CGE POOR

377 views
Skip to first unread message

Kenneth Crichton

unread,
Apr 30, 2016, 9:02:28 PM4/30/16
to Open PHD Guiding
 Hi all!

Im beginning to wonder where my year of astronomy has gone, three cge pro's since December and im still getting the same issues. The Dec and Ra rates are both working at different speeds which for me indicates backlash but on three separate mounts?? Could i have a faulty cable etc etc? Phd recommends i guide in only one direction but ive noticed the backlash setting after running the assistant always varies, sometimes considerably! Im also going to try and plot my periodic error tomorrow as that could be another issue :( i have the software to resolve that though !

Thanks ken
 
PHD2_DebugLog_2016-05-01_010145.txt
PHD2_GuideLog_2016-05-01_010145.txt
PHD2_GuideLog_2016-05-01_001252.txt

Andy Galasso

unread,
May 1, 2016, 1:22:22 PM5/1/16
to Open PHD Guiding
Hi Ken,

Your RA calibrations are showing some non-uniform motion in RA. The mount seems to move forward a few increments then jump forward:

Inline image 1

One possible cause for this could be poor balance in RA.  Are you perhaps near the weight limit for the mount?  I'm not sure if it would help but you could try biasing the weight a bit to the east (very slightly east-heavy) to see if that helps.   You should probably also check in the CGE forum and see if any other CGE Pro users have some more specific tips.

I believe the non-uniform RA response is causing RA oscillation during guiding:

Inline image 2

The Guiding Assistant run is a good illustration of the RA behavior:

Inline image 3
The RA curve (blue) shows some regular spikes, much larger than seeing (Dec curve, red).  It might be interesting to load your Guiding Assistant data into PECPrep to measure the period of the spikes (it looks to me about 14 seconds) to see if that corresponds with some part of the RA gear train and to see if that provides a clue about where the problem lies.

Andy

Kenneth Crichton

unread,
May 1, 2016, 3:55:24 PM5/1/16
to Open PHD Guiding
Thanks Andy ,

I will have a look at the pecprep when I'm home this evening .

I have a c11edge with focal reducer mounted at the moment without guide scope so I'm only at 1/3 of the weight capacity . I did balance the RA well but I didn't have it east heavy (I will try that out).

I cannot find the cge forum for some reason ? This is obviously my issue but any chance you could send me a link ?

I still don't understand why the Dec is moving considerably slower than the RA ? I mean this is the third cge pro I've tested this year all with the same issue , could it be something else ??

Bryan

unread,
May 1, 2016, 8:25:49 PM5/1/16
to Open PHD Guiding
Kenneth

The only CGE forum of which I am aware is on Yahoo groups.


Bryan

Kenneth Crichton

unread,
May 2, 2016, 11:50:33 AM5/2/16
to Open PHD Guiding
Just another thought , could this be Stiction ??

Cheers Ken

bw_msgboard

unread,
May 3, 2016, 12:30:48 AM5/3/16
to Kenneth Crichton, Open PHD Guiding
Hi Ken, I don't think so. Andy has directed your attention toward the RA
behavior, which is the larger problem. Your RA gear system is in constant
motion, running at sidereal rate +/- 0.5X sidereal. Hence, there shouldn't
be any "static friction" involved. I think this is the place to
concentrate, starting with a correction for periodic error.

The short snippets of guiding in your log don't seem to show massive
problems. Are you having trouble with unacceptable images or are you just
focused on the calibration alert you are getting from PHD2?

Good luck,
Bruce
--
You received this message because you are subscribed to the Google Groups
"Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to open-phd-guidi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Kenneth Crichton

unread,
May 3, 2016, 5:37:04 AM5/3/16
to Open PHD Guiding
Hi Bruce ,

I will use PEC to start with . I was just concerned the guiding assistant said my backlash was to high to compensate and I should guide in one direction . Plus the RA and Dec are at different speeds like before ?? Should I be concerned ?

The connection to the mount kept dropping for some reason (loose connection ??)

Cheers Ken

bw_msgboard

unread,
May 3, 2016, 11:48:00 AM5/3/16
to Kenneth Crichton, Open PHD Guiding

-----Original Message-----
From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Kenneth Crichton
Sent: Tuesday, May 03, 2016 2:37 AM
To: Open PHD Guiding

Subject: RE: [open-phd-guiding] CGE POOR

 

Hi Bruce ,

 

I will use PEC to start with . I was just concerned the guiding assistant said my backlash was to high to compensate and I should guide in one direction . Plus the RA and Dec are at different speeds like before ?? Should I be concerned ?

 

Uncorrected PE in RA can affect the calibration, resulting in an RA guide rate that’s too high or too low.  So if yours ended up being too high, that could trigger the calibration alert on rates.  In the end, these alerts are just suggestions for things to consider, they’re not messages of doom.  You should judge your guiding results based on your imaging results.  We’ve seen lots of Celestron mounts that have these behaviors but people seem to do ok with them in the end.  If the Guiding Assistant reports a lot of Dec backlash, you should go with the flow and try guiding in only one Dec direction.  I don’t think you should be reluctant to try this, you can get very nice imaging results that way – many imagers do it.  

 

Attached is an example from a uni-directional Dec guiding run.  This was done with my crappy Meade fork mount and a Meade 10-inch SCT 8 years ago.  I’m not claiming this is a world-class image by any means, but I can tell you I was very happy to get it at the time.  Really, uni-directional Dec guiding shouldn’t be a big problem.

 

The connection to the mount kept dropping for some reason (loose connection ??)

 

What?  What connection are you talking about?  What error message are you seeing?  This should never happen with the serial link to the mount, something’s not right.  It could be the cable but we’d need to know what message you’re seeing and from which app it’s coming.

 

Good luck,

Bruce

Mean_M77_LRGB_1.jpg

Kenneth Crichton

unread,
May 3, 2016, 12:14:16 PM5/3/16
to Open PHD Guiding
Hi Bruce ,

During part of a guide log I posted PhD 2 came up with a warning something along the lines of "guide pulses not being received by mount " then the tracking became very good like 0.15 RMS . I disconnected PhD and re calibrated then it started to work on again .

bw_msgboard

unread,
May 3, 2016, 6:22:13 PM5/3/16
to Kenneth Crichton, Open PHD Guiding
Alert messages in this area are fairly rare and don't usually have anything
to do with the cable connection to the mount. I'm guessing you saw this
one:

"PulseGuide command to mount has failed - guiding is likely to be
ineffective."

This means the most recent guide command wasn't completed in a reasonable
length of time. It can be due to a problem with the mount controller or the
driver, it depends, and it's probably more likely to occur if a huge guide
pulse was issued. In most cases, guiding does continue without needing any
intervention from the user. In other words, it's a problem of some kind but
usually not a fatal one.

For future reference, when you see an error message that isn't entirely
clear to you, just grab a screen snapshot of the current window
(print-screen or alt/print-screen). Then, when you are done with what
you're trying to do, past the snapshot into a document or e-mail message so
we can see exactly what you saw. These messages may not mean much to an
end-user but they can tell us a lot.

Cheers,
Bruce

-----Original Message-----
From: open-phd...@googlegroups.com
[mailto:open-phd...@googlegroups.com] On Behalf Of Kenneth Crichton
Sent: Tuesday, May 03, 2016 9:14 AM
To: Open PHD Guiding
Subject: RE: [open-phd-guiding] CGE POOR

Andy Galasso

unread,
May 3, 2016, 6:30:22 PM5/3/16
to Open PHD Guiding
Also, all alert messages are recorded in the PHD2 debug log so it is always a good idea to post the debug log and let us know approximately about what time the error occurred so we have a clue where to look in the log.

Andy

Kenneth Crichton

unread,
May 4, 2016, 5:30:46 AM5/4/16
to Open PHD Guiding
Hi Andy ,

I've been looking at the debug log I posted. Would the following be of relation to loosing control of the mount ?

02:25.267 00.000 768 UpdateImageDisplay: Size=(968,608) min=0, max=255, FiltMin=6, FiltMax=228
01:02:25.267 00.000 768 Resizing image to 562,353
01:02:25.267 00.000 4296 IsSlewing returns 0
01:02:25.267 00.000 4296 IsGuiding returns 0
01:02:25.282 00.015 4296 PulseGuide returned control before completion, sleep 408
01:02:25.282 00.000 768 UpdateGuideState exits: m=6988 SNR=49.6 Saturated
01:02:25.282 00.000 768 OnExposeCompete: CaptureActive=1 m_continueCapturing=1
01:02:25.282 00.000 768 ScheduleExposure(2000,3,1) exposurePending=0
01:02:25.282 00.000 768 Enqueuing Expose request
01:02:25.704 00.422 42

Andy Galasso

unread,
May 4, 2016, 3:05:45 PM5/4/16
to Open PHD Guiding
Hi Ken,

Nothing out of the ordinary there in that fragment of the debug log. You can search for the string "Alert:" to find the error messages in the debug log.

Andy

Kenneth Crichton

unread,
May 4, 2016, 4:18:01 PM5/4/16
to Open PHD Guiding
Thanks again Andy!

The only alert i can find is -
Calibration alert details: Expected ratio at dec=14.2 is 0.970, actual is 1.424

Im assuming the debug log holds all the data from the three guide logs i made during the evening?

Im confused as i had a spiky graph for 5 minutes after calibration then suddenly everything smoothed out and i was apparently guiding at 0.15rms. It was at that point i remember some sort of alert that guide pulses weren't being received by the mount. I thought phd had reverted to a dummy run of guiding ??

I had a look at the pec software you recommended the other evening and its about 10arc sec although im going to try a longer unguided sequence before i try any correction :)

Cheers ken

Andy Galasso

unread,
May 4, 2016, 6:59:46 PM5/4/16
to Kenneth Crichton, Open PHD Guiding
On Wed, May 4, 2016 at 4:18 PM, Kenneth Crichton <kennethc...@gmail.com> wrote:

Im assuming the debug log holds all the data from the three guide logs i made during the evening?

There is one debug log and one guide log per session. They have matching file names, except for the "DebugLog" or "GuideLog" part. If you had three guide logs, then there are three debug logs.

Andy

Kenneth Crichton

unread,
May 5, 2016, 5:27:32 AM5/5/16
to Open PHD Guiding
Cheers Andy ,

I found the other debug log and I will post this evening as I'm using my iPhone at the moment.
I found the alert which is
Pulse guide command to mount has failed - guiding is likely to be ineffective.

Kenneth Crichton

unread,
May 5, 2016, 12:33:06 PM5/5/16
to Open PHD Guiding
Hi Andy,

As promised the other debug log.

Thanks for your help!

PHD2_DebugLog_2016-05-01_005559.txt

Andy Galasso

unread,
May 5, 2016, 2:06:53 PM5/5/16
to Open PHD Guiding
Hi Ken,

The debug log shows a single instance of the pulse guide command failure, among numerous successful pulse guide commands.   Although it may be something you'll eventually need to be concerned with, I believe right now it is a bit of a red herring.

The uneven "spiky" RA motion that we pointed out earlier in this thread is the primary issue: you can't guide that out.  You can see the spiky RA tracking with guide output disabled, like with the guiding assistant. See the third graph (the one from the guiding assistant run with guide output disabled) in this post.  I personally do not know enough about CGE Pro mounts to suggest what could be the cause of the uneven RA tracking. One clue we have is that the period of the spikes is about 14 seconds (you can get the period more precisely by loading the guiding assistant run section of the guide log into PECPrep).  Sometimes knowing the period can point to a specific component in the gear train.

Andy

Kenneth Crichton

unread,
May 6, 2016, 6:39:52 AM5/6/16
to Open PHD Guiding
Hi Andy ,

I've had a look on pecprep and the periodic error is 10 arc sec is this what your referring to? I gather from your wording "period of the spikes" your meaning this is only happening during a particular period of the tracking ?

I'm assuming I would need to do a longer unguided sequence with PhD before I could correct periodic error ?

Near the end of the unguided sequence you highlighted for me, there is a period of a flat graph , should I delete this before attempting pec ? What is it actually ?

Ken Self

unread,
May 6, 2016, 9:37:41 AM5/6/16
to Open PHD Guiding
In your guide log you start with guiding on then there is a period with guiding off then it is switched on again. The flat bits are where guiding is on.
If you remove the guided portions of the data you've got a bit more than 2 cycles with 8.5 arcsec p-p (+/-4.25). A couple more cycles would be better but that is still a pretty good PE.
The calibration looks a bit out but I wonder if that is due to the small pixel scale which means the periodic error, small though it is, is having an impact as it is equivalent to over 3 pixels.
With a bit of number crunching it might be possible to work out a better calibration angle based on the RA drfit - PECPrep gives a trend figure of 0.0191 arc-seconds per second

Andy Galasso

unread,
May 6, 2016, 7:45:00 PM5/6/16
to Open PHD Guiding
Ken,

If we look at your guiding run RA data again, you can see that it has two major periodic components.  There is a long period slowly varying error with an amplitude of perhaps 8-10 arc-seconds. That slowly varying error is completely normal and easily corrected by PHD2 and/or PEC.  This is not a problem:

Inline image 2

Superimposed upon this gradual oscillation is a higher frequency oscillation. The high frequency oscillation is what I was calling "spikes". Here are some of the spikes highlighted with vertical orange lines:


Inline image 4

These oscillations are a problem because PHD2 will not be able to correct them since they are large -- about 5 arc-seconds --- and rapid.  I would recommend focusing your attention or correcting these spikes.  The observation that the spikes have a period of approximately 14 seconds (based on a quick visual look at the log) may be a clue regarding the mechanical origin of the spikes.  PECPrep can measure the period of the spikes more accurately and that was what I was recommending using PECPrep to do. I would not bother using PECPrep for PEC yet until you can eliminate or reduce the 14-seconds spikes which are giving you the problems with calibration and guiding.  The trick is going to be to find the mechanical origin of the spikes and fix the mechanical problem.

Andy

Kenneth Crichton

unread,
May 7, 2016, 4:50:02 AM5/7/16
to Open PHD Guiding
Thanks Andy and Ken !

So the issue is mechanical :(
I will need to contact celestron! being that it's a new mount under a 2 year warranty I would void this if I opened the gear box etc . Unless anyone knows an alternative?

Ken Self

unread,
May 7, 2016, 6:33:31 PM5/7/16
to Open PHD Guiding
I edited and loaded the unguided section of your log and loaded into PECPrep. Attached is the PECPrep formatted version that you can load into PECPrep as a PERecorder (RA) file

The main PE is the second harmonic of the worm drive  - the peak at 167 seconds period
The spikes Andy noted are the peaks at 16.9 and 32.2 seconds period. I couldn't find any info on the CGE-Pro gear train but at a guess I'd say the 16 second period is a gear mesh period and the 32.2 is a subharmonic of that.
You can apply a high pass filter at anywhere around 40-75 seconds to isolate the spikes. What shows up is that there is a big spike at the same point in each worm cycle with smaller ones in between. 
So, with lots of caveats, it could be some grit or a burr on one of the gear teeth causing the big spikes. 

PHD2_GuideLog_2016-05-01_010145B_RA_EQMOD.txt

Kenneth Crichton

unread,
May 14, 2016, 3:56:25 AM5/14/16
to Open PHD Guiding
This is the response from celestron, so looks like I need to loosen the RA meshing.

"Hello Kenneth,
Thank you for your patience.

Please see response below. I have attached instructions on how to adjust the mount. Before attempting any adjustment, contact the dealer to find out about your specific warranty if the mount is brand new.


"It’s possible the R.A. gear is a little too tight and binding a little. If the tension is backed off a tiny but, it can dramatically improve the tracking. The overall drift over the 14-minute track log is actually pretty normal, as there is instrument flexure that exceeds any periodic error. But spikes or jumpy movements in tracking is something the mount should not do."

Kenneth Crichton

unread,
Oct 8, 2016, 2:20:16 PM10/8/16
to Open PHD Guiding
Just an update ....

So Celestron initially made some adjustments to the RA which although not great I eventually decided just to get on with it . I got to the point recently where I was consistently getting between 0.7 and 1" rms which i was pleased with. Although just the other night the ra motor decided to start jumping (10" spikes every 10-15 mins) you can see my video of the spur gear attached to the Pittman motor . I contacted Pittman because I measured movement of the motor shaft to be 0.5mm and pecprep was telling me this was where my issue was coming from . Pittman US told me that the motor should have a maximum play of 0.2mm and any more then the motor is likely faulty or at end of life . so I've contacted Celestron who have asked me to return the mount . I cant help but think this was always the issue that was causing the poor guiding (the motor was on the way out ) as I noticed this movement of the motor shaft months before . Anyway fingers crossed for a good outcome and thanks to David Hinds for a speedy response !

Kenneth Crichton

unread,
Oct 9, 2016, 5:43:49 AM10/9/16
to Open PHD Guiding
Oh just for clarity the mount spur gear hasn't always moved like this. When I'm talking about 0.5mm play in the motor shaft it's side wards movement which you cannot see in the video . The clicking of the spur gear in this video has started to happen recently and in my opinion it is all an issue with the motor . There is considerable backlash in the gear so it's not stiction or an overly tight meshing . Obviously backlash is never an issue in RA anyway .

P.s. the guiding in Dec has been exceptional since it was adjusted for backlash , I never get over 0.5" guiding.it was only the RA letting it down somewhere between 0.8 and 1.4" . Backlash has been measured in PhD to be 1 second so I'm very pleased with that !
spur gear spike.mp4
Reply all
Reply to author
Forward
0 new messages