Help improving guiding

[macd...@yahoo.com]

Last night guiding didn't fail, but it wasn't clean, either.  Here are my logs:

https://openphdguiding.org/logs/dl/PHD2_logs_i5de.zip

And here is my gear:

Mountain Instruments MI-250 w/ Gemini I

CFF 350mm f/15 classical Cassegrain (5250mm focal length)

Feather Touch 2" R&P focuser

Optec Third Lynx focus motor

QHY medium OAG (8mm prism) --not currently in use--

80mm f/7 guide scope

QHY 485C mini camera for guiding

QHY filter wheel

QHY 268M imaging camera

Last night I tried two major things: I used my previous profile with the RA algorithm set to hysteresis and the declination algorithm set to resist switch.  I also tried a new profile with both axes using the hysteresis algorithm.  A friend of mine is set up that way and he gets good results.  I did new calibrations for both profiles (I had to redo calibration for my old profile because the orientation of my guide camera was changed).  I also ran Guiding Assistant for each profile and Accepted the recommendations.

The profile using hysteresis/ resist switch worked better than the hysteresis/ hysteresis profile.  If you look in my log you'll see I started with the CA/ GA recommended minimum moves, but made those progressively shorter in an attempt to get my mount to settle.  I was somewhat successful. You'll also notice something went really crazy at 3:00am and I didn't recover.  

Just like last week, declination was better behaved than RA, which I don't understand.  In the coming weeks I'll be able to borrow a small counterweight which I can adjust to make sure my mount is slightly imbalanced on either side of the pier, which may help.

Do you think any of my settings are obviously wrong?  Suggestions for changes?

I got a recommendation from a friend who consistently gets good results that I change my guide camera exposure to something much longer, between 5-10s.  What are your thoughts?

Another weird thing was a few times during the night my SNR ratio dropped from >300 to less than 10 and I lost guiding.  I checked the sky, it was clear (at least to the naked eye) every time.  I also check my guide scope lens and it was dew-free.  Any thoughts about what could have caused this?  

[Brian Valente]

@macdonjh

>>> The profile using hysteresis/ resist switch worked better than the hysteresis/ hysteresis profile. 

that's entirely expected, these are the default algorithms for a mount such as yours. 

>>>   If you look in my log you'll see I started with the CA/ GA recommended minimum moves, but made those progressively shorter in an attempt to get my mount to settle.  I was somewhat successful. 

I agree - your longest session of 1 hour 35 mins showed fairly low overall RMS of 0.75". That's probably pretty close to what's possible with your setup. There are a few suggestions below

>>>declination was better behaved than RA, which I don't understand

That is typical: RA axis is always moving at sidereal rate so you have to deal with the effects of periodic error and other related mechanical issues, whereas Dec is only moving when commanded to do so (guide pulses) where you are dealing with reverses and backlash.

>>>>I got a recommendation from a friend who consistently gets good results that I change my guide camera exposure to something much longer, between 5-10s.  What are your thoughts?

I would not do this - your mount needs to be guided reasonably aggressively. 5-10 seconds would be fine for a well behaved mount or one with encoders. I think somewhere in the 1.5-3 second range is good for you.

>>>Do you think any of my settings are obviously wrong?  Suggestions for changes?

Here are some thoughts:

1. It would be helpful if on your next outing you did a calibration and then ran the guiding assistant for 30 minutes uninterrupted. It would be good to see the unguided performance of your mount
   
   
2. One of the issues I see is very high guidestar SNRs and huge variability. Something seems amiss here. One thing that would help is enabling the diagnostic imaging logging for your next run, available in Advanced Settings. I'd suggest enabling these options:
   
   
   
3. the MI-250 is a classic losmandy design with 239 second primary period. Consider switching your RA algorithm to PPEC with a fixed period of 239 seconds, disable auto adjust period. 
   
   
4. Getting good results with a scope like this is going to be a challenge: It's a very long focal length and a slow optic, so it's going to take a lot of integration time. It's a scope optimized for planetary imaging/observing. 
   
   
5. can you send a pic of your guide scope setup? your guiding assistant showed some bleed between ra and dec, i'm wondering if the guidescope is well affixed. Hopefully not the guidescope rings.

--
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/4377344d-69ff-4406-b11f-bfe2686bc2f1n%40googlegroups.com.

--

Brian 

Brian Valente

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

nature portfolio brianvalentephotography.com 

other astronomy https://www.brianvalentephotography.com/astrophotography/ 

[macd...@yahoo.com]

Thank you, Brian.  I've printed your reply and am hopeful of decent weather this weekend so I can try your suggestions.  

I understand your comment about using hysteresis vs. resist switch for declination.  My friend gets good results, so I tried it his way for a while.  Then switched back.  I've got a PEC curve stored in my Gemini controller and had activated it this past weekend.  I had hoped my RA response would be smoother for that reason.  The huge jumps, 3"-4", also surprised me.  Perhaps a PEMPro license is in my future.  My other friend's mount is well behaved and his entire system is well-tuned, so he gets good results.  I may try the long guiding exposure just to see what happens.  I can also try PPEC w/o auto adjust for RA.  I can certainly run Guiding Assistant for an extended period as well as enable Diagnostic Image Logging.  My SNR has been super high since I stared using a guide scope.  I am just as confused the variability I saw this weekend and was unable to figure out a cause.  The only thing I'm left with is a momentary power outage, but i didn't notice any of my other gear disconnecting, so I don't think that was it.

Yes, I know, this is the exact wrong scope for DSO imaging.  I get good enough results to keep trying, though.  A new scope is not on my horizon.  I may try it for what it's intended for one day: lunar and planetary.  :-)  At least with the super-sensitive sensors I get reasonable results pretty quickly.  It'll be fun to see what happens when I can reliably get longer total integrations, though.

I don't have a good photo of just my guide scope.  I do not use guide scope rings.  I have fixed rings in a custom cut plate which forms part of my main scope's frame.  The rear fixed ring is secured with a single bolt to create a pivot.  The front ring is secured by two bolts which "ride" in curved slots to allow my +3 degrees of azimuth adjustment.  I've got my guide scope adjusted and shimmed so whatever object I point to falls on both my imaging chip (10' x 15' FOV) and my guide scope chip.  

[Brian Valente]

>>>   I've got a PEC curve stored in my Gemini controller and had activated it this past weekend. 

how recently was your PEC done? it may be invalid. MI-250 isn't exactly a new mount ;)

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/bb34d70b-6a5a-4f63-940d-690771014f6cn%40googlegroups.com.

[macd...@yahoo.com]

Just a couple of months.  I made a curve when my trial subscription was active.  If it turns out to be important, I guess I'll purchase a license.  I have a couple of other suggestions from you to try first.

[Brian Valente]

>>> Just a couple of months.  I made a curve when my trial subscription was active.  If it turns out to be important, I guess I'll purchase a license.  I have a couple of other suggestions from you to try first.

that's good - it really depends on how well your PE curve corrected any measured error. the pempro files would tell you more here

iirc the Gemini 1 has a somewhat limited PEC function, compared with Gemini 2 L6 firmware

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/8d818cdf-694d-40cf-acc2-8dffc53f88bfn%40googlegroups.com.

[macd...@yahoo.com]

Logs: https://openphdguiding.org/logs/dl/PHD2_logs_Ssba.zip

Last night I started with a new calibration and then ran Guiding Assistant for fifty minutes.  I have some questions about the INFO: GA Result in my log.  Perhaps someone can help me with these: 

• When I accepted the GA suggestion to apply 650ms of backlash compensation I was surprised to see, after guiding had been going for a while, my backlash compensation had been changed to 20ms.  Is that all part of the "adaptive algorithm" described in the "book"?  I calibrated near the meridian but was guiding in northern Cygnus.
• I looked at my log and don't know what RA HPF-RMS means.  What does HPF stand for?
• RA Peak (pixels), is that my mount's tracking error?

The RA error plot when I was running Guiding Assistant was quite periodic, perhaps my PEC curve isn't right.

I tried PPEC for the RA axis twice.  In each case I let PHD run for twenty and then three-and-a-half minutes and RA didn't stop oscillating either time.  That was frustrating.  I eventually switched back to Hysteresis for RA, but then the sky got squirrely with wide-spread clouds coming and going.  You can see that, I'm sure in the star mass and SNR results after midnight.

I'm as confused by my SNR results as Brian V. is.  I forgot to enable Diagnostic Image Logging, but have it turned on now.

Would changing the focal length of my guide scope make any relevant difference?  I've seen posts recommending guide systems have an image scale of 0.5"/pixel.  Perhaps that's a suggested minimum.  My current image scale is a bit more than 1"/pixel.

What's currently causing me the biggest problem is the big oscillations.  SGP won't start a frame until guiding is within +1" maybe +0.75", so my system spends nearly all its time waiting for guiding to settle.  I've posted in that forum asking if there is a way to change that.

[Brian Valente]

You started with an error-free but fairly poor calibration result, so many of your results are somewhat compromised. The main issue continues to be you are losing your guide star. 

>>> I forgot to enable Diagnostic Image Logging, but have it turned on now.

Enabling this would have shed light on this issue. One thing I can see is you are guiding on heavily saturated/too bright stars. Make sure you have saturation by profile enabled. 

 

>>> When I accepted the GA suggestion to apply 650ms of backlash compensation I was surprised to see, after guiding had been going for a while, my backlash compensation had been changed to 20ms.  Is that all part of the "adaptive algorithm" described in the "book"?

yes, that's exactly how it's supposed to work. Your Dec looks reasonably good

>>>Would changing the focal length of my guide scope make any relevant difference?  I've seen posts recommending guide systems have an image scale of 0.5"/pixel.  Perhaps that's a suggested minimum.  My current image scale is a bit more than 1"/pixel.

No. 0.5" is a minimum scale for centroid calculation, not a recommendation. The only recommendation is keeping within 5:1 or lower image scale between guide scope and imaging scope respectively. 

>>>>What's currently causing me the biggest problem is the big oscillations.  

look at your unguided results: your mount has a large periodic error, measured at 58". So i'd say your mount PEC (if enabled) is invalid and making things worse, or other hardware-related issues are causing this. You need to take your raw PE tamed before you are going to make more progress in your guiding

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/03cc13be-bb9f-4537-ad98-c1f94f0ee629n%40googlegroups.com.

[macd...@yahoo.com]

Thank you for all your help, Brian.  Comments to your comments are below in red: 

You started with an error-free but fairly poor calibration result, so many of your results are somewhat compromised. The main issue continues to be you are losing your guide star. 

That's funny.  I did a little happy-dance when I got a "good" result the first time last night (previously I generally got "poor" once or twice before getting "acceptable").  What are the things to look for (in the log?) which would tell me the difference between a "good good" calibration and a "poor good" calibration?

Enabling this [diagnostic image logging] would have shed light on this issue. One thing I can see is you are guiding on heavily saturated/too bright stars. Make sure you have saturation by profile enabled. 

 Enabled now, hope for clear skies tonight to get some images for this.  You've commented about huge SNR numbers before.  Last night was very strange: there were times I had huge SNR, >600, and other times when SNR struggled to be higher than 10.  I checked for clouds, for dew, and didn't see either.  I have no idea why the signal changed when there was no change in hardware or profile and no environmental changes I could see.  

yes, that's exactly how it's supposed to work. Your Dec looks reasonably good

OK, at least that I understood.  Initially I thought PHD was ignoring what I'd asked for and got irritated.

No. 0.5" is a minimum scale for centroid calculation, not a recommendation. The only recommendation is keeping within 5:1 or lower image scale between guide scope and imaging scope respectively. 

Got it.  What characteristics of guiding are affected by the ratio of image scales?

look at your unguided results: your mount has a large periodic error, measured at 58". So i'd say your mount PEC (if enabled) is invalid and making things worse, or other hardware-related issues are causing this. You need to take your raw PE tamed before you are going to make more progress in your guiding

That graph's from the Log Viewer?  Where is it, I couldn't find it.  The periodic error peak I see in the RA-Dec graph (when GA was running) is generally 10"-15" (~15"-17" peak-peak).  I have time before dark to get a PEMPro license and can redo my PEC curve tonight.  I'll also guide for a while without PEC enabled using Hysteresis/ Resist Switch so there's some "clean" data.

For interpreting my GA results, what does RA HPF-RMS mean?  Specifically the HPF part.  Also, in my log, when GA was running, my RA Peak-Peak was 17.3", if that's not my mount's PE, what measurement is that?

I'll also have photos of my guide scope installation tomorrow, when I can get them off my phone (didn't bring that cable with me).  

[macd...@yahoo.com]

Logs: https://openphdguiding.org/logs/dl/PHD2_logs_i37K.zip

I don't know how to upload Diagnostic images, are they attached to the logs (it doesn't look like it).  I haven't searched my hard drive yet.  If I find them, I'll post some or all.

I took some more data this weekend and have done some preprocessing.  Brian V.'s comment about guiding fighting with mount issues appears to be spot-on.  I took an hour's worth of data guided and another hour's worth unguided.  "wg" in the filename means "with guiding".

Most everything else I was working on fell into place over the weekend.  I pulled my imaging gear off and put an eye piece in and checked collimation (yes, I know I could have done that with a camera; tried it and failed so I went back to what I know).  Collimation was off a bit, so I improved that.  After that autofocus started working properly, and I was already getting good plate solves so I could finally focus on one thing.  

That's the reason behind both unguided and guided imaging.  During the run I did with guiding I let everything work for four to five minutes (one worm period plus a bit) then I'd change something.  Mostly I changed the length of my guiding photo: initially 3s but I worked down to 1.5s.  I think guiding was more stable at 1.5s, but as Brian said, there are things fighting.  Next time out I'll have more knowledge of PEMPro and I'll get a good PEC curve and try that.

If you look at my logs, I think most of the data are suspect.  There was a light breeze, variable but present almost all the time which pushed my mount around.  Luckily the wind stopped after midnight when I decided to try taking some photos.  

[macd...@yahoo.com]

...and here's that guided image.  I couldn't post them together, files are too big.  I'll resize them next time.

[Brian Valente]

It would be more informative to post single subs rather than integrated masters. 

Another thing you can test for is differential flexure: integrate your subs but do not register them first. That will show you clearly how much DF you have.

Otherwise, I think you've done about all you can do in terms of improving guiding and anything I can offer you here. I suggest you take a break here and focus on your mount performance and working on that. There's not a lot more we can say here that hasn't already been said. 

The logs look pretty much the same as before, slightly improved. If you can tame the primary pe to be lower, that will probably go a long ways towards improving things. 

--
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/d79a45d4-a91c-4830-9741-099d0d8a6222n%40googlegroups.com.

[Brian Valente]

PS

>>>I don't know how to upload Diagnostic images,

Choose Open Log folder from Help menu

you will see folder(s) like this, inside are the diagnostic images

--
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/d79a45d4-a91c-4830-9741-099d0d8a6222n%40googlegroups.com.

[macd...@yahoo.com]

I agree, and I thank you for all your help.  I only posted those frames as follow-up to all the suggestions you've made.  I think it's poor manners to ask for help and not respond to those who've helped you with your results, whether good or bad.  

Time to read about proper PEMPro operation so my next attempt at PEC will be better.

[Brian Valente]

>>> Time to read about proper PEMPro operation so my next attempt at PEC will be better.

head over to the Losmandy user groups.io, there's a pdf on how to optimize PEMPro use for losmandy mounts

(I forget if it's in the losmandy user group or the gemini 2, but either way it applies for you)

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/affae58c-5c14-45fa-9312-c8044ee62007n%40googlegroups.com.

[macd...@yahoo.com]

Thank you, again, Brian, I'll look.  I'm a member of both groups.io forums.

A follow-up: I troubleshooting late last night after working with PEMPro and noticed my guide camera is setup with a gain of 100.  If I read the QHY literature correctly, the setting for unity gain is 16.  If I reset my gain to 16, (reduced by a factor of 7), is it likely my SNR in PHD would drop by a similar factor?  If so, during last night's work, by SNR would have dropped from ~350 to ~50.  If that's the way it works, perhaps I can use that when it's time to try my OAG again: set my guide camera's gain to say 48 or 64 to increase my SNR from ~6 to ~18-24.

Finally, can I change the guide camera gain in my profile without messing things up?  Or should I create a new profile with a new gain setting?

[Brian Valente]

SNR is going to depend on the selected guide star, primarily. I wouldn't sweat the gain on your guide camera. If you are getting guidestars with SNR >20 it is fine. 

If you are getting lost guidestars frequently, you can increase your gain. You do not need to make a new profile when changing the guide camera gain, although you probably should (i.e., your darks are not going to match your guide camera). I'd figure out what gain you want, then create the profile and dark library

To view this discussion on the web visit https://groups.google.com/d/msgid/open-phd-guiding/f17a4d4d-7cd1-4684-b0ed-46097f6bd535n%40googlegroups.com.

[macd...@yahoo.com]

Thank you again, Brian.  Lost guide stars are one reason I am working with a guide scope instead of an OAG at the moment.  I'll keep this post in mind.