Re-Greasing my CGX Worm Drives

[Graeme1858]

My CGX sat in the garage for nine months whilst I was otherwise engaged and now it's sitting on top of my new pier. Trouble is, it's got a problem. It slews ok on the hand controller and from NINA but it is sadly unable to guide. The PHD2 Star Cross test produces an image of star Ts! Whilst calibrating it is unable to move adequately North and fails. This has been a most frustrating time! I posted the log on the PHD Open PHD Guiding Google Group and Bruce came to the conclusion that "There's something significantly wrong with the behavior of the Dec axis on your mount" and produced this image from the log:

So I sent copies of the images and a description of the problem to the original supplier to ask if they could advise a way forward or a telescope mount maintenance provider in the south of England. They forwarded my email to Celestron UK. They forwarded my email to Celestron US. I'm still waiting for a response! After much searching and watching videos, I decided to clean and regrease the worm drives.

On the advice of the supplier I ordered some Geoptik Equatorial Mount Grease, it comes with a pot for the worm gears and a less viscous pot for the bearings and shafts. I thought I would sort out the worm gears first and see how that goes and leave the bearings for now since they're less susceptible to dust ingress and it's a much more invasive operation.

This is the Declination worm drive with the cover off: (click the image for a close up)

The first problem was the Declination belt was a lot looser than the Right Ascension belt, probably caused by the grub screws to the right of the belt, there's two of them, between the motor block and the worm drive block, they were half way out instead of tightened against the worm drive block. There's three Allen bolts that hold the motor block in place but they need the grub screws to be tight to maintain the correct adjustment.

Then the cog on the motor drive shaft was too close to the motor. As a result the belt was only half way on the cog, only held in place by the flange on the worm drive shaft cog. I took the belt off, moved the drive shaft cog away from the motor a couple of mm and put the belt back on. Then I checked the belt tensions compared to each other, in both directions, with my eyes closed, using the force!

Once I was happy that everything was as it should be, I loosened the clutches and set the mount motors turning towards the home position. whilst they were spinning I used a tooth brush and a cloth to remove as much of the old grease as I could and reapplied the new grease. Again, I could have dismantled more and been more through but I wanted to keep things as least intrusive as possible at this stage.

So, I waited for darkness to fall then fired up NINA. I aligned the mount, slewed to Deneb, ran an auto focus and plate solved. Then I fired up PHD2 and was pleased to see stars in focus, captured by my new ASI220MM mini guide camera. It was good to see that my OAG fiddling appears to have been about right! I clicked Auto Select Star and Begin Guiding thinking it would start a new calibration but it went straight into guiding. After a short period of dodgy guiding it settled down to fairly straight lines both within 2" peak to peak. But the back lash was bad and the
Target plotted dots in a straight line slope. After watching for a while I stopped guiding and started a forced calibration. Then the clouds came over and all the guide stars disappeared!

The forecast is partially cloudy for tonight, so hopefully I can try a calibration and see if my fettling has been successful!

It was the most fun I've had in ages!

Graeme

[Lady Fraktor]

Glad you could get a start on fixing things. Hopefully you will have it running well again soon.
Reading that I am understanding how you AP guys are a bit off
There is something calming about just putting an eyepiece into the telescope and viewing

[Gordon]

Hopefully everything will be back in working order. I need to do some work on mine also!

[Graeme1858]

Glad you could get a start on fixing things. Hopefully you will have it running well again soon.
Reading that I am understanding how you AP guys are a bit off
There is something calming about just putting an eyepiece into the telescope and viewing

Yes, AP can be the most frustrating hobby invented. But then when it all comes together, then it's such a buzz!

Hopefully everything will be back in working order. I need to do some work on mine also!

Thanks Gordon

Fingers crossed for a clear night!

What's the problem with your mount?

[SkyHiker]

The calibration curve gives you a solid handle on the mechanics. Once you fixed the belt tension, worm engagement etcetera, this problem should be solved. Too bad the fog set in just as you were going to recalibrate!

The RA curve doesn't look stellar either. I wonder if the mechanics have problems, if the step size is too small, or if the guide scope is part of the problem (noise, sampling properties). Something to keep in mind.

[StarBru]

Graeme,

I'm glad you could have some fun with all that and hopefully fixed the problem! Anxious to know how it all works out!

[Graeme1858]

The calibration curve gives you a solid handle on the mechanics. Once you fixed the belt tension, worm engagement etcetera, this problem should be solved. Too bad the fog set in just as you were going to recalibrate!

The RA curve doesn't look stellar either. I wonder if the mechanics have problems, if the step size is too small, or if the guide scope is part of the problem (noise, sampling properties). Something to keep in mind.

Yes, the RA curve could be better, I'm hoping the re-greasing of the worm drive will help.

It slipped my mind, Celestron have offered an initial response to my dilemma and suggested backlash was the problem. They advised "Set the calibration steps field to something like 10, and the “Calibration Step [ms]” to something like 2000." I did that but 2000ms was far to long to not loose the guide star, I tried 1000ms with the same result. Then I tried 500ms and the calibration completed with warnings. The resulting guide RMS error was >2 and the captured image data was unusable. I sent them the results and that's the response I'm waiting on now.

The guide scope was part of the problem I was using an old ALTAIR GPCAM2 which Bruce suggested I should bin at 2x2 but this camera doesn't bin. But the new ASI220MM mini is doing a great job.

[Gordon]

What's the problem with your mount?

I have 2, my EQ5pro just doesn't want to guide in dec very well, my guess is I just need to fiddle with the dec backlash.

My EQ6pro was working fine last year but it's time to do some maintenance since it's just been setting.

[JayTee]

Graeme,

I think that you're minimally invasive approach to start the fix of the problem was absolutely correct. I'm very interested and how this maintenance turns out.

I'm now afraid to take something so far apart that I can't with confidence put it back together. I've only done that around a hundred times in my life. It's a hard lesson to learn apparently.

[Graeme1858]

What's the problem with your mount?

I have 2, my EQ5pro just doesn't want to guide in dec very well, my guess is I just need to fiddle with the dec backlash.

My EQ6pro was working fine last year but it's time to do some maintenance since it's just been setting.

Good luck!

Graeme,

I think that you're minimally invasive approach to start the fix of the problem was absolutely correct. I'm very interested and how this maintenance turns out.

I'm now afraid to take something so far apart that I can't with confidence put it back together. I've only done that around a hundred times in my life. It's a hard lesson to learn apparently.

Funny how it's so easy to take something apart but so hard to put it together again!

[Juno16]

Hi Greame,

Very nicely documented work in your thread.

I surely hope that your work has been successful!
I know that adjusting backlash at the worm can be tedious, but once you hit the sweet spot, you are good to go!

Looking forward to hearing some good news soon.

[Graeme1858]

Looking forward to hearing some good news soon.

Thanks Jim, me too!

We have rain forecast for tonight (my marrows and beans will be pleased!) Then it's cloud free for a week!!!

Graeme

[Graeme1858]

Graeme,

I'm glad you could have some fun with all that and hopefully fixed the problem! Anxious to know how it all works out!

Thanks Bruce

It was fun but it will be an even greater experience if my guiding is fixed! Hopefully I will find out this weekend.

Graeme

[Juno16]

Looking forward to hearing some good news soon.

Thanks Jim, me too!

We have rain forecast for tonight (my marrows and beans will be pleased!) Then it's cloud free for a week!!!

Graeme

I went back and re-read your original post.
Sounds like you had some significant issues with the dec deive belt assembly! Surely sounds like that was a problem or was getting ready to be!
I read that you forced a calibration. What did the graph look like?
It will be interesting to see what the PHD2 Guiding Assistant reports.
Fingers crossed Graeme!

[Graeme1858]

I read that you forced a calibration. What did the graph look like?
It will be interesting to see what the PHD2 Guiding Assistant reports.
Fingers crossed Graeme!

I didn't get a graph because I stopped it as soon as it started.

But with Henk's help I now have the PHD Log Viewer installed so I can look up previous logs when I'm indoors.

I'll post up some graphs when I get calibrated.

What sort of guiding figures are you getting Jim?

[Juno16]

I read that you forced a calibration. What did the graph look like?
It will be interesting to see what the PHD2 Guiding Assistant reports.
Fingers crossed Graeme!

I didn't get a graph because I stopped it as soon as it started.

But with Henk's help I now have the PHD Log Viewer installed so I can look up previous logs when I'm indoors.

I'll post up some graphs when I get calibrated.

What sort of guiding figures are you getting Jim?

Hi Graeme,

The PHD2 log viewer is an excellent tool. You can do all sorts of troubleshooting with it. Including when the guiding was lost due to clouds!

I use it during a session to check my guiding error average for a run and also to check PA.

My guiding is as good as it has ever been. My last run, the average total rms error was 0.6 on both sides of the meridian. Previously, I had my best ever at 0.5 on both sides of the meridian. Thant's with terrible seeing. The mount guides equally well with either the 102mm or 61mm as long as they are well balanced.

I have an HEQ5 with the Rowan belt kit installed. I had to make three shots last year at getting the backlash out of the DEC worm gear, but now I consistently get <150ms DEC backlash when I run the PHD2 Guiding Assistant. PHD2 backlash compensation easily handles that!

Looking forward to some good news soon from you!

[SkyHiker]

Huh, that ASI220MM has an SC2210_BW sensor, my QHY5III200M has an SC2210 but it is BW so I presume it is the same. You are right, it is a fine autoguider that produces lots of stars. Yours is USB2.0 type C with no memory listed, mine is USB3.2 type C with 512 MB DDR for caching and it cost 10 bucks more. It looks like they are pretty similar.

I think you may be able to test it in the daytime, using the manual steps in PHD2 if you can get them of the same size as your calibration step.

[Graeme1858]

Well we had a clear night last night and I set up to calibrate my guiding. I must say, the new ASI220MM mini is really good. So I set the guiding calibration going looking 0° Dec just West of the Meridian. It was disappointing to see that guiding failed due to not enough movement North. The exact same problem was still there!!!

So I upped the Calibration Step size to 1000ms and tried again. Nearly made it. Upped to 1500ms and tried again, star lost due to neighbour's smoke bush! I slewed back to nearer the Meridian and tried for a forth time. This time the guiding calibration completed! Albeit with a warning that the calibration wasn't very good.

I ran the Guiding Assistant a couple of times, applying the suggestions. Then recalibrating and Guiding Assistant again. On the third attempt one of the suggestions was to try a Dec Backlash Compensation of 1720ms. I clicked to apply and watched the guiding graph. After two and a half hours of messing about it finally looked like I was guiding. The Total Error dropped to 1.35 (0.78").

So, finally, at about midnight I slewed to M31, auto focussed, plate solved and set up a NINA Sequence to try a first light on my new ASI1600MM and LRGB filters. 180 seconds was ok, 240 seconds blew the core of the galaxy out. So I put 4 x 180 second subs on each filter in a loop till astronomical dawn. I chose M31 to keep things simple because at this time of year it is in the sky all night without crossing the Meridian. It was that or M76 for the same reason. And away it went!

At half 12ish I was so excited I set up a Google Meet and sent an invite to a few people.

Just after 0100 the guiding jumped to 2.3! Then after a few more minutes up to 3.6 and the guide star was lost! I did an auto select, found a guide star and away it went again. Greg @AstroBee was in the Google Meet and suggested the poor Total Error was probably due to bad seeing. The jet stream was overhead so it's a fair bet! On Greg's advice I tweaked the Dec and RA aggressiveness settings to attempt to compensate. We tried 4 second exposures but nothing changed much, I didn't get sub 1.0 guiding all night! But it did drop to 1.1 by 0300.

Just before astronomical dawn I stopped the sequencer and I was going to capture some flats but my Pegasus Astro software had lost connection with the power box. I'm glad it was just the front end dashboard and not all the supplies! The plan was to do the flats then set up a darks sequence, leave it running, lock up and go to bed. But without the PPB I couldn't switch on the light box, so I left the telescope parked and went to bed anyway.

Not a perfect night of astrophotography but the showstopping problem appears to have been overcome and I'm able to guide without replacing my CGX! A result then!

Despite the poor guiding, the data looks ok, the stars are round and there's no sign of noise. I need to watch some LRGB PI processing videos now and see what we got!

Thanks for reading.

Regards

Graeme

[AstroBee]

Glad the night was mostly successful.

[SkyHiker]

Congratulations! I would expect better guiding from a CGX than 0.78 RMS, once you get it nailed. You need better performance for an SCT. One suggestion, while PHD2 can compensate for DEC backlash to some extent, it cannot achieve the rapid motor speedup provided by the Nexstar backlash compensation. You may want to give it a try for DEC, though it can backfire when not set accurately. I did this with my AVX, and it seemed to work OK. First of course I would check if the DEC worm can be tighter without binding up.