CEM70 Guiding Troubles- Large RA peaks causing Elongated and Bad Calibrations

[Vince Tambellini]

Hello,

Here is the link to my logs: 

https://openphdguiding.org/logs/dl/PHD2_logs_yd2v.zip

Basically, I have been having troubles for the past few weeks on my CEM70.

I am running:

840mm scope at f7

ASI2600mm camera

OAG with 220mm guide camera

CEM70

Summary:

I am getting decent overall guiding, around 0.4RMS, but the peaks in mostly RA and now DEC are elongating my stars, in especially RA. They are very frequent. I have worked out some of the error, so now the elongation is less than it was, doing things like changing the guide settings. However, I can still see some smearing of stars at 60 second exposures, and at 5 mins they do not look great. There has been a weird period wave on + and - error of RA, and now DEC has really huge adjustments frequently.

Current status:

I do not have an issue with Polar alignment. The dec backlash might be an issue or an artifact of that period stuff.
A lot of the parameters are pretty well fined tune now. Ignoring that first period could be a few minor things including chasing seeing or the dec backlash which may or may not be real. Need some help on the RA axis and setting changes probably are not going to do much for it (at least not with PPEC) at this point. Hard to say though and unsure. 

On the logs:

One of the guide logs really started at around 1 or 2am after tuning settings. The one from tonight is much shorter. Another issue is whenever I use 2 seconds things get a lot worse, so been using 1 seconds I suppose. m

On Calibrations:

Also, calibrations tend to be poor- it also seems like phd2 when trying to point to 0/5 for best calibration, that's actually the incorrect location- so I've been manually slewing to 0/5 then calibrating. Also, the calibrations are almost never orthogonal. 

Not sure what else to mention- please let me know if you need additional information. 

Thank you,

Vince T

[Vince Tambellini]

Just wanted to follow-up here.I would really appreciate it if someone could assist with looking at my logs to pick anything out or find any suggestions to improve things. 

Summary:

My DEC corrections are huge, > +-2 or even 3" around every few seconds or so, even though the average for DEC is 0.41".

For RA, I'm getting many spikes > +-1" probably every 5 seconds as well- which is looking slightly okay on 60 second subs but really bad on 5 min subs. There's a lot of bouncing around that makes it not really possible for 5 mins subs. Below is from tonight. Logs in the previous email were from 2 previous nights. 

I'm not sure if these dec corrections are even necessary, are a signal of something else wrong physically, or a setting somewhere is off. 

Thank you kindly!

Vince Tambellini

--
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/13908362-7fb9-416a-afe2-ae2a2df07268n%40googlegroups.com.

[Bruce Waddington]

It's been less than 24 hours since you posted your first complaints so I think it would be good if you could have some patience.  I've already spent a considerable amount of time looking at your logs and I would say you have presented us with a smoking mess.  The consequences of that are that it's going to be difficult to figure out exactly what's wrong although it's already clear it's a combination of mechanical problems with the mount, operational errors, and some misunderstanding on your part about what you're seeing in the graphs.  I will continue to work on this but I don't expect to have a response before tomorrow.

Bruce

[Vince Tambellini]

I apologize, Bruce. Thank you very much for your help. 

Vince Tambellini

To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/da63a3b4-527f-4b88-bc60-faf03745dd58n%40googlegroups.com.

[Bruce Waddington]

This is going to be a long response.  We can start with some operational problems and mis-conceptions you have:

1. Your guide camera is running in 16-bit mode but you've specified a saturation-adu of 255, which is only appropriate for 8-bit cameras.  This degrades the auto-find procedure and forces PHD2 to use sub-par guide stars.  Change the saturation-adu value to something appropriate near 65500.

2.  The Calibration Assistant will always do a better job than you can.  When you slew manually, you're probably not clearing the Dec backlash/stiction sufficiently in your mount which leads to problems such as orthogonality error.  In one of your calibrations, you got a lot of lost-star events so obviously that calibration was useless.  If you have some reason to want the calibration slew to go elsewhere,  just change the 'target' values in the CA window - that's why they are there.  You can even save and re-use them as a "custom" location.  The important thing is to stay close to Dec=0 and not pointing down close to the eastern or western horizon.

3.  Your worry about large Dec corrections is unwarranted.  Those are Dec backlash compensation pulses and yes, they are needed for your mount because of its mechanical behavior.  You should focus on the results of the guiding, especially the RMS values of both RA and Dec rather than guide pulse sizes.

4. Your fiddling around with guide settings has been generally futile and in the case of PPEC, counter-productive.  The problems you're having at the moment are coming from the mount and you will not be able to mask them with guiding.

5.  The last guide log you sent was done with shift-lock enabled in NINA, presumably because you were trying to track a comet.  This makes it nearly impossible to analyze your guiding and/or mount behavior.

Now to the mount-related mechanical issues:

There is a lot of periodic error in RA, I would say considerably more than what is expected with an iOptron mount:

This is the unguided tracking behavior of the mount showing a peak-peak periodic error of nearly 30 arc-sec.

It is mostly attributable to a periodic error that roughly matches the worm period on the mount:

This large RA periodic error creates RA drift during calibration which is probably the main source of your orthogonality errors. Although PHD2 is doing a reasonable job of trying to cope with this, it was hampered by your fiddling with the guiding parameters.  This, coupled with using 2-sec exposures, leads to pretty consistent under-correction in RA - PHD2 was generally falling behind the tracking errors:

I think you should discuss the uncorrected PE with iOptron - I think they will agree it is too large and will help you improve it. They are certainly familiar by now with the kinds of graphs I'm showing you.  In the meantime, you can try to help PHD2 manage it a bit better by using 1-sec exposures and using PPEC parameters of 0.9 for predictive gain and 0.7 for control gain.  Set the period length to 348 and disable auto-adjust.  Again, this is just a temporary measure, you should find a way to improve the mount mechanics.  We have seen many cases where iOptron mounts were simply meshed too tightly and could be improved rather easily.  I suspect this is also what's behind the occasional large RA excursions you are seeing.  Beyond that, I strongly discourage you from doing any other fiddling around with guiding parameters.

If you really want to get a clear view of the mount's limitations, you can follow the procedures in the attached document.  Note that the procedure must be done with no imaging and no involvement from your imaging application.  I would probably extend the recommendations in the document to suggest getting 20-min guiding sessions on both sides of the meridian.

Regards,

Bruce