Ad Astra Observatory

Now that the moon’s big again, it’s time for a little maintenance.

I had two main things on the to-do list. First, the new drive system has been making a constant high-pitched whine at idle, ever since I got it. I suspected the gear mesh, and besides, who doesn’t love a little white lithium grease? Second, I wrote some tested Python to drive the dome around, so I was hoping to try it out, and see if I can tweak the parameters a bit, to make Lesvedome more accurate. It’s pretty good now, but there’s a small delta between where the dome should point, and where it ends up, especially over a long imaging run.

First, I tackled the RA motor on the mount. I pulled off the housing, greased things up, and immediately noticed that the frequency of the whine is way higher than the speed at which the two exposed gears are running. So, I opened up the motor housing a bit, to see what’s what, and there is a 12:1 reduction gearbox in there; that seemed a likely culprit. Good thing I didn’t tear into it, though, because it turns out that a careful read of the motor software parameters in the user manual led me right to a setting that made the whine go away, instantly.

OK, so first I tried switching the motor from “Microstep” to “Full” and “Half” step.

…

that was a mistake. Man, does the motor get loud and chunky when it’s in full step mode! 🙂

However, this gave me confidence that changing the settings can definitely create new and interesting tones out of the stepper motor, so I pored through really carefully, and found that the Freq2 parameter said something about changing it if the motor gets louder… so I gave it a go. Manual says typical settings are 30Hz. Mine was set to 0Hz. I tried 10Hz, just to see if changing it would help… and the motor went dead silent immediately. Nice!

With a smoother-running RA motor, I decided to tweak the speed settings, to see if I could goose a little more slew rate out. I don’t think it made a huge difference, but I did change the max rate from 200x to 260x, because it sounded fine. My understanding is that max speed is closely tied to input voltage. I am running a 15v/25A power supply, I will invest in a larger one (it is good from 9-30v) as I start moving to a “permanent installation”.

In short, the mount maintenance went quite well.

While I was scratching my head about the mount, I hacked my way to getting the dome moving under Python control. I figured out immediately after trying my first dome slew, that I needed a “please stop rotating now” script, as it was very easy for the dome to get to a “I’m just going to keep turning” place. I added in a bunch of timeout code, too.

I also had to write my own “counter” (based on polling) driver, as the one from Velleman just didn’t do what I wanted.

But, by the end, I was able to do slew-to-home both CW and CCW, and slew-by-amount (in something reasonably like degrees, kind of). I was also able to take a few full-rotation readings. I don’t quite know how to translate all the numbers, but I got a similar number of encoder “ticks” on 5 full rotations, so I guess I’m counting something.

The dome still needs some attention, before I can fully grok what I’m seeing. But I have a (very) basic control system now, and that’s good stuff.

Spring rains, bright moon… maintenance went well. 🙂