Ad Astra Observatory

The NJP’s state when it got back from “repair” was pretty sad. I mean, even a poorly working NJP is better than I’d ever really gotten with the G-11, but it needed some TLC, badly.

Before I could get to work on the mount, though, I had to put the observatory back in order. I had never really cleaned up after the trip to GSSP last year, and stuff was really starting to be piled up, making it tough to move around, let alone do intricate mechanical work.

So I spent some time getting stuff put away and horizontal surfaces cleared and things moved to the walls where possible. It’s not totally spic and span, but it’s looking a lot more like normal now.

So. The mount. I dismounted Veronica, pulled off the saddle and adapter, and started looking at the Dec axis first. I couldn’t really tell what was going on right away, so I kind of just started ripping things apart until I figured it out. I found out that the gear attached to the motor actually has a little play in it, which means that I was not going to be able to get all the play out of the Dec axis. But it was pretty minor, so I decided that I’d just move forward.

I put some thin paper in between the gears (I used party streamer, because I couldn’t find my Zig Zags). I pulled out my feeler gauges, and made sure that both pins were seated the same amount, with the gears meshed. It took a couple of tries, but the gear mesh is now nice and tight, both bracket pins have the exact same spacing, and, wonder of wonders, when I put the gear cover bracket on, I was able to adjust it so that it no longer scrapes! While I had it apart, I noticed that someone had taken a file to the edge of the cover. Not really a normal method of repairing this part; I could have done that myself if I’d known that was the method to be used. I think I could get a little more clearance, by holding the bracket in a different position relative to the mounting posts as it got tightened up. It’s clearing now, so I’m not going to worry about it.

The adjusted Dec axis feels really nice; I powered up the mount and confirmed that it would run in both directions.

With the Dec axis completed, I stopped to think a bit about the RA axis. The RA axis came back to me with a lot of slop. I really wanted to take out the backlash, but there are a lot more moving parts involved in getting the RA axis diagnosed. As I’d already had the axis mostly apart before, as I tried to figure out if I could adjust the reticle myself. I had decided that the reticle was too complicated to mess with, but this time, I thought I might be able to find out what was wrong with the backlash.

It turned out to be a pretty simple repair.

First, I unscrewed the rear cover, which revealed the main axis cover. The main axis cover is mottled black, and has 4 set screws that hold the machined brass RA axis in place. With the setscrews loosened, the main axis cover unscrews, to reveal a conical bearing, conical bearing seat, and a few large washers. From my many times messing with the G-11, I realized that, at this point, I could probably just pull the RA axis out.

This suspicion turned out to be well-founded.

I found a soft place for the Dec axis (with RA shaft still attached) to land (this needs to be thought-out, because the cable harness is still attached), and out it came.

Doing this revealed the main RA gear, the upper conical bearing, and a thrust bearing setup (washer-bearing-washer). Under the main RA gear, there is a “ring” of metal that gets squeezed by the axis lock. That ring’s housing attaches to the Dec axis housing via a very large machine bolt.

With the gear train of the RA axis all apart, I started hunting for sources of backlash. There were essentially 4 places that could have been problematic:
1) play in the main gear attached to the motor shaft
2) gap between the motor shaft gear and the worm shaft gear (the “normal” place to adjust the axis)
3) improper centering of the worm gear housing (more on this in a bit)
4) play in the RA housing somewhere (not something I’d originally considered)

There was a tiny amount of play in the main gear. A little less than the play in Dec. Again, I decided there was nothing I could really do about it.

The gap between the gears seemed about right; I later ran a paper between them, and it seemed fine. There was still a little play present, but with the gears gapped properly, This didn’t really seem like something that I was going to mess with.

The worm gear housing is definitely not centered in the main gear housing gap. I don’t know what that’s about, but I decided not to change it; I was looking for something “simple”.

And I found it.

The large bolt that holds that RA shaft lock housing to the Dec axis was very loose. The two housings could shift past each other by several millimeters, probably as much as 1/2″. This is just shoddy work putting the mount back together. Nothing else could really explain it. I used my finely-calibrated index finger to confirm that I had the two housings lined up properly, popped a big Allen wrench into the bolt, and was able to tighten it about half a turn before things felt secure. I re-checked to make sure I was all lined up, gave the bolt another tweak, and started reconstructing the axis to see if I’d made a difference.

Putting the axis back together was pretty straightforward; get all the junk on in the right order, make sure everything is lubed properly or it won’t fit (more on tolerances in a bit), put all the parts together, and once I screwed the mottled black RA main axis cover, everything tightened up and into place. I did a quick test to see how the axis felt, and was impressed. I re-tightened the set screws, reinstalled the rear cover, and spent a little time getting all the greasy fingerprints (some mine, some left by the professional “repairs” I’d just had done) off of the mount. With everything tight and back in place, I checked the backlash in RA. It was down to well under 1mm. There is still the tiniest amount of play in the axis, but it’s certainly down in the “acceptable” range. I might try to tighten it up a little more, if I find it to be problematic, but I was ready to declare the mount repair a success!

While I had everything off of the mount, I thought that I should rotate the pier top a little; the way it’s set up, I have to run the mount’s Azimuth all the way to one stop. I pulled the mount, figured out which direction to rotate the top, rotated it, then re-leveled the pier. I actually kind of like leveling, when I have a good tool and a good surface to work with. I had both, and the leveling only took a couple of minutes. I put the mount back up, and moved on to the next task.

I thought about whether to remount Veronica, and I decided that I should polar align with the mount unloaded first. So I left Veronica sitting there, waiting for the sun to set.

I had one last thing to do before sunset, though; the polar scope illuminator has always been a little flaky, and I decided that I needed to fix it. I messed around with it quite a bit, and finally found an LED that would work as a replacement. I de-soldered the original LED, put in a new one, and was pleased to find that the new LED looks very bright. I’m glad that I didn’t ask for this to be “repaired” professionally. For the price of an LED and a couple minutes with a soldering iron, I probably saved myself a few hundred bucks. The new illuminator is visible even during the day, so I was able to get it reinstalled on the mount and all ready to go for sundown.

After sunset, the sky was clear enough to polar align, so I did; using the new reticle with the fixed illuminator made the process simple. I reset the mount’s bubble level to be properly arranged for the new reticle; it’s nearly in CW-down position, but not quite.

I added some Velcro to the saddle and to the USB hub and DSUSB, so now I can just put the cable harness right where it needs to be.

I played around a little with saddle placement. I think that eventually I will want to figure out how to offset the saddle, but for now it’s centered.

I got Veronica back up and cable harness attached. Spent some time getting the balance right. Once Veronica was all mounted up, I pointed over to the moon and focused the camera.

Suddenly, I was ready to start imaging!

Well, ready to test out the imaging rig with the newly-repaired mount, anyway.

I pointed over at Alkaid in The Big Dipper, sync’d the mount, and then did a GOTO to M51 nearby. 3min unguided left some star trails, so I fired up the guider and re-shot a 3min exposure. The guiding accuracy was down in the 0.07pixels (about 0.4″) range in both axes. The stars came out very round, as you’d expect.

I realized that I didn’t have the LPS filter in place, so I reinstalled it, pointed back at the moon, and refocused. Turns out the LPS filter gives a little extra backfocus. Nice! Back to Alkaid (you gotta love the simple pointing model), over to M63 this time, and another 3min shot. Very nice looking galaxy. Exactly what you expect from a 3min shot.

I need to tweak the polar alignment with WCS. I need to get autofocus and plate solving working again. I need to clean the camera optics. There is always stuff to do.

But I hereby declare Ad Astra Observatory at Newark, CA up and running once more.