CEM26 guiding issues

[Tom van Peer]

Hi All,

Up until last summer guiding with my 4 year old CEM26 were pretty ok, usually an RMS around 1". Since I move my complete setup from my home to the garden, I typically did a calibration followed by GA and that usually gave some advices for min-move but never anything to panic about like changing dec guide mode. Then when nights got longer, the guiding got worse, up to around 2". Since the mount was already 4 years old, I thought it might be a good idea to replace the belts. After that the real trouble started: very erratic guiding and a huge DEC backlash between 2 and 3 seconds, so it seems.

I tried several things, tightening the belts, loosening the belts, tighten the clutches, then loosening them again, but that didn't help. I tried PEC, tried resistive switch instead of PPEC for RA, nothing seems to help. If I engage the gear switches, I can't feel any play.

I have uploaded a session where the GA ran way longer than my mounts RA period, which is 600 seconds.

Can someone please help to explain what the graphs are showing and how it could relate to mount issues?

The logfiles can be found here: https://openphdguiding.org/logs/dl/PHD2_logs_eLrF.zip

[Brian Valente]

Hi Tom

you have a lot going on. Here are some observations:

- i suggest you recreate a new profile using the profile wizard. Somehow you ended up with resist switch for both axes, that isn't a good fit for RA and the aggression settings don't make sense.

- i suggest you re-do your calibration using the calibration wizard. The guiding assistant warned you about a number of problems including guide camera focus and rates mismatch, you need to address those and not skip over them. On the rates specifically, look at what your RA is doing (blue) - the first 9 calibration steps started on the opposite side. When you slew to the calibration sky area, let it settle for about 30 seconds (a bit long for now, but see how that goes). it could be your ra drive has backlash and needs to take that up. 

- any reason you aren't using a dark library for your guide camera?

- your seeing appears to be pretty poor? See the graph below, your guidestar mass is in yellow (star snr matches this). guidestar values here should usually show flat line with high frequency SNR, but yours is all over the place. 

- increase your minimum snr for auto find to 30 or higher. There were a number of runs later on where your terrible results were because it was guiding on a non-star. See the star mass below: from one of those runs. You can see it goes to nearly 0 and any guidepulses are nonsense at that point.

aside from those operational issues, phd did reasonably well guiding.

In RA you have a couple higher frequency residual (uncorrected) errors, one at 119.6 seconds, one at 21.3 seconds. They aren't enormous but they are the constraint to better RA guiding

Your Dec backlash is measured at over 3 seconds, which is more than you want. so look at tightening up your dec axis (belt tightening?) 

so overall i'd summarize as:

- cockpit errors in PHD, fairly easy to fix, start with new profile via wizard, new calibration, resolve the issues the guiding assistant points out (and mentioned above)

- once you address those i suspect things will improve considerably, and then you can decide if the mechanical fixes are worth pursuing. It depends on your imaging setup's image scale and what RMS you can live with.

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 visit https://groups.google.com/d/msgid/open-phd-guiding/ad9d9a14-6966-4ddf-96f7-e46c2d658ea4n%40googlegroups.com.

--

Brian 

Brian Valente

astronomy portfolio https://www.astrobin.com/users/bvalente/ 

[Tom van Peer]

Hi Brian,

Thanks for all the advice. I once again opened up my mount yesterday, to get rid of some friction in DEC, greased the wormwheels (probably for the first time in 4 years), and checked the belts (should be fine).

Although the seeing last night was again quite bad, I managed to get a calibration done, and a run of GA with a new profile as suggested. I also created a new dark library. I don't know why the old one wasn't used. I had been playing around to create a bad pixel map, and probably had not switched the dark library back on after 'playtime'.

You can find the file here: https://openphdguiding.org/logs/dl/PHD2_logs_5QRq.zip I had the SNR auto find set to 30, but switching back to 6 made it possible to do multi star selection. I think the graph of the later runs where PHD was guiding on a non-star, were probably caused by the telescope pointing to the ground... 

The calibration took a couple of attempts due to lost stars, but finished without complaints/suggestions. The graph looks better, as the RA 'excursion' to the opposite direction is gone now. What are your thoughts on this result?

Then I ran a GA for 11 minutes, but I once again had quite a few star losses, so is that reliable? The DEC backlash is still high with 2396ms. So still work to do.

As for the frequency spikes at 21.3 and 119.6. The latter is almost exactly 1/5 of the worm period. Could that be related?

I then tried to let guiding run for a while, but seeing only got worse and I broke it off after 10 minutes.

Here is a noob question: PHD2 had quite a few star losses, which was no surprise with all the high clouds, but sometimes it reported 'star lost' even though I could clearly see a star in the Star Profile window. What is happening that to make PHD2 decide that the star is lost? And how do star losses impact calibration and GA?

Once again, many thanks for your help and I hope I can further improve the guiding with the help of your analysis.