Celestron has tried to fix my CGX twice

[Nick Gatel]

Mount is about 18 months old. Used to guide at 0.7 to 1.3 as/px RMS, depending upon seeing. A couple months ago it started to get worse. Could never guide below 1.3" RMS. I live in a desert Bortle 5 location.

Send several files to Celestron. They had me send the mount in for warranty repair: they adjusted Ra and Dec axis.

Got it back with no improvement. Sent more files. They had me send it back for more repairs. This time they replaced the RA motor.

I am using a WO 50mm guide scope with ASI120MM Mini camera mounted on top of an Esprit 100ED. The guide scope is mounted on a Vixen bar bolted to the top of the Esprit clamps. Very solid. The guide scope is fitted inside two clamshell rings - very secure - I cannot turn the guide scope by hand.

I need help before I contact Celestron again. Attached are two logs. Both of them I did the following after I got the mount back:

• Calibrate PHD2 near the celestial equator and south meridian
• Let PHD2 guide for 10 minutes
• Using the Celestron PECTool, ran 10 worm cycles, then uploaded the PEC file to the mount
• Ran Guide Assistant for 10 minutes and accepted all suggestions
• Guided at 1s and 2s exposures 

https://openphdguiding.org/logs/dl/PHD2_logs_f4Rt.zip

[Bruce Waddington]

Sorry you're having so much trouble.  I think you may be conflating results because of your immediate use of the PEC tool.  If we look at the longest guiding session, we can see the raw RA tracking is fairly poor:

The frequency analysis shows a very large tracking error at a period of about 480 seconds and a lower component at about 240 seconds.  These are probably harmonics of the worm period.

I think it's likely the PEC you did was bad, perhaps making things worse.  I think you should forget about PEC for now and carefully measure the native tracking error of the mount without any software corrections.  The other thing I would do is create a new PHD2 profile for testing purposes, one that guides through the main scope.  This might be a nuisance to put together but if you can connect the guide camera on the main scope, it completely eliminates the guiding assembly as a source of error.  To create the testing profile, run the new-profile-wizard and be sure to specify the correct focal length, the focal length of the main scope.

Good luck,

Bruce

[Nick Gatel]

Bruce,

Thank you for taking the time to look at this. So you are saying that the PEC Training was bad? Celestron said to do PEC Training with their tool right away. Because guiding was so poor the first night (3/24), I did a new PEC Training session on 4/1.

I'll try your recommendation of guiding through the main scope, Won't have good skies until next week. 

[bw_msgboard]

Hi Nick.  I just looked at your earlier log from 3/24 and it looks pretty much the same - sorry I didn't do this earlier.  So the PEC training didn't accomplish anything but I can't say it made things worse - that assumes there was no PEC active in the 3/24 log.  I really don't understand why Celestron would consider this level of tracking error to be acceptable, it really looks like a fundamental mechanical problem with the RA drive system.  But I'm not an expert on these mounts, that's just what it looks like to me.  Personally, I think they should be looking at the unguided, native RA tracking performance without trying to band-aid things with PEC curves. 

 

Good luck,

Bruce

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Nick Gatel
Sent: Saturday, April 02, 2022 4:18 PM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re: Celestron has tried to fix my CGX twice

--
You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/0qQigHdDiXY/unsubscribe.
To unsubscribe from this group and all its topics, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/8a3fe0e8-7514-456f-a25a-bf9e0602390an%40googlegroups.com.

[Nick Gatel]

Bruce,

I think I figured it out. You said, " I think you may be conflating results because of your immediate use of the PEC tool." 

I had to look up the definition of conflating: "combine (two or more texts, ideas, etc.) into one." That was an Aha moment.

I looked through Celestron's documentation for PEC training with CPWI or the hand controller, and the PECTool manual and I don't see any mention anywhere (maybe I missed it), that the mount's PEC should be disabled when training. Plus Celestron didn't specify I shoaled do this in our email exchanges. 

So what I must have been doing is PEC training the mount while it was already playing back the PEC, thus conflating the data. Is this correct?. 

Tonight I had about one hour of below average-to-average seeing before the clouds rolled in. So this is what I did/found:

• Disabled PEC playback in the mount and guided for 10 minutes. RMS was 1.18" RMS. PHD2 was showing SNR in the 12-14 range, which is pretty low compared to what I usually get.
• Then, using CPWI, I did PEC training and uploaded it. CPWI runs PEC training for about 1 worm cycle and uploads the file. The SNR had dropped down to 9-11 or sp. I was worried that PHD2 would lose the star mass, but it didn't.
• Then I ran the mount guided with PEC active for about 10 minutes. RMS was around 1.20" RMS. RA changes were kinda gradual compared to what I had been getting previously.
• Lastly I ran the Guiding Assistant for 10 minutes or so. The high frequency Star RMS was 0.80" — so seeing was poor. But my RA peak to peak was significantly improved compared to what I had been seeing. I accepted the suggestions.
• Then I started guiding, but clouds started rolling in and PHD2 started to lose the stars, so I ended the session.

So I think I solved the issue. Should have good skies in a few days and will start over.

THANK YOU SO MUCH!

- Nick

[Bruce Waddington]

Hi Nick.  Whatever PEC curve is computed and uploaded to the mount must represent the totality of the desired corrections.  So you can't "layer on" iterative corrections.  More sophisticated PEC training programs can iteratively compute the best correction curve, a process called "refinement", but what is eventually uploaded to the mount still represents the total picture.  Unless you're using that kind of program and are doing refinement, the PEC in the mount must be disabled.  I recommend that you take things a step at a time so you can see what's happening. 

1. Disable PEC in the mount and do a GA run for at least one full worm period. 

2. Try programming PEC in the mount and enable PE corrections. 

3. Immediately do another GA run for at least 1 worm period.  Compare the results - did the PE training help or not? 

You should do all of this in the same region of the sky.  It's probably best to start on the east side of the pier pointing near Dec=0 and close to the celestial meridian.  This will give you time for these iterative runs and will also give you the best view of the tracking errors in the mount without introducing complications of meridian flips or other large movements of the mount.

Good luck,

Bruce

[Nick Gatel]

Bruce,

You have been so helpful. I did upload a 10 worm cycle PEC session to my mount using the instructions you provided. Doesn't seems to be a whole lot of difference. But things are somewhat reasonable now.

The night after the PEC Training, I ran the mount for about 6 hours using 2 minute exposures (Apr 6 2022). Results were poor with RA RMS more than 2 times that of Dec. I sent the file to Celestron and was told with my mount I should use 0.5 or 1.0 second exposures and also guide east heavy. I challenged this base on all I have read. They came back and said some mounts work better with such short exposures. To me, and I may be wrong, such short exposures might be a bandaid for a mount that has excessive mechanical problems.

So anyway, the next night (Apr 7) I followed their instructions:

• Ran guided for about t0 minutes (my worm cycle is around 8 min) at 0.5 seconds
• Ran the GA for over 10 minutes and PHD2 recommenced RA min move of .20 and Dec min move of .30 (I have never gotten a suggestion of over ,20 for any min move previously)
• I accepted the suggestions
• Ran PHD2 guiding for about 2 hours with 500ms exposures
• Ran PHD2 guiding for the rest of the nigh — I should have done another GA session at 1000ms, and will do that tonight
• At around 12:30AM a meridian flip occured and I forgot about east heavy, so I moved a counter weight during the focusing run, which created a big blip in guiding at that point 
• I then let it run until around 4AM

My concern is possible mechanical issues, especially since I have used the mount for at least 400  nights over the past 18 months or so. I used to be able to guide at 0.7 - 0.8 on good nights. I can not longer do that, which means I can't use my C8 with reducer if the mount never goes below 1"/pixel.

I uploaded the two logs and would greatly appreciate your feedback. Thanks in advance.

- Nick

https://openphdguiding.org/logs/dl/PHD2_logs_zatU.zip

[Bruce Waddington]

Hi Nick.  I really can't see that you're making a lot of progress here.  This is the latest FFT analysis of your RA tracking:

I guess this poses the question of whether you're still having procedural problems with the PEC software. I really don't know.  But in terms of your mainstream guiding performance, I think the biggest problem is the 32-34 second oscillation of over 1 arc-sec.  This isn't something that will be handled well by PHD2 or in all likelihood by any sort of PEC software.  With this fairly abrupt error every 34 seconds, it's unlikely you'll drive the RA guiding RMS below 1 arc-sec.  I suggest you take this up with some of the people on the Mounts forum of Cloudy Nights.  Someone will probably know what component of the RA drive system is causing this and perhaps what you can do to mitigate it.

Good luck,

Bruce

[Nick Gatel]

Again, thanks.

I went back through my Guide Logs for the past couple of months. Every session shows that same > 1.5" amplitude spike at 34.2 seconds, with a couple happening at 34.3 or 34.4 seconds. Since the mount is still under warranty, I'll build a table of these peaks in a spreadsheet and send it to Celestron. If they disagree it is a problem, then I'll go ahead and post my difficulties on CN.

[Nick Gatel]

Bruce, I want to thank you for your help. Often these threads just die and no one knows if the problem was solved or not. I solved mine.

If you remember, I had sent the mount to Celestron twice under warranty. The first time they did adjustments. No improvement at all. After going through several procedures at their direction and sending them files, they had me return it again. The RA motor was replaced. Again, no improvement. This was the point I posted here.

With what you taught me about the high frequency amplitude at around 32sec, I put together several charts and sent those and the logs to Celestron. Here is their response:

- - - - - - - - - - - - - - 

"Hello Nick,

 Thank you for the logs.

 We have reviewed it but there is nothing in there showing a defective mount. The 34.2 second spike you are seeing is due to the design of the pulley system(14 teeth per ~480 second worm cycle). PEC should take care of most of the spike; however we do not guarantee < 1 arc-second performance.

 It does looks like tweaking some setting would improve the performance a bit. I would suggest increasing the aggressiveness and trying .5sec exposures. It also looks like you can set the RA axis slightly more east heavy. Doing a new PEC training with more cycles averaged may should also help (you mentioned you did 10 cycles the first time before the repair but not sure on the cycles for the recent one)"

- - - - - - - - - - - - - - 

Reviewing my options, I didn't feel comfortable selling the mount to someone, given the problems I was having. Not knowing how long it would take, or even if I could fix it, I ordered a couple polyurethane drive belts and would install them and start working on the mount when they arrived (took a couple months).

In the meantime, I bought a new EQ6-R Pro mount.

Once I got the belts, I opened up the mount to inspect the RA and Dec motors. I found a great variance between each in the adjustment of the backlash, worm-blocks and belt tension. I replaced the belts, adjusted the backlash on both axes, and using a dial indicator adjusted the worm blocks. I also used a digital caliper to adjust the belts so there was equal deflection on both sides of each belt. Given how it had been adjusted by Celestron, I decided not to go any further and test the mount to see if this improved things.

I set up the tripod using three pavers with a dimple drilled in the center of each for each leg of the tripod. These were set in the gravel of my backyard. A gravel base in the heat of one of hottest deserts in the US isn't the best idea, of course, especially if one is imaging in the summer. Because of the location of the mount in close proximity to a fence on the north side of my backyard, I had to extend the legs of the tripod all the way to gain clearance over the fence to Polar Align. Success! Guiding is now almost always under 1.0" RMS and on really good nights can get down to the 0.6" RMS range.  I have been imaging with the mount on the tripod almost every night, all night, for the past 4+ months. I haven't re-checked the Polar Alignment either. I have found most nights the EQ6-R Pro, which is mounted on a pier is usually guiding 0.2" to 0.3" better than the CGX. I would expect better guiding with the CGX if I moved it to the pier.

I don't think replacing the belts solved the problem. It was proper adjustment of backlash, worm blocks and belt tension that solved the problem. All fairly simple.

With the CGX I must use the PHD2 PPEC algorithm and my best guiding is achieved with 1.0 second exposures. My EQ6-R Pro has not been PEC Trained and I guide that mount at 2.5 seconds using the Hysteresis algorithm. For now I am going to keep the CGX. I also DID NOT set up the mount to be east heavy and I get similar guiding on each side of the mount. Note that Celestron recommended 0.5 second exposures, east heavy balance and more aggressive guiding. 

Here is a typical FFT RA high frequency graph nowadays. At 34.7sec the amplitude is 0.4" and is marked by the white dot. Without your input I could not have resolved the problem. Thanks, again.

[Bruce Waddington]

Hi Nick.  Thanks for reporting back on this.  You’re right, we don’t often hear the outcome for these more complicated problems.  It seems to me you’re the one who did all the hard work, that sounds like some significant adjustments you had to do on that drive system.  Congrats on getting it done and I hope you continue to enjoy good results on both mounts.

 

Regards,

Bruce

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/4a91dde1-e61d-4076-8ef0-b3403f00a1ffn%40googlegroups.com.