PHD2 and encoder mounts ( AP MACH2) best settings and algorithm

[Andluc]

Hi,

I've just tried PHD guiding with NINA, and I am  new  with the program.

Even if I was able to get round stars, the guide was not on par with what I had till now.

I suppose it depends on the parameters and algorithm.

I noticed that Lowpass2 is suggested in DEC for this kind of mounts.

in my case, errors were oscillating both in  AR and DEC.

Worth to mention , the corrections were very big, about 3-4 times the magnitude of the errors in the cycle before.

this is from my log:

X guide algorithm = Hysteresis, Hysteresis = 0.100, Aggression = 0.700, Minimum move = 0.190

Y guide algorithm = Lowpass2, Aggressiveness = 80.000, Minimum move = 0.190

Backlash comp = disabled, pulse = 20 ms

Max RA duration = 2500, Max DEC duration = 2500, DEC guide mode = Auto

I see in the log aggression 0.7 (0-1) and aggressiveness in LP2 80.000 (0-100) (is it ok this scale difference )

It would be help ful if someone could post here the best algorithm to date for AR and DEC axis and the parameters to key in as starting point.

I am attaching a log if useful. (real guiding start at 23:43

Thank you in advance

Andrea

[Andluc]

https://openphdguiding.org/logs/dl/PHD2_logs_rtWP.zip

[bw_msgboard]

Hi Andrea.  In general, well-made mounts with absolute encoders don't benefit from aggressive (rapid) guiding and some may actually suffer from it.  You were trying to guide with 1-second exposures and that's not a good approach with your mount, mostly because you were probably just chasing seeing effects.  Mounts such as yours often produce the best results with long-cadence guiding, meaning that guide corrections are issued on intervals of, say, 5 to 10 seconds.  The goal is to guide out the low-frequency errors that arise from polar alignment error, atmospheric refraction, flexure, etc.  There's no need to use rapid guiding because the encoders should be keeping the mount on track.  This guiding cadence is established by using two parameters: 1) the guide camera exposure time and 2) the time lapse property.  If you use longer exposure times, you will dampen the effect of seeing variations right away - a good starting point for you would be 2-3 seconds or so.  Then you adjust the time lapse property to create the final delay between guide corrections: camera exposure time + time lapse value.  If you find that the guide star drifts too much in declination, you can reduce the time lapse value or improve the polar alignment.

 

You've also missed some other important best practices that will improve your results.  First, you should be using multi-star guiding which means you must allow PHD2 to make its own selection of guide stars - don't click on a particular star, click on the 'Star' icon to force an auto-selection.  Then, especially when you're getting started, run the Guiding Assistant to see how the mount is performing on its own and also to get a recommendation for appropriate min-move values.  In addition to the exposure time setting, the min-move values are your best protection against chasing seeing.  You should also specify a min-HFD value for guide stars to help PHD2 distinguish between faint stars and camera sensor noise. 

 

All of that said, your guiding wasn't bad, not something to be too excited about.  In the 3.5 hour final session, excluding the problem at the end where you simply lost the guide star, your total guiding RMS was 0.6 arc-sec with a close match between RA and Dec.  So you should be have been getting nice round stars.  It looks to me like this was pretty much seeing-limited. It's not uncommon, particularly in the summer, to see substantial changes in the seeing conditions from hour to hour - so you can't expect results to necessarily remain the same all night or from night to night.  If you follow the recommendations above, I would expect to see your guiding results improve but of course you will always be limited by the local seeing conditions.

 

Regarding your question about the different scales for aggressiveness, that's an inconsistency we should fix in the UI.  But it's only a UI problem, in your example the RA was being guided with 70% aggressiveness and the Dec with 80%.

 

Here are some good reference materials in case you haven't seen them:

 

https://openphdguiding.org/documentation/

 

https://openphdguiding.org/phd2-best-practices/

 

Hope this helps,

Bruce

 

---

From: open-phd...@googlegroups.com [mailto:open-phd...@googlegroups.com] On Behalf Of Andluc
Sent: Tuesday, September 14, 2021 3:17 AM
To: Open PHD Guiding
Subject: [open-phd-guiding] Re: PHD2 and encoder mounts ( AP MACH2) best settings and algorithm

--
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/5eaad0fc-0283-412b-8c97-fb0469f2cdf0n%40googlegroups.com.

[Brian Valente]

Hi Andrea

i have an AP1600 with absolute encoders and it acts very similarly to the Mach2. There was a healthy discussion with Roland over at the AP forums. (you might check over there)

After quite a bit of testing we both concluded guiding is a very light touch with encoder mounts. Roland refers to it as "bump" guiding, but it's basically doing a guide exposure, and then pausing about 10 seconds between guide exposures (the setting is 10000ms in time lapse under advanced settings->camera)

Regular guide exposures (one after the other, 2-5 second exposures) show considerably worse performance in both our experiments.

If you would like more details you are welcome to contact me directly

Brian

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/AB442ED66AEB42B4A1EFB716F503382E%40HomeDesktop.

--

Brian 

Brian Valente

portfolio brianvalentephotography.com 

[Andluc]

This is all very help full.

Bruce, I fully agree with you. it was the first time with PHD for me and I was lucky enough to get one image beside my total ignorance.

Now I have some clear directions to understand better the program and fine tune. Thank you.

The only thing I don't understand at present is the magnitude of the corrections applied. It is just a matter of aggressiveness?

Brian, I will have a look at the discussion and in case I will drop you an email.

Thank you very much

Andrea

[Andluc]

Forget my comment on Correction magnitude:

I've seen the log and it makes sense. I don't know why in NINA the graphic showed them amplified.

Andrea