Help with Auto-PEC on EQ6 Mount

[Howard Schneck]

I don't know if this is considered off-topic but I am trying to improve my RA guiding by turning on Auto-PEC on my EQ6 mount.  Unfortunately it doesn't seem to improve things..  I've imported the PHD log file into PECPREP to try and understand what is going on and how to generate a good correction file but I guess I don't understand exactly how to do this despite looking at the  EQMOD VS-PEC manual.

Here is my log file.  I typically get oval stars because my DEC guiding is sub 1 arc-sec but the RA is > 1 arc-sec.

https://openphdguiding.org/logs/dl/PHD2_logs_4sbD.zip

Thank you for any help you can provide.

Howard

[Terry Fitzpatrick]

Hi,
I've done this on an EQ6-R and most recently an AZ-EQ5, and the process is relatively straightforward with EQMOD.
1. Make sure PPEC is turned off in EQMOD.
2. Start guiding on a star, probably ideally near zero DEC (within 20 degrees), and let everything settle.
3. In the EQMOD control panel, change to the PEC display using the green plus sign in the upper left corner.
4. Hit the record button (red circle) and go and have a coffee / beer as the process runs over five worm cycles, which is around 50 minutes. When it's finished, EQMOD does an analysis of the accumulated data to produce a PEC curve which it then applies - you should note that tracking rate should be Sidereal+PEC.
5. Stop PHD2 guiding as EQMOD will be supplying corrections based on the curve it's generated.
6. Open up the right-hand part of the EQMOD panel and on the right, check the box that says ”Record PPEC” and leave it for a while. This could take up to 20 minutes, so I'd leave it for two full worm cycles as training will only start when the encoder on RA reaches a pre-defined position, but I don't know what that position is). The flashing light on the mount shows the status, but you'll need to check the manual to figure it out.
7. Once it's done, delete the PEC file from EQMOD otherwise you'll be applying it twice, and remember to Enable PPEC at the start of each session.
This gave a brilliant result on my AZ-EQ5 recently, so I may have a go at redoing the EQ6-R Pro in the observatory.

[Howard Schneck]

Terry,

Thanks you for replying.  I will give fowl your steps and try it out.  When you say brilliant result, are you below 1 arc-sec?

Howard

[John Sim]

Hi Howard

The final steps in Terry's advice are to do with storing the PEC curve in the mount itself and let the mount apply the corrections - permanent PEC. 

If you are going to be using EQMOD anyway (the only alternative is Skywatcher's ASCOM drivers, or ST4), I would begin with using EQMOD to apply the PEC and then consider permanent PEC later after you have everything sorted to your satisfaction. Personally I don't see that permanent PEC has any advantages over EQMOD applying PEC but opinion is divided on that point.

Re setting up AutoPEC, I would train it over the full 9 cycles rather the minimum of 5. This should give a cleaner and more accurate PEC curve. I looked at your logfile using PHDLogViewer and checked its frequency analysis of the long 3h 47m guiding session. It shows a single clean peak at around 480 seconds which is the worm period for the EQ6R and should be very easy for PHD2 to guide out, and later for PEC to correct.

However before you embark on PEC I think you need to take a good look at your guiding. Your RA guiding is an endless sawtooth of corrections. I note you are using Predictive PEC as the RA guide algo and I am not familiar with that algo and its configuration but a sawtooth pattern like this usually indicates that guide corrections are too aggressive. 

I have the EQ6R too, and the similar AZ-EQ6, and I use the default hysteresis RA guide algorithm with low settings for aggression. I have experimented with PEC but found that it made little or no improvement to my overall guiding. I do find that these mounts generally work better with short, 1s, exposures and I only increase the exposure time if that is the only way to improve the SNR of the guide star. For me the biggest factor is seeing and my RMS can be anywhere between 0.6 and 1".

My recommendation would be that you change from the PPEC algo to Hysteresis and experiment with its settings (i.e. aggression and MinMo) to achieve the best guiding and only then try PEC, or revert to Predictive PEC but this time with a better idea of what your mount is capable of.

cheers

John

[Terry Fitzpatrick]

Howard,

It depends on seeing obviously, but I can get 0.8 arc-secs or so using a QHY5LII-M guide camera on and Orion 240 mm FL guide scope. I have considered moving to an OAG, but since my main scope is only 1000 mm FL, I'm not sure that I would get any improvement - in fact, I'm sure it would just introduce a million other problems for me to have to sort out.

Terry

[Terry Fitzpatrick]

Howard,

Yes, you're right, I was applying permanent PEC. I used to use PEC in EQMod on my older EQ6-Pro mount (no PPEC), and it worked well, but if I ever lost power or had any kind of interruption before parking the mount, I lost synchronisation and would end up having to re-record the PEC file which was a bit of a nuisance. Now I have a newer EQ6R-Pro and an AZ-EQ5 GT both of which support PPEC, so I no longer have that worry. The EQ5, in particular, seems to have a very smooth PE curve so running EQMod PE recorder over 5 cycles gave a very smooth curve to program into PPEC. I take your point though that it is probably a good idea to use EQMod's playback facility to check out how good the PEC curve is before recording it to the motor controller.

Terry

[Howard Schneck]

Terry,

Thanks again for your feedback!

Take care,

Howard

-- 
You received this message because you are subscribed to a topic in the Google Groups "Open PHD Guiding" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/open-phd-guiding/osFzkxi8lv8/unsubscribe.
To unsubscribe from this group and all its topics, 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/beac16e7-1cb4-4b7a-94db-6cb344a466d9o%40googlegroups.com.