Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

PHD2 reports 14s of Dec backlash but guiding seems ok?

  • Please log in to reply
17 replies to this topic

#1 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 19 July 2019 - 04:45 PM

I've been easing my way into guiding and generally it's been going OK. Yesterday I switched my Orion MMAG guiding setup between scopes and re-ran calibration which reported "too little Dec movement". I then ran the guiding assistant to learn more with the following results:

 

PHD2 - 2.JPG

 

14s of backlash seems a little unlikely and the PA error isn't terrible at 2.3 right? I pressed on and starting guiding anyway and the results look fairly normal to my untrained eye. I was running 60s exposures in SharpCap and didn't notice any drift. 

 

PHD2 - 4.JPG

 

I would appreciate any thoughts from more experienced guiders on where I've gone wrong here.

 

Laptop was connected to my EQ6-R via EQMOD to the handset port. Happy to post any settings or logs if needed.

 

Thanks



#2 RedLionNJ

RedLionNJ

    Soyuz

  • *****
  • Posts: 3565
  • Joined: 29 Dec 2009
  • Loc: Red Lion, NJ, USA

Posted 19 July 2019 - 05:01 PM

Unless I'm mis-reading the last picture (the guiding graphs), I don't see any corrections being issued in Dec.

 

The results in the image subs are what really matters, here. I would expect significantly large stars, given the tolerance in both RA & Dec. But if they're round and you're happy with them, I'd let sleeping dogs lay.


  • cshine likes this

#3 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 19 July 2019 - 05:31 PM

Unless I'm mis-reading the last picture (the guiding graphs), I don't see any corrections being issued in Dec.

 

The results in the image subs are what really matters, here. I would expect significantly large stars, given the tolerance in both RA & Dec. But if they're round and you're happy with them, I'd let sleeping dogs lay.

Thanks for the reply and good point on the Dec corrections..  I did take another screengrab 10 mins before the one above which does show some Dec corrections but not many.

 

PHD2 - 3.JPG

 

On the star roundness.. I just checked the individual subs more closely and my bad I actually do have a drift issue, didn't notice this when monitoring the guiding graph and the stacked image in SharpCap.

 

BadStars.JPG

 

So does the question now become why is PHD not giving more Dec corrections?



#4 OldManSky

OldManSky

    Apollo

  • *****
  • Posts: 1497
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 19 July 2019 - 06:22 PM

Even that second one only has one visible DEC correction.  Notice that one happened when the DEC error got above 3" -- your minmove setting for dec, at .4, means that PHD will do nothing until you're over 3" of DEC error.  That's probably not nearly good enough.  I'd suggest setting your DEC minmove to about 0.1 or so, and turning the DEC aggression down a little as well (to 80% to start with).  See how that goes.


  • cshine likes this

#5 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7353
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 19 July 2019 - 06:43 PM

Do you have the correct pixel size for the guide camera and the correct focal length for the guidescope in PhD2? 

 

It's very unlikely for PhD2 to suggest 0.32 and 0.4 minmos with the MMAG (Magnificent Mini). I never got one greater than 0.15. The higher MinMos are more like what you get with an OAG. Check your guide log and make sure you have the above data correct. 

 

Incidentally, "star did not move enough" is one more sign of that (wrong F/L input). 


  • cshine likes this

#6 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 19 July 2019 - 07:05 PM

Do you have the correct pixel size for the guide camera and the correct focal length for the guidescope in PhD2? 

 

Thanks Stelios, yes those settings look to be correct. However when checking this I noticed the suggestion to compute a value for "calibration step (ms)" using the Advanced button. When I did that PHD suggests a very different value (4600ms) to the current setting (1850ms):

 

phd2 calibration step.JPG

 

Is that suggestion based purely on pixel size and FL?



#7 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 19 July 2019 - 07:07 PM

Even that second one only has one visible DEC correction.  Notice that one happened when the DEC error got above 3" -- your minmove setting for dec, at .4, means that PHD will do nothing until you're over 3" of DEC error.  That's probably not nearly good enough.  I'd suggest setting your DEC minmove to about 0.1 or so, and turning the DEC aggression down a little as well (to 80% to start with).  See how that goes.

Will do, thank you.


  • OldManSky likes this

#8 Stelios

Stelios

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 7353
  • Joined: 04 Oct 2003
  • Loc: West Hills, CA

Posted 20 July 2019 - 02:07 AM

Is 0.2x sidereal the proper guide speed for your mount? Seems very low.



#9 AhBok

AhBok

    Vanguard

  • *****
  • Posts: 2216
  • Joined: 02 Dec 2010
  • Loc: Lakeland, TN

Posted 20 July 2019 - 08:05 AM

I have an EQ6R Pro and mine will not even calibrate properly at .2 Sidereal. I’ve worked with other users who say the same. I’d bump it up to .5-.7 Sidereal.

#10 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 20 July 2019 - 08:37 AM

Is 0.2x sidereal the proper guide speed for your mount? Seems very low.

 

I have an EQ6R Pro and mine will not even calibrate properly at .2 Sidereal. I’ve worked with other users who say the same. I’d bump it up to .5-.7 Sidereal.

 

Thanks Stelios and Randy.. I actually have it set to 0.5 in EQMOD and correcting this parameter on the Advanced screen gets me back to a Calibration Step recommendation of 1850ms. I'm not sure how or why it was changed from 0.5 to 0.2 in there.

 

So in a summary attempt, the pixel size and fl were correct and the measured total RMS was respectable at 0.66. The guide rate was wrong in the Advanced calibration screen but it was correctly set to 0.5 in ASCOM EQMOD and that's what really matters. However the high backlash measurement probably resulted in a very high minmove recommendation which then caused significant under corrections in DEC.. the log file confirms this, vast majority of required DEC guide pulses were below minmove and ignored.

 

I've set minmove back to 0.1 as suggest above. Aggression was at 70, I've tweaked that to 80.

 

Still not clear on what caused PHD2 to record such a high backlash value.


Edited by cshine, 20 July 2019 - 09:16 AM.


#11 elmiko

elmiko

    Mercury-Atlas

  • -----
  • Posts: 2981
  • Joined: 27 Jul 2009
  • Loc: Arizona

Posted 20 July 2019 - 09:17 AM

Try changing your RA and DEC settings in Eqmod to something in the 20 or 30 range  the sliders in the aggression in the pulse guiding box. Found in the Eqmod toolbox. This could be the problem.

Mike



#12 StephenW

StephenW

    Viking 1

  • -----
  • Posts: 827
  • Joined: 04 Feb 2005
  • Loc: San Jose, CA

Posted 20 July 2019 - 10:23 AM

Can you upload the logs files that include the calibration and GA run that lead to those large minmove recommendation?  It would be interesting to see what PHD2 was "thinking" when making those recommendations.

 

And it would also be good to upload those same logs to the Open PHD2 forum so the devs can have a look at it and figure out what is going on.



#13 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 20 July 2019 - 05:53 PM

Can you upload the logs files that include the calibration and GA run that lead to those large minmove recommendation?  It would be interesting to see what PHD2 was "thinking" when making those recommendations.

 

And it would also be good to upload those same logs to the Open PHD2 forum so the devs can have a look at it and figure out what is going on.

Appreciate the response Stephen.. I've attached the 2 interesting log sections. I'll post these over on the PHD2 forum too.

 

Here is the GA result for reference:

 

23:10:23.405 00.000 12392 BLT: South pulses ended at Dec location -754.83
23:10:23.405 00.000 12392 BLT: Drift correction of -0.12 px applied to total north moves of 58.74 px, -0.005 px/frame
23:10:23.405 00.000 12392 BLT: Empirical north rate = 1.22 px/s
23:10:23.405 00.000 12392 BLT: Trial backlash amount is 17.35 px, 14179 ms, sigma = 0.1 px
23:10:23.405 00.000 12392 BLT: Clearing pulse is very large, issuing max S move of 12256

 

Just to stress - I'm pretty new to guiding so there's a good chance I've messed up on some basic setup step somewhere..

Thanks

Attached Files



#14 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 20 July 2019 - 06:08 PM

Try changing your RA and DEC settings in Eqmod to something in the 20 or 30 range  the sliders in the aggression in the pulse guiding box. Found in the Eqmod toolbox. This could be the problem.

Mike

Hi Mike, I'll check that out, thanks. I don't remember seeing aggression settings in the EQMOD pulse guiding area, just guide rates. I only have access to the simulator settings right now (while away from mount) but this is what I see there:

 

EQMOD.JPG

 

Here are the current PHD aggressiveness settings:

 

PHD2 settings2.JPG

 

 

Cheers


  • elmiko likes this

#15 elmiko

elmiko

    Mercury-Atlas

  • -----
  • Posts: 2981
  • Joined: 27 Jul 2009
  • Loc: Arizona

Posted 20 July 2019 - 08:11 PM

Try increasing the DEC guide rate in Eqmod to 30. Yours is set at .10. Try .30 

I meant the guide rates, not agression.

Mike


Edited by elmiko, 20 July 2019 - 08:14 PM.


#16 elmiko

elmiko

    Mercury-Atlas

  • -----
  • Posts: 2981
  • Joined: 27 Jul 2009
  • Loc: Arizona

Posted 20 July 2019 - 08:16 PM

https://youtu.be/WX2uXMu0crw

Check out this video by Chris Shilito.



#17 cshine

cshine

    Vostok 1

  • *****
  • topic starter
  • Posts: 132
  • Joined: 26 May 2018
  • Loc: Houston, Texas

Posted 20 July 2019 - 08:30 PM

Try increasing the DEC guide rate in Eqmod to 30. Yours is set at .10. Try .30
I meant the guide rates, not agression.
Mike

Hi Mike, the ascom simulator mount guide rates shown above are indeed set to .1 but I have set mine to .5 per other recommendations

thanks for that video link I'll check it out
  • elmiko likes this

#18 StephenW

StephenW

    Viking 1

  • -----
  • Posts: 827
  • Joined: 04 Feb 2005
  • Loc: San Jose, CA

Posted 21 July 2019 - 11:29 AM

Appreciate the response Stephen.. I've attached the 2 interesting log sections. I'll post these over on the PHD2 forum too.

 

Here is the GA result for reference:

 

23:10:23.405 00.000 12392 BLT: South pulses ended at Dec location -754.83
23:10:23.405 00.000 12392 BLT: Drift correction of -0.12 px applied to total north moves of 58.74 px, -0.005 px/frame
23:10:23.405 00.000 12392 BLT: Empirical north rate = 1.22 px/s
23:10:23.405 00.000 12392 BLT: Trial backlash amount is 17.35 px, 14179 ms, sigma = 0.1 px
23:10:23.405 00.000 12392 BLT: Clearing pulse is very large, issuing max S move of 12256

 

Just to stress - I'm pretty new to guiding so there's a good chance I've messed up on some basic setup step somewhere..

Thanks

 

 

Looks like you uploaded snippets from your debug logs, not the guide logs.  The guide logs are more useful as we can visualize what's going on using the PHD2 log viewer.

 

Anyway, the debug logs do contain the final GA recommendations:

 

23:11:21.450 00.000 12392 Guiding Assistant results follow:
23:11:21.450 00.000 12392 SNR=47.0, Samples=356, Elapsed Time=734s, RA HPF-RMS=  0.08 px (  0.55 arc-sec ), Dec HPF-RMS=  0.05 px (  0.36 arc-sec ), Total HPF-RMS=  0.10 px (  0.66 arc-sec )
23:11:21.450 00.000 12392 RA Peak=  0.98 px (  6.51 arc-sec ), RA Peak-Peak   6.59 px ( 43.61 arc-sec ), RA Drift Rate= -0.41 px/min ( -2.70 arc-sec/min ), Max RA Drift Rate=  0.12 px/sec (  0.76 arc-sec/sec ), Drift-Limiting Exp=   2.8 s
23:11:21.450 00.000 12392 Dec Drift Rate= -0.07 px/min ( -0.47 arc-sec/min ), Dec Peak=  0.66 px (  4.35 arc-sec ), PA Error= 2.3 arc-min
23:11:21.450 00.000 12392 Uncorrected Dec RMS=0.307px, Linear-fit Dec drift=-0.071 px/min, Drift-corrected Dec(raw) RMS=0.185px, R-sq=0.677
23:11:21.457 00.007 12392 Recommendation: Try to keep your exposure times in the range of 1.0s to 3.0s
23:11:21.464 00.007 12392 Recommendation: Try setting RA min-move to 0.32
23:11:21.474 00.010 12392 Recommendation: Try setting Dec min-move to 0.40
23:11:21.488 00.014 12392 Recommendation: Backlash is >= 14179 ms; you may need to guide in only one Dec direction (currently North)

 

It's definitely strange that the recommended min-moves were so large.   For RA it could be partially due to the relatively large/fast Max RA drift rate (caused by some steep/fast part of your PE curve).  Doesn't seem to be anything in DEC that would justify such a large min-move.

 

Will be interesting to hear what the PHD devs have to say.


  • cshine likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics