RA Issue with CGX-L

[Norman McCall]

Hello all,

I've been fighting (for months) on my own to get my CGX-L mount and C11 EdgeHD SCT system to Auto Guise well with PHD2 for astrophotography.

I've believe I have the balance right (East heavy & Front of scope heavy) and seen great improvements in Dec.  I akso adjusted to a 3 sec exposure which also helps.  I have not yet applied PEC as the last time  itdid not help.  (Although, with tonight's results in Dec I believe this is the next step.) Tonight I tried to go through a variety of small changes to monitor and keep what worked. 

If someone can make a suggestions for RA that would be appreciated.

BR,

Norm

https://openphdguiding.org/logs/dl/PHD2_logs_3axG.zip

[Brian Valente]

Hi Norm

sorry to hear you are having problems.

i don't know if i have a definitive answer for you, but i do have some observations

generally you seem to be on the right path - the calibrations are in the ballpark, and once you have a good calibration, you do not need to continue to recalibrate

Your DEC could benefit from turning on backlash compensation. I think there was that recommendation in one of your GA runs, but it looks like you turned it off later one. For right now, that's not critical because DEC is outperforming RA by a factor of 3, so improving it right now will only make your stars more oblong - just something to keep in mind for future

RA is definitely your challenge. Specifically you have a ~21 second oscillation - it exists in both your guided runs as well as your unguided runs (guiding assistant) and results in this thrashing that looks like this (segment 9 from your guidelog):

Normally this kind of thing would be the result of some bad PHD parameters, but I don't really see any in your case. Since you can see this without any guiding, i'm thinking the mount has some sort of 21.5 second issue that is getting worse when guided. your best bet is to track that down with celestron. All the other recommendations i offer have to do with minimizing that in your guiding:

I can see you fiddling with the RA hysteresis aggressiveness, which is a good instinct and may help alleviate this some. But small changes a few points aren't going to make much of a difference. i'd say try for a high value and see if that helps, maybe 70-80. 

For such a short periodic error, your longer 3 sec exposure isn't going to help you to tackle this. You might try 1 second exposure with the higher hysteresis aggressiveness and see how that performs- Another possibility is you can try the PPEC algorithm and set a fixed period length of 21.5 seconds (turn off auto-adjust period). this is a bit of a brute force approach, but it could be effective at 1 sec guide exposure.  there's definitely a tradeoff in there as far as chasing seeing with a shorter exposure, but right now i think this thrashing is the limiting factor.

Your GA run also suggested better focusing of your guidestar so you might look into that as well

There are other people who may offer more insight on why this happening, but if you can solve this thrashing, and then add back in your dec backlash compensation, you'll be in good shape

Brian

--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/71d6f240-b61f-4fed-a6cb-f6b49af58a78%40googlegroups.com.

--

Brian 

Brian Valente

portfolio brianvalentephotography.com 

[Brian Valente]

PS Norm-- i did a little searching and realized the CGX-L is a belt-driven system. Small problems in the belt could definitely explain something like this. you might look into celestron user forum or contact celestron regarding this 21.5 second problem and see if there are any belt-related issues or fixes you could apply there. 

hth

Brian

On Tue, Sep 24, 2019 at 9:04 PM Norman McCall <njmc...@gmail.com> wrote:

--

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.
To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/71d6f240-b61f-4fed-a6cb-f6b49af58a78%40googlegroups.com.

[Andy Galasso]

Hi Norm,

I agree with Brian that you should contact Celestron for help with the mount.   The RA drive is oscillating with a 21 second period moving back and forth by 5-6 arc-seconds or more each cycle.  This is a mechanical problem in the RA drive that really cannot be corrected by guiding.  The RA axis is simply moving too fast for guiding to follow -- around 2 arc-seconds per second or more.

(As a side note, I will mention that without any mechanical fixes, you will get your best results by guiding RA aggressively: keep your guide exposures short, about 1s, lower your RA min move to about 0.1 px, keep your hysteresis low (10% or lower) and set your aggressiveness high, about 70-80%.  With these settings you might be able to reduce your RA RMS error in half, to about 0.8",  but you're still going to get significant star elongation in your imaging camera and you'll ultimately need to find a fix for the underlying mechanical problem.)

Andy

[Hugo Post]

I had a brand new 2 times a  CGEMII, the RA and DEC where mutch to mutch screwed down, now it turns smouthly just as hypertuned.

Pls read this:

I assume that you are talking about the black metal rings that are hidden by the counterweight bar and the polar scope cover.  It is OK to loosen those up a bit.  Sometimes it makes a difference sometimes it doesn’t.  Celestron does not publish and torque specifications for this and I do it by feel and experience.  In general, I tighten the rings by hand to the point that the axes start to get stiff.  Then I unscrew the rings slowly until the axis is reasonably loose.  Do not back the rings out so far that the axis shaft can move up and down or side to side.  As long as you do those things you should be fine.  You will want to recheck things once you place a load on the mount.

Op woensdag 25 september 2019 06:04:00 UTC+2 schreef Norman McCall:

[Hugo Post]

In the beginning I had a AVX with oscilation problems, I did get my money back and buy a CGEMII.

The first CGEMII  had some sand inside the DEC axes and the next one was to mutch screwed down.

Now it is very good, Guiding below 0.8.

Regards Hugo

Op woensdag 25 september 2019 06:04:00 UTC+2 schreef Norman McCall:

Hello all,

[Norman McCall]

Hugo, 

You make an interesting point.  I never thought about the friction in the rotation of RA (and to a lesser extent Dec).  I do note RA does not have a free mechanical rotation. In fact the balance on each side (East & West) differs due to friction.  I will look more carefully at the mount to determine if there is an easy way to improve the mechanical rotation. 

Norm

--

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.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/8318ebfc-e0b3-481e-a013-6fb1aa577021%40googlegroups.com.

--

___________________________________________________________________________
Norman McCall, London ON Canada  M: 519-630-6366 Email: mailto:njmc...@gmail.com

[Norman McCall]

Brian, 

Thank you for your thoughtful analysis.  I need to look at the segment of the graph more carefully tonight at home.  Currently I have a hard time seeing a 21 sec. harmonic. To me I see a lot of harmonics but no clear fundament waveform.   For example, to make the analysis less subjective does there happen to be any s/w that does an analysis of the (or a spreadsheet) that derives the fundamental frequency mathematically? 

Firstly I was wondering which program utility are you using to do the analysis.  As I note you have a time-stamp across the top which I do not see in my use of the PHD2 Log Viewer program.  (I therefore thought is must be another utility.)  

As Celestron claims from the tracking data I supplied them that "the mount is fine", I'd like to present them good data that proves otherwise. Hence my question on the 21 sec harmonic.  

Lastly, as It stands right now (having struggled with this for 8 months without much success) if I cannot expect reasonable performance from the CHX-L, I think I'll move to the Paramount MyT once I've saved up the money. (Note, in Canada things cost 35% more than in the US due to the exchange rate.)  

Best regards,

Norm

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i4fYO059WGQ4LHxz%2B%2BM-yy630W8RoKk1VYpgZO0%2BrxMXA%40mail.gmail.com.

[Brian Valente]

sure great question

i am using the PHD logviewer, which is freely available https://openphdguiding.org/phd2-log-viewer/ 

I can bring up the raw RA performance, and the 21 sec fundamental sticks out pretty clearly:

even during the GA run (where no corrections are applied) you can still see it (the shape and visual size are a little different, but it's roughly the same magnitude)

so that tells me there's something in the mount, and it's not interacting well with PHD

After reading about the belt drive, and assuming that's true for you, i would imagine there are some adjustments or feedback specifically around that mechanism. I've seen other mounts where a short periodic error like this was caused by belt-related issues or mis-adjustments. 

hth

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAD-pWYv5tVx7fOvyA32OB8Awoi1PvPx4HsUVCkJwsk668DtUHw%40mail.gmail.com.

[Norman McCall]

Brian,

 

Thanks for the help in using PHD Log Viewer. I now see how to do the analysis myself.  (I had been looking on the menu for a selection w/o success.)  

 

I now need to search on-line for help regarding the harmonic or contact Celestron to see what they can offer.

 

BR,

Norm

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i6NGejW9UpSRt9796diuJzixKXvtEoPZs5q9Nh44q5f8Q%40mail.gmail.com.

[Norman McCall]

Andy,

 

Great and helpful input.  I’ll apply your suggestions and see what happens.

 

Thanks,

Norm

 

From: open-phd...@googlegroups.com <open-phd...@googlegroups.com> On Behalf Of Andy Galasso
Sent: September 25, 2019 2:20 AM
To: OpenPHD Guiding <open-phd...@googlegroups.com>
Subject: Re: [open-phd-guiding] RA Issue with CGX-L

 

Hi Norm,

--

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.

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAKemMNUwKNOJKi4aVs3_DmwRATPP46RRWXvJdCXzJMXkYs_JTQ%40mail.gmail.com.

[Norman McCall]

Brian,

 

Firstly, I have greatly appreciated the feedback which has allowed me to focus onto what might be a core issue with my CGX-L mount.  I have also learned ho to access features o the PHD2 Log Viewer which I was unaware of.

 

I thought I’d give you some feedback on the harmonic issue and what I believe might be the source.  

 

The period of the harmonic is actually 21.3s and given that one period of the RA worm gear is 300 seconds, then 300/21.3 ~ 14.0 seconds.   In looking at the drive belt I note there are 14 sprockets on each drive wheel.  I therefore believe the issue is that imperfect meshing of the sprocket and the belt causes a bump in RA guiding.  Probably it is either an alignment issue with the belt and the wheel(s) or a poor fit with the belt tread into the sprocket (for example If the belt has stretched). 

 

I have contacted Celestron TS and hope they will provide useful guidance. 

 

Norm

 

From: open-phd...@googlegroups.com <open-phd...@googlegroups.com> On Behalf Of Brian Valente

Sent: September 25, 2019 12:40 PM

To: Open PHD Guiding <open-phd...@googlegroups.com>
Subject: Re: [open-phd-guiding] RA Issue with CGX-L

 

sure great question

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/CAJa45i6NGejW9UpSRt9796diuJzixKXvtEoPZs5q9Nh44q5f8Q%40mail.gmail.com.

[Brian Valente]

Norm - great sleuthing and exactly what i'd expect the answer to be. 

hopefully it will be a simple mechanical adjustment or fix for you

well done

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/132e01d57626%2473de9720%245b9bc560%24%40gmail.com.