What's Wrong with the Celestron CGEM Mount?
Posted 04 December 2011 - 03:03 PM
By Greg Marshall.
Posted 04 December 2011 - 11:13 PM
Great article. My mount does it too. I wish there was a good solution to it.
Posted 05 December 2011 - 06:11 AM
My CGEM worked just fine, but I a fortunate enough to live in Arizona so I have Starizona near by. When I bought it from Starizona, Dean Konig looked it all over then lubed and set the gear adjustment very carefully. I never had a cogging problem and once I got good with the polar alignment routine I got lots of great images and enjoyed my CGEM.
After taking all the images I wanted to take, I was done with astrophotography and returned to just viewing. I have a simple tracking mount and I enjoy some wide field imaging, but I am happy that I returned to just looking at the sky.
It is unfortunate that there is not one simple "fix it" for this problem.
Clear skies to us all;
Posted 05 December 2011 - 09:29 AM
Tnakfully my CGEM does not exhibit this behavior when guiding - works great!
Posted 06 December 2011 - 02:16 AM
Posted 13 December 2011 - 03:20 AM
I was asked whether a mount that does not exhibit the problem should be upgraded and I think the answer is "no" - at least until it has been proven to work reliably. I say that because I believe that the changes needed were extensive.
Posted 16 December 2011 - 07:11 AM
did u get a new motor from celestron or just
swap RA motor with dec motor?
Posted 23 December 2011 - 07:19 PM
Posted 23 December 2011 - 07:24 PM
Posted 01 January 2012 - 09:41 PM
I'm no astrophotographer but still enjoyed reading your clear exploration of such a subtle issue.
I wonder if one could source high-end servos satisfying the specs of the CGEM servos. Seems like upgrading to motors that consistently avoid cogging would save a lot of APers a lot of hassle.
Posted 11 January 2012 - 04:22 AM
I enjoyed your fine review of the cogging problem. I have a "mild" case of this - in that my jumps are normally within +/- 1 pixel, but sometimes dec will wonder off 2 or 3 pixels. I upped the Guide rate on dec in the mount to 90% and upped the PHD's Min Motion to .40 - and both of these measured seemed to help - making the jumps less severe. And as you noted, upping PHD's Max Dec to 1800 or 2000.
I believe it was Ed who has communicated with a firmware engineer at Celestron who acknowledged the problem and is supposed to be working on a fix.
Would be nice - then it would be reasonably good mount.
Posted 17 January 2012 - 12:19 PM
There have been serious efforts to find an alternative motor for the CGEM, but AFAIK, none has been found.
Doug, yes, the degree of cogging will vary from one motor to another. I'm skeptical about improving the situation with adjusted guide parameters. Have you measured the long-term performance with and without the changes?
Agreed, the CGEM would be a fine mount if they could solve this problem - or if you are lucky enough to get one with good motors. I've seen and tested some with good motors, but both of the units I owned had bad motors on the DEC axis.
Posted 21 January 2012 - 01:18 AM
Have not used the mount since - we've had lousy weather, so can't comment on long term performance. But it clearly dampened the dec excursions during the session when I made the adjustments.
Several days ago, I did receive an email from Derik of Team Celestron regarding the cogging issue. He responded - "As for the DEC guiding problem, I am working on it as fast as I can. I hope to have some alpha code out next month."
I've signed on for Beta testing, so hopefully we will see a fix in a month or two.
Posted 29 January 2012 - 07:20 PM
Posted 14 February 2012 - 07:34 AM
My CGEM has been flawless since day one. The only time I have had any issues with guiding, it's been a balance issue on my part.
Posted 09 April 2012 - 12:56 AM
Posted 01 August 2012 - 08:36 PM
Greg or anyone,
Any good alternatives?
Also, short of buying one, how would we have any assurance that Celestron has fixed the problem or minimized it?
Posted 09 September 2012 - 11:15 PM
Posted 10 September 2012 - 06:47 PM
Posted 11 September 2012 - 08:52 AM
Was wondering, is this just a problem with the CGEM or is it indicitive of the CG-5 ASGT mounts also. I get something similar with my CG-5, but always thought it was a sticky gear problem (even after cleaning and buffing the gears etc.
I've been wondering too. I've been lurking here for a while, and while I don't own a CGEM I've seen this problem with my CG5. I find that I can usually resolve it by either "perfect" polar alignment (drift alignment refining the All-Star routine, which theoretically eliminates the need for DEC corrections entirely) or by deliberately inducing DEC drift (by mis-alignment) while unbalancing with force in the direction of the drift.
I believe that this issue is caused by the mount "springing back" after a correction, when the DEC is imperfectly balanced in the direction opposite the correction. By unbalancing in favor of the correction, the problem goes away.
I don't think that "motor cogging" has anything to do with it. Maybe the geartrain is "cogging" due to binding or other interaction in the spur gears, but motor cogging only relates to a periodic error that is related to the number of poles in the motor's rotor. Since the motor's instantaneous rotation rate exceeds 30 RPM even at the slowest correction rate and its cogging rate would therefore be at least 1 Hz (since there are at least two poles), active guiding cannot possibly act fast enough or precise enough. It's usually not a problem, except possibly in the most precise drives that strive to obtain accuracy to within an arcsecond. As an aside, some audiophiles can hear the effects of motor cogging even in the best direct drive turntables (which perform a function similar to that of a mount's RA drive), but only because the human brain can be quite sensitive to audio clarity (flutter/jitter).
A well-engineered analog DC servo should be capable of exceeding the performance of any equivalent-sized stepper motor in terms of both torque and flutter, but I don't think that the Celestron servos are optimized. Their control loop is digital rather than analog; that introduces digital jitter even if the software CONSTANTLY monitors the encoder and controls the shaft position: the software might not be capable of that in this case because it's doing many things at once. But the software could certainly stand further development; it should at least be able to hold the servo output stationary after a correction.
Posted 23 September 2012 - 03:02 AM
Posted 08 March 2013 - 05:27 PM