After autofocus auto guider takes an age to resume

[kirkster501]

Hi all,

Guiding graph is now quite good and got good stars (at last).  I get a steady drift and will investigate that. 

So I have it set up so that guiding is paused during autofocus.  However, when guiding resumes again after autofocus "resuming the auto guider" I watch PHD2 and it goes a wandering around the guide star, trying to find it again.  It can take ages.  It never used to do this.  I cleared the calibration and reforced it and now I keep guiding ON whilst autofocus runs. Seems OK.  

Any thoughts please guys?  I'm having a torrid time with guiding lately.....

Steve

[RayD]

5 minutes ago, kirkster501 said:

Hi all,

Guiding graph is now quite good and got good stars (at last).  I get a steady drift and will investigate that. 

 

So I have it set up so that guiding is paused during autofocus.  However, when guiding resumes again after autofocus "resuming the auto guider" I watch PHD2 and it goes a wandering around the guide star, trying to find it again.  It can take ages.  It never used to do this.  I cleared the calibration and reforced it and now I keep guiding ON whilst autofocus runs. Seems OK.  

Any thoughts please guys?  I'm having a torrid time with guiding lately.....

Steve

I used to have the same Steve if pausing guiding during focus.  

I don't pause guiding during focus and everything is fine.  I narrowed mine down to only Ha filter, for which focus routine is quite long, and I think PHD2 struggles to recenter the star as it is too close to force a new star, but too far to get a good centre.

[kirkster501]

Oh OK Ray that is good to know. Thanks.

This is the state of my guide graph when left running during autofocus.

[kirkster501]

..what also concerns me a bit is the nature of the graph.  It's reasonably flat but the blue and the red seem to be separated on this rig - FSQ85/MOravian G2-8300 and OAG with Lodestar - and rarely criss-cross. The red is generally north of the X axs and the blue south of it.  I don't get why.....?

On my TEC/Atik/OAG/QHY5 rig the blue and red were closer to the X axis (implying better guiding) and criss-crossed each other all the time.  I thought with its longer focal length the TEC would have been harder to guide.  But that is not what I am finding.

[RayD]

1 minute ago, kirkster501 said:

 

Oh OK Ray that is good to know. Thanks.

This is the state of my guide graph when left running during autofocus.

 

Yes if you use an OAG it will do this, Steve as your guide star is changing also (I believe you do use OAG)?

I guess ideally you should pause it then, but this can have the issue noted where it struggles to pick up a star.  Have you noticed whether it happens on all filters, or only Ha?

 

[kirkster501]

This is on Blue Ray.

This is my "behaving" graph.....  The bottom line is that the subs are OK.  But I am an engineer.  And engineers do not like not understanding!!

[RayD]

3 minutes ago, kirkster501 said:

..what also concerns me a bit is the nature of the graph.  It's reasonably flat but the blue and the red seem to be separated on this rig - FSQ85/MOravian G2-8300 and OAG with Lodestar - and rarely criss-cross. The red is generally north of the X axs and the blue south of it.  I don't get why.....?

On my TEC/Atik/OAG/QHY5 rig the blue and red were closer to the X axis (implying better guiding) and criss-crossed each other all the time.  I thought with its longer focal length the TEC would have been harder to guide.  But that is not what I am finding.

I know it sounds basic stuff, but I assume you do have two separate PHD2 profiles and force a new calibration when switching between them?

[RayD]

Do you auto-select a star, Steve, and do you have a bad pixel map in place? 

[RayD]

3 minutes ago, kirkster501 said:

But I am an engineer.  And engineers do not like not understanding!!

Me too, and I haven't stopped fiddling since starting in the hobby!

[kirkster501]

Yes I do Ray.  I always force a calibration at the start of the night anyway.  I just think it is good practice.  The subs seem OK.  This is M37, 2 minutes on Blue, raw sub.

 

[kirkster501]

2 minutes ago, RayD said:

Do you auto-select a star, Steve, and do you have a bad pixel map in place? 

Yes to both Ray.

Going to check the mount balance tomorrow.  we are in for some crappy weather next few days so will check it out.  But i think it is all ok in that department.

I am also within 8 arc min of the Pole so PA is not am issue.

[RayD]

Well the results look absolutely fine. 

SGP recommend pausing guiding if using OAG, but when I did that I had the same issue as you.  

[RayD]

The drift you are seeing does point towards PA, so may be worth checking just to rule that out.

[kirkster501]

Yes there is a steady "saw-tooth" that is cyclical every three minutes or so until the guiding pulls brings it back.  I checkled the PA three weeks ago and it was bang on.  I will check it again on a less than ideal night.  Want to make the most of tonight since it is a good night.

EDIT: Note the drop in guiding at the extreme right when autofocus starts.....

[RayD]

5 minutes ago, kirkster501 said:

Yes there is a steady "saw-tooth" that is cyclical every three minutes or so until the guiding pulls brings it back.  I checkled the PA three weeks ago and it was bang on.  I will check it again on a less than ideal night.  Want to make the most of tonight since it is a good night.

Don't blame you Steve.

[michael8554]

Hi Steve

I notice that your graphs have very few corrections showing, yet RA and Dec are making noticeable excursions  away from the axis.

This would suggest to me that your Min Moves are set too high ?

Michael 

[RayD]

5 hours ago, michael8554 said:

Hi Steve

I notice that your graphs have very few corrections showing, yet RA and Dec are making noticeable excursions  away from the axis.

This would suggest to me that your Min Moves are set too high ?

Michael 

Yes I agree with this.  I was able to check mine last night as I couldn't remember, and mine are almost half this (0.1 on both).  Also my aggression is much lower.

[kirkster501]

Thank you Michael and Ray.  I will check into the min moves and some of the other parameter.  I have never had to fiddle with those before but will research into it.  It makes sense what you say.