PHD2 guiding calibration error

[SStanford]

Hi All,

Lovely clear night tonight; first opportunity to use my ZWO ASI120mm which I recieved this week.

Connected the EQ3pro Synscan to my comuter, controlled by Stellarium, slewed to a star in the South (which then starts tracking), Set-up PHD2, got a good focus on the stars, calibration running smoothly until it "steps north".

PHD2 "steps east" (around x50), backlash (around x20) then "steps north" (around x60) before I get "Calibration failed" in the bottom left and the message: "Dec calibration failed: Star did not move enough".

At the backlash stage on PHD2 the mount starts clicking every so often, as it does for the whole "steps north" process too 😬.

I tried slewing to different parts of the sky but encountered the same failure three times. 

I'm completely stumped with this one so any guidance would be hugely appreciated! I've attached a screencap of the interface below in case theres anything I've overlooked here.

[knobby]

Have you entered guide camera pixel size and guidescope focal length correctly ? Maybe try increasing calibration step size and / or nudge north before calibration.

How is the mount connected ? St 4 cable or eqmod ?

[knobby]

Sorry just noticed eqmod on screenshot 😁

[KP82]

What about balance on the DEC axis? Did you pick a star just above the equator and near the meridian for the calibration?

[kens]

Start with this: https://github.com/OpenPHDGuiding/phd2/wiki/EQASCOM-Settings

Then have a read of: https://openphdguiding.org/phd2-best-practices/

If you still have problems: https://openphdguiding.org/getting-help/

or if you prefer you can upload your logs to the forum. 

[SStanford]

1 hour ago, knobby said:

Have you entered guide camera pixel size and guidescope focal length correctly ? Maybe try increasing calibration step size and / or nudge north before calibration.

Thanks @knobby.

I havent tried any of these things yet! I'm a complete newbie with PHD2, do you know if theres a guide that'll gove some pointers in this direction?

1 hour ago, KP82 said:

What about balance on the DEC axis? Did you pick a star just above the equator and near the meridian for the calibration?

@KP82; I did indeed pick a star near the equator and meridian (just east of the meridian, North of the equator). Balance in bothe Dec and RA are good.

1 hour ago, kens said:

Start with this: https://github.com/OpenPHDGuiding/phd2/wiki/EQASCOM-Settings

Then have a read of: https://openphdguiding.org/phd2-best-practices/

If you still have problems: https://openphdguiding.org/getting-help/

or if you prefer you can upload your logs to the forum. 

@kens, thanks for these. I'll implement them tomorrow.

Log files are detailed below if anyone can take a look?

https://openphdguiding.org/logs/dl/PHD2_logs_b2VW.zip

Edited January 23, 2021 by SStanford
Wrong Log file!

[michael8554]

Might be in one of those links above, but default EQMOD Guide Rate is 0.1X , set it to 0.5X or higher.

And before Cal, slew mount north at lowest speed until guidestar moves, to take up Dec backlash.

Michael

[kens]

1 hour ago, michael8554 said:

Might be in one of those links above, but default EQMOD Guide Rate is 0.1X , set it to 0.5X or higher.

And before Cal, slew mount north at lowest speed until guidestar moves, to take up Dec backlash.

Michael

It is indeed in the first link along with screen shots showing how to do it.

And from inspection of the guide log that is the first issue to resolve..

[SStanford]

Thanks both @michael8554, @kens.

I'll make these changes and post my results. 

[SStanford]

I've set RA & Dec rate to x0.80. Does it really matter if I were to put this up to x0.90? would guiding be more accurate at x0.50 for example?

"sideofpier" is set correctly to "Pointing ASCOM" as is Guiding "ASCOM pulse guiding".

I have also ensure that focal length and pixel size is accurate for the guide scope and cam in PHD2 (3.75 and 120mm (here)).

At next clear night I'll be focusing on a portion of the sky +/- 20deg from 0dec and near the meridian. I'll also be slewing north via EQMOD at slew rate 1 for 20 seconds before running calibration.

Should this not resolve the issue I'll be trying all the details outlined here and here. I'll post my results after.

[Skipper Billy]

7 minutes ago, SStanford said:

I've set RA & Dec rate to x0.80. Does it really matter if I were to put this up to x0.90? would guiding be more accurate at x0.50 for example?

AFAIK its a trade off between pulse guide settings in ASCOM and step size in PHD2. If you increase the pulse settings in ASCOM you will need to reduce the step size in PHD2. 0.8 in ASCOM is a very good starting point.

Once you have it up and running run the guiding assistant and implement its recommendations - it is usually pretty much spot on.

[SStanford]

59 minutes ago, Skipper Billy said:

Once you have it up and running run the guiding assistant and implement its recommendations - it is usually pretty much spot on.

Thanks @Skipper Billy.

[kens]

23 hours ago, kens said:

Start with this: https://github.com/OpenPHDGuiding/phd2/wiki/EQASCOM-Settings

Then have a read of: https://openphdguiding.org/phd2-best-practices/

If you still have problems: https://openphdguiding.org/getting-help/

or if you prefer you can upload your logs to the forum. 

 

[SStanford]

Calibration successful! thanks for your help all. Brief opportunity before the clouds rolled over tonight.

With one issue resolved, two more pop up. All part of the fun.

1. I ran the guiding assistant tool and had two different outcomes. The second outcome was worse having tweaked PA via the Drift alignment tool (reducing recommeded exposure times from 2-4 seconds to 1-3 seconds 😐, unacceptable!) It also recommended amending the RA and Dec min move, before I apply these changes has anyone had any experience with these amendements?

2. The drift alignment tool is very tricky to manage. I had c.40 mins to play with Azimuth adjustment; when I got anywhere near a stable Dec line in the graph, the slightest adjustment in the Azimuth would cause the graph to give a wild reading.  In fact, the lowest PA error I could get was 4.64' (38px), see image below. Any tips with this?

[kens]

You need to attach a guide log for any analysis. The screen shots have too little information to give any sensible guidance

[KP82]

I agree a guide log would be useful.

As for the min-move recommendations by the guide assitant, I'd apply them. They compensate the backlash in both axes.

Drift align is quite tricky. First you need to let it drift while the atmoshpere is steady. If there is a strong turbulance, it could throw off the DEC line giving erroneous readings. You then need to alternate between azimuth and altitude drift/adjustment until you get a good result. The best I've ever done on a steady night was around 0.8 - 0.9 arcmin PA error. During windy nights it would go up to 1.8 - 2 arcmins.

Edited January 26, 2021 by KP82

[michael8554]

PHD2 Guide Assistant measures PA and only complains if the error is greater than 5 arcmins.

That's because the developers say that is good enough to guide with.

So the 4.64arcmins in the screen shot is fine.

And yes, changing the adjustment when you are that close is very tricky, another reason to stop while you're ahead.

You only need to spend time getting a very low PA error if you're imaging without guiding.

As I understand it, the recommended exposure times are based on the High-frequency Star Movement measurements., to prevent "Chasing the Seeing" conditions.

That figure in the second screengrab is nearly twice the first.

And until you gain experience, why would you doubt the recommended MinMo settings ?

Michael

[SStanford]

20 hours ago, kens said:

You need to attach a guide log for any analysis. The screen shots have too little information to give any sensible guidance

Thanks Kens, guide logs below for reference.

18 hours ago, KP82 said:

As for the min-move recommendations by the guide assitant, I'd apply them. They compensate the backlash in both axes.

Drift align is quite tricky. First you need to let it drift while the atmoshpere is steady. If there is a strong turbulance, it could throw off the DEC line giving erroneous readings. You then need to alternate between azimuth and altitude drift/adjustment until you get a good result. The best I've ever done on a steady night was around 0.8 - 0.9 arcmin PA error. During windy nights it would go up to 1.8 - 2 arcmins.

I'll apply the recommended minmove now.

There was a lot of whispy cloud around last night and I feel this probably interfered with the readings. I'm also going to up the exposure times to 2 seconds rather than 1 to see if that helps.

10 hours ago, michael8554 said:

PHD2 Guide Assistant measures PA and only complains if the error is greater than 5 arcmins.

So the 4.64arcmins in the screen shot is fine.

I'll aim for <5mins going forward.

10 hours ago, michael8554 said:

As I understand it, the recommended exposure times are based on the High-frequency Star Movement measurements., to prevent "Chasing the Seeing" conditions.

I'll chalk this up to the aformentioned clouds. Will the increased exposure time and a steadier atmosphere help the High-frequency Star Movement measurements?

PHD2_GuideLog_2021-01-25_194449.txt

Edited January 26, 2021 by SStanford

[kens]

22 hours ago, SStanford said:

Calibration successful! thanks for your help all. Brief opportunity before the clouds rolled over tonight.

With one issue resolved, two more pop up. All part of the fun.

1. I ran the guiding assistant tool and had two different outcomes. The second outcome was worse having tweaked PA via the Drift alignment tool (reducing recommeded exposure times from 2-4 seconds to 1-3 seconds 😐, unacceptable!) It also recommended amending the RA and Dec min move, before I apply these changes has anyone had any experience with these amendements?

2. The drift alignment tool is very tricky to manage. I had c.40 mins to play with Azimuth adjustment; when I got anywhere near a stable Dec line in the graph, the slightest adjustment in the Azimuth would cause the graph to give a wild reading.  In fact, the lowest PA error I could get was 4.64' (38px), see image below. Any tips with this?

Your seconds GA run is a mess and totally unreliable. It looks like your mount got bumped or something slipped as it moved around 80 arcseconds on two or three occasions. Maybe the Alt/Az bolts were loose?

Due to your bad initial PA you should recalibrate once your PA has completed. A bad PA affect the accuracy of calibration. 

Do the recalibration before you run the Guiding Assistant. To recalibrate, hold down the shift key while clicking the Guide button