Help, why does my guiding continue to suck

[lrt75914]

Hey guys,

I had hoped to circumnavigate my ever-present guiding issues by buying an expensive mount. Corona made short work of
that pipe dream and now I'm stuck with ye olde AZ-EQ 6.

I decided to strip down that old hunk of junk again to make sure that everything was lubed up and ready to go once the
clouds buggered off. Tonight was the night and as always I didn't take any pictures. Instead I was futzing about trying
to find the source of all my woes.

Polar alignment was done using polarmaster and the guiding parameters were set according to the recommandations
of the guiding assistant. I used AutoPec to rerecord a PEC curve and recorded both the guided and unguided performance
pre and post PEC.

Here are the results:

Unguided without PEC:

Guided without PEC :

RA:   1.12"  (0.54 px)
Dec:  0.60" (0.19 px)

(Un)guided with PEC:

RA:   1.06"  (0.54 px)
Dec:  0.35" (0.19 px)

 

Guiding Assist Output pre PEC:

Guiding Assist output post PEC:

 

 

Can anyone help me? What am I doing wrong?

[Davey-T]

Have you actually tried imaging with it, have you followed recommendations ? particularly exposure times, one second is too short.

Guiding doesn't look that bad once it settles.

Dave

[Ibbo!]

As Dave says try increasing expousre time and actually see what an image looks like.

What is the f/l of the guider.

[bottletopburly]

What do you have pulse guide settings at ,default in Eqmod are 0.10 up them to 0.60 for both Dec and Ra 

[lrt75914]

Thanks for the quick responses guys!

7 hours ago, Davey-T said:

Have you actually tried imaging with it, have you followed recommendations ? particularly exposure times, one second is too short.

Guiding doesn't look that bad once it settles.

Dave

6 hours ago, Ibbo! said:

As Dave says try increasing expousre time and actually see what an image looks like.

What is the f/l of the guider.

I did apply all the recommendations and took a couple of images of Regulus to test my guiding performance. With an
integration time of 2 minutes most images showed elongated stars. I did try to increase the exposure time in a pre-
vious run but I did not see any major improvement that couldn't have been down to a difference in seeing.
Guiding parameters were set to default prior to the guiding runs that I posted to make sure that the issues did not
stem from my phd2 setup.

I followed the following steps to record the PEC curve:

1.  Polar Align
2. Point South near celestial equator
3. Calibrate guide camera in PHD2
4. Run guiding assistant for >5mins and apply reccomendations
5. Run auto-guiding

6. Initiate AutoPEC for 9 cycles
7. Switch to sidereal + PEC once finished and turn off auto-guiding
8. Record PPEC
9. Once finished turn off PEC
 

The gear I used is:

Telescope: Skywatcher Esprit 100
Camera: ZWO ASI 1600 MM Cool

Guide Scope: Borg Guide Scope (250 mm)
Guide Camera: ZWO ASI 178MM

(Batman is not part of the setup)

 

4 hours ago, bottletopburly said:

What do you have pulse guide settings at ,default in Eqmod are 0.10 up them to 0.60 for both Dec and Ra 

RA and Dec Rates are set to 0.5

 

[Davey-T]

I see the problem the scopes upside down 😄 sorry couldn't resist.

Which way are the stars elongated ? in RA presumably.

Did you leave guiding to settle before starting imaging ?

Dave

[lrt75914]

33 minutes ago, Davey-T said:

I see the problem the scopes upside down 😄 sorry couldn't resist.

Which way are the stars elongated ? in RA presumably.

Did you leave guiding to settle before starting imaging ?

Dave

You got me worried there for a second. 😄

Elongation is in RA and I do leave time for the guiding to settle before taking the images.
 

[Davey-T]

How's the balance ? if you run it unguided and rest a finger on the weight bar you can push gently one way and the other to see if you can hold the RA more on track if so try altering the balance to mimic it.

Dave

[Clarkey]

As Dave has said the guiding does not look that bad once settled and I am not sure it would be bad enough to cause obvious egg shaped stars. Looking at the trend there is clearly some oscillation in the guiding, which does suggest a mechanical issue with the mount. Is there any obvious play in the RA axis? My AZ EQ6 regularly needs tweaking to stop too much backlash.

The other point to consider is whether it is definitely the guiding giving poorly shaped stars? Is the camera completely perpendicular to the scope? A little movement could give wonky stars.

[Ibbo!]

I think your choice of star is too bright and saturating.

i run my old EQ6 without PEC as it seems to work better.

Apart from that and rechecking the fitment of things nothing comes to mind striaght away.

[rickwayne]

You've already done many of the tests and steps but in case you didn't have this one to hand (and for others reading this thread): https://openphdguiding.org/man-dev/Trouble_shooting.htm

1 second is not necessarily that short an exposure these days. Notice that the GA is recommending 1.8s as a drift-limiting exposure. I've found that 1s works pretty well with multistar guiding in PHD2, but of course that's with my rig and YMMV.

[lrt75914]

4 hours ago, Davey-T said:

How's the balance ? if you run it unguided and rest a finger on the weight bar you can push gently one way and the other to see if you can hold the RA more on track if so try altering the balance to mimic it.

Dave

I did fiddle about with balance but usually keep it east heavy. Never checked the balance by pushing on the weight bar with my finger, though so I might as well give that a try. Thanks for the tip🙂

3 hours ago, Clarkey said:

As Dave has said the guiding does not look that bad once settled and I am not sure it would be bad enough to cause obvious egg shaped stars. Looking at the trend there is clearly some oscillation in the guiding, which does suggest a mechanical issue with the mount. Is there any obvious play in the RA axis? My AZ EQ6 regularly needs tweaking to stop too much backlash....

The mount was rebuild last month and both axis should be as tight as I can possibly get them without running into binding issues. I was thinking about readjusting the worm gear tension/engagement and belt tension while tracking a star but I'm not sure if that is advisable.

3 hours ago, Clarkey said:

....

The other point to consider is whether it is definitely the guiding giving poorly shaped stars? Is the camera completely perpendicular to the scope? A little movement could give wonky stars.

There is no obvious play in the image train and I'm fairly certain that the camera is perpendicular to the scope. However I did notice that there is some play in my guide scope focuser. At least I can see some movement if I push on the guide camera itself. Is there a way I can check if any of these problems affect my guiding?

43 minutes ago, Ibbo! said:

I think your choice of star is too bright and saturating.

i run my old EQ6 without PEC as it seems to work better.

Apart from that and rechecking the fitment of things nothing comes to mind striaght away.

The weather this weekend seems to be fairly consistent so I'll try and see if a different guide star will do the trick

[Clarkey]

I think if your guide scope is focused and is locked in position there should not really be any noticeable movement. If there is, it's worth doing whatever you can to prevent it - even if it means taping it in place.

I did have to adjust the worm gears on my HEQ5 when it was virtually brand new as there was some play. It certainly helped. I think the main thing is not to over tighten or it will lead to premature wear.

[kens]

@lrt75914Can you post screenshots from PHD2 Log Viewer using the "Analyze selected frames" and Frequency analysis option? 

[Pompey Monkey]

I tried to use AutoPEC with my AZEQ6GT a few years back.

It failed miserably, basically because there is no worm position information coming from the mount.

The encoders only "tell" ASCOM where the telescope is pointing, not what's happening to the worm.

You are on a hiding to nothing and, as other have said, the guiding is OK without the AutoPEC.

 

[Starflyer]

I had an AZ EQ6 for a few years and they're great mounts, at the time the best mount available under £2.5k.  It handled 16KG of 10" newt / camera / filterwheel / OAG etc with round stars at 600s exposures.

First thing; disable the encoders, they're such low resolution that they're only useful for visual use and will adversely affect your guiding.  The option in EQMOD didn't used to exist, it was added specifically to solve the encoder / guiding problems on AZ EQ6 and EQ8 mounts when they first launched.

Second; don't overtighten the clutches, that will also mess with your guiding on this mount.  The purpose of the clutches isn't to lock the axis solid, they only need to hold the axis so they can drive the scope without slipping, and if balance is reasonable a light lock is all that's needed.

Third; you say you've had it apart a couple of times, so did I and this is what I learnt.  More grease isn't better, if you pack the bearings then stiction will cause jumps in your guiding as the friction builds until the drive can overcome it, a light smear on the bearings is all that's needed.  This also makes the axis feel looser and so easier to balance.  Set the belt tension correctly, easier said than done, but too tight is as bad as too loose.

Don't aim for perfect balance, it's often said that you should have perfect balance on all belt driven mounts, it's simply not correct.  All a belt drive does is to significantly reduce the backlash from the transfer gearing, it does nothing to reduce the worm to worm ring backlash and this what offsetting balance helps with.

The final thing, are you sure everything's solid and it's not differential flex between the main and guidescope?  The guidescope looks solid enough but the focuser on the main scope is out a long way, maybe an extension tube would allow it to be further in and better supported.

Persist with it, they're great mounts and it will easily handle your frac.  If you insist on upgrading then I'll help you out and take it off your hands for £500 😉

[rnobleeddy]

I got similar results last night, where as the night before I'd been at 0.4" in RA/DEC. I attributed that to the fact it was rather windy outside.

Otherwise the existing advice seems solid. As the issue is in RA I'd try purposefully unbalancing a little. 

Otherwise I'd be interesting to know if the behavior in consistent across the sky - I realize some level of deviation is normal, but my ageing EQ6 appears to perform quite differently depending on where in the sky it's pointing and the orientation - to the point where guiding is often markedly better one side of a meridian flip.

[alacant]

On 12/02/2021 at 01:33, lrt75914 said:

What am I doing wrong?

Hi

Remember to look at the images rather than the numbers, but anyway...

Lose the external autopec and instead guide RA using PHD2's PPEC with an initial period of 480s. 

If not, as per @kens' request.

Cheers

 

 

Edited February 13, 2021 by alacant