and 1.0 thou smaller than the bore.

None of this probably matters too much for visual use, except that it makes offsetting from one object to another using the setting circles not very accurate (and I have memories that this was the case).  After seeing how this is all set up, I imagine there could easily be a degree or two of perpendicularity error.

 

The second image is a CAD drawing I made showing the idealized original Cave shaft interface, to help me design the new one.  You can see all the dimensions in the left column.  As usual the OnShape CAD file is public here: https://cad.onshape....59b590c09b4186f

 

I'll publish the new design after I finish up a bunch of details.  It's going to use brass setscrews onto a ground flat on the shaft to get a much more definite location, along with a proper 1.5" spec 304 stainless shaft (+0, -15 microns), and maybe double the engagement, which will also create room to use a tucked-under dec motor with pulley drive.

 

Edited by Dave Cook, 13 February 2021 - 09:13 PM.

[macdonjh]

Very cool.  Thank you for posting all of this.

[ngc7319_20]

 

• The non-concentricity is real.  The 1.5" saddle bore is laterally off-center by about 0.10" in the casting.  The walls of the rim are about 0.587" on one side and about 0.684" on the other.  This can also be measured on the other side by looking at centering of the bolt hole.

Great stuff!  Thanks for posting!

 

I'm not sure any of these non-concentricities / offsets in the saddle will affect anything.  Probably the most important alignment will be the orthogonality of the RA and Dec axes.  I haven't taken mine apart, but I assume the Dec. housing / RA shaft joint is probably similar to the casting / machining on the saddle.

[Dave Cook]

The offset doesn't actually hurt, but I think the tube-to-dec-axis orthogonality may be kinda far off in the original setup.  So many things lack precision that there's really no way to tell.  Visually it affects star hopping over considerable distances, as the polar alignment will never seem to be very good.

 

Anyway I'm working on making it better <evil grin>.

 

Face milling the new dec shaft interface block with the big fly cutter.

 

 

Photo below shows checking out the parallelism and flatness of the faces with my tool height setter rig (with its tiny surface plate).  The digital indicator is kinda cheesy but if you threaten it enough you can get it repeatable to maybe half a thou.  Overall the surfaces came out good to +/- 0.5 to 1 mil...way WAY better than I expected...did extra measurements because I was convinced the indicator was frozen.  I did do a careful job of setting the X tram on the tilt-head mill to within a thou in 6 inches, but the Y tram on a column mill is basically non-adjustable.  Fortunately the factory seems to have done a decent job.

 

I have a Losmandy style saddle on the way from ADM that will bolt on top of this block, and a matching D-rail that will bolt to the outer rotating rings.

 

A rough calculation is that if I can hold the alignment of the shaft in the block within .001 over a 2.0" engagement, the angular error will be about 1.7 arcmin.  Likewise if the mounting face for the saddle is within .001 in ~6" from perpendicular to the bore, the error there will be around 0.6 arcmin.  How they combine depends on the sum of the error vectors but it seems like mechanically we might achieve 2 arcmin before considering the ADM D-saddle and rotating rings.  Overall I'm pretty confident that things will be a lot straighter than before.

 

 

Oh yeah, my AstroTech AT80ED aux scope showed up from Astronomics. (Don't shoot me for posting a non-classic scope but imma gonna bolt it to the Cave, so there).  I couldn't resist sticking it on a tripod and taking it outside for a look at a few things.  I practically fell into the Panoptic 19;  I think it's better than my eyeballs.

 

Edited by Dave Cook, 19 February 2021 - 12:34 AM.

[Dave Cook]

Got back around to the mirror clips.  Machined the slots and wet-sanded the factory mill marks off the extrusion.  They've joined the stack of things ready for the Electric Homer Bucket of Anodizing ™.  I've recently collected all the bits n bobs needed to anodize and dye small aluminum parts...weirdest Amazon order ever with lead sheet, ping pong balls, lye, an aquarium air pump, titanium wire...probably got me onto half a dozen watch lists  Hoping to try it out over the next couple of weekends.

 

 

 

[Dave Cook]

Here's the first complete rev of a new dec shaft to OTA interface, replacing the original saddle.  Drawing may be grabbed here: https://cad.onshape....04819296ad0488c

 

This uses two 3/8" setscrews to hold the shaft against the side of the precision machined bore.  The shaft will have a setscrew flat to make the positioning more definite, and a 5/8" bolt in thru the end that is really just a safety feature to prevent the OTA from falling off the shaft when past the meridian. 

 

It's not shown in the drawing, but the setscrews will be in 3/8" stainless steel EZ-loks so that we won't be turning against tapped aluminum when adjusting them.

 

The end bolt is 5/8" rather than the original 1/2" mostly because that's the smallest tap size that Misumi has in their standard builds for a 1.5" shaft. I wish I didn't need this bolt because it makes the interface block 0.5-0.6" taller than it would otherwise be, but the bolt is the positive retention for the OTA and thus is not optional.

 

The new shaft will also be 30" long (longest available as standard order) instead of 27" to help the counterweight with the heavier OTA and taller interface block.  And I'm ordering it with the 5/8" tapping on both ends so that I can bolt on a counterweight safety retainer disc at the other end.

 

Last thing I have to do before ordering the shaft is make sure that the dec bearings in the casting will actually accept a true 1.500" shaft in place of the undersized 1.497" original.  I'm fairly confident.  Fairly...

 

Now that the design is done I can go do some more actual machining on this block.  Logistics have been against me this week; I just had to order a replacement for a boring head order that went missing in the caverns of the FedEx Nashville hub (rare), and the ADM brackets I need to finish the OTA have been rattling around somewhere in the USPS, which was helpfully telling me for the last few days that my package is "in transit" and "late".  [EDIT] The brackets have suddenly reappeared in USPS tracking and might actually arrive today.

 

Edited by Dave Cook, 26 February 2021 - 07:46 PM.

[Dave Cook]

Knobs, saddles and rails, oh my!

 

Hand knobs for bolting the mount head to the pier.  I will be doing this frequently and don't want to mess with Allen keys in the dark.

 

 

Really big knob for 5/8" stud to hold a counterweight safety retainer disc on the bottom of the new dec shaft.  It weighs over 500gm and is a counterweight in itself.  I could not find a SS knob with a shoulder over 1.5" dia. to act as the retainer itself.  Instead I'll just lop a cookie disc off some 2" stock and put a 5/8" hole in it.

 

 

Combo saddles, radius blocks and a 15" long Losmandy style rail from ADM.