AVX Guiding Performance

[jimjam11]

I have had my AVX for a few months now and it is great. I haven't really adjusted PHD2 settings and am getting an average RMS of around 1.3" which is below my imaging scale. However, I noticed the manual recommends adjusting the anti-backlash settings to elevated values for guiding. I have tried this on two occasions and guiding was terrible, particularly on the dec axis. I seem to get best results between 10 and 25. Anything higher than this and it lurches wildly from one side to the other (anywhere up to 12")!

Has anyone else noticed similar behaviour, what kind of settings are you using and what kind of stats are you getting?

If the rain ever stops I am going to look at the guiding rate (default to 0.5x, so will try 1x), and then I need to look at the PHD settings themselves.

TIA,

[orley]

Had mine runnung with default settings and it guides beautifully. Haven't yet found the need to tinker. The only problem I've had was with a dodgy cable. Admittedly I'm not going for image of the year but using a QHY5//finder combination and an old Canon DSLR I get nice tight, round stars from 10 minute subs. If only I coukd get rid of the darn light pollution that is forever my bugbear! One day...

[PeterCPC]

I would suspect that any guiding issues are down to PHD and not the mount.

Peter

[knobby]

Had mine runnung with default settings and it guides beautifully. Haven't yet found the need to tinker. The only problem I've had was with a dodgy cable. Admittedly I'm not going for image of the year but using a QHY5//finder combination and an old Canon DSLR I get nice tight, round stars from 10 minute subs. If only I coukd get rid of the darn light pollution that is forever my bugbear! One day...

With what scope ? I've just started dabbling in the dark art of guiding [emoji3] . I'm using a finder guider /ASI 120 mc and a William optics Zenithstar 71.

[jimjam11]

I agree it guides well, but I think there is room for improvement. In general I find it overcorrects, albeit by a small amount. I think with some tweaking of the configuration even better performance can be achieved.

I also suspect some of my issues are down to the guider chasing the seeing (using and st80 as guidescope) because my exposure time for guiding is usually around 1-2s which I think is to fast.

[orley]

With what scope ? I've just started dabbling in the dark art of guiding [emoji3] . I'm using a finder guider /ASI 120 mc and a William optics Zenithstar 71.

SkyWatcher 100ED. It's a bit slow but perfectly servicable!

[knobby]

Ta mate, so a wee bit longer than my W.O. 71 , there's hope for me yet then !

[orley]

I agree it guides well, but I think there is room for improvement. In general I find it overcorrects, albeit by a small amount. I think with some tweaking of the configuration even better performance can be achieved.

I also suspect some of my issues are down to the guider chasing the seeing (using and st80 as guidescope) because my exposure time for guiding is usually around 1-2s which I think is to fast.

I rarely go above 1 second.

[MattJenko]

1 second is quite short for guiding from my experience. I actually use 4 seconds as a general rule. Longer exposures prevent the sending of too many corrections to the mount, meaning it is not being micromanaged, as I have found my mount (HEQ5) cannot respond that accurately in that short a time frame and it doesn't seem to behave that well being moved small amounts back and forwards all the time. It also appears to smooth out atmospheric distortions so you are not chasing local seeing conditions.

[orley]

I get a pretty smooth graph! Perhaps though I'll try longer exposures and see what the difference is.

[jimjam11]

This video was useful and confirmed my suspicion that I was chasing the seeing. After watching this I changed my guide parameters to the following:

RA Aggressiveness - 55

RA Min Move - 0.56

Dec Aggressiveness - 100 (Unchanged)

Dec Min Move - 0.6

This has significantly improved my RMS figure (0.9"-1.0"), but more significantly my peak-peak has dropped a lot. It tended to lurch from one side to the other, sometimes by a fair amount (perhaps 4"). It now tracks within 2" all the time with no wobbles. Balance also seems less critical so I think it is working much better, although I guess only time will tell!

I think I will try increasing exposure time next to see what difference that makes.

[libraryman]

Try running the guiding assistant, its very good and will likely recommend good settings, including an exposure around 2/3 seconds.

As an aside, my EQ8 guides around 0.8 RMS average.

Ray

[jimjam11]

I did, and used its recommendations as my starting point which really helped. It did also complain that my polar alignment was off, but that didnt seem to make any difference!

[libraryman]

If you have run it long enough, I assume that you have; then its measure of your PA is likely accurate, of course your mounts pa error is also in the altitude adjustment not just azimuth and whilst it may not impact heavily on your guiding; it could well be......depending on how far out it is be the difference your looking for!

Ray

[jimjam11]

It seemed to think I was about 10' out after a 2 star align, 1 calibration star and aspa which I thought that was acceptable.

[libraryman]

I guess this depends on how I read that number (10'} I would say that was very unacceptable for imaging! 10 arc min's ?

Rayp

[jimjam11]

I thought that was pretty good, lol!

[Rob63]

As another AVX user, can I just jump on this thread and ask about PHD2 calibration.

When I do my calibration it runs through west/east fine (about 13 steps) and then moves to north/south.

It completes north in about 16 steps but when doing south the star barely moves - if it has moved 25px in North it may only shift back by 5px.

is this normal or do I have a problem?

[Freddie]

The difference in N and S is the mount DEC backlash. Once it has moved N, the first few commands to move S do nothing as the backlash in DEC is taken up. Sounds like a lot of backlash but I have no experience of that mount.

[jimjam11]

As another AVX user, can I just jump on this thread and ask about PHD2 calibration.

When I do my calibration it runs through west/east fine (about 13 steps) and then moves to north/south.

It completes north in about 16 steps but when doing south the star barely moves - if it has moved 25px in North it may only shift back by 5px.

is this normal or do I have a problem?

Check the anti backlash settings on the hand controller and maybe tweak them. Using an st80 and qhy5lii and a calibration step of 750 I got 11 west steps followed by 10 north.

[PeterCPC]

I read on the Team Celestron site that a 10' error should be capable of being dealt with by PHD. That's thrown that grenade!

Peter

[jimjam11]

Lol, my guiding was excellent.

Interestingly phd guiding assistance calculated my PA error as 10', but phdLab calculated it as 8.8',2.2',2.92', and 5.6'. This is connected to the mount through ascom so the dec angle and positioning data was known to phd.

[Rob63]

Re: DEC backlash. I don’t think this is the issue, 

I have taken the cover off the DEC motor and the gears are nicely meshed, there are no excessive gaps and there is no wiggle between them.

I ran dec at a slow motor speed and adjusted the anti backlash settings to 10 in dec +ve and -ve (i.e. not very much). 

When I use north/south buttons on hand controller the response is instant with no jumps or delay in changing direction.

But when calibrating is still get the problem that south doesn’t "unwind" very much BUT there is no delay in starting movement, it just doesn’t move very much.

i.e. if it has moved 24px in 15 steps north - when moving south it does start moving immediately but only by about 0.5 - 0.75px at a time and never catches up.

Maybe it's a balance issue, pretty difficult as the ED80 is small and light and the AVX dec shaft is very stiff (seems they all are).

I don’t know if I should care too much, my guiding is around 1" RMS with my finderguider and 6min exposure looks pretty good with slight egging at the corners.

Any thoughts? 

Rob.

[jimjam11]

Re: DEC backlash. I don’t think this is the issue,

I have taken the cover off the DEC motor and the gears are nicely meshed, there are no excessive gaps and there is no wiggle between them.

I ran dec at a slow motor speed and adjusted the anti backlash settings to 10 in dec +ve and -ve (i.e. not very much).

When I use north/south buttons on hand controller the response is instant with no jumps or delay in changing direction.

But when calibrating is still get the problem that south doesn’t "unwind" very much BUT there is no delay in starting movement, it just doesn’t move very much.

i.e. if it has moved 24px in 15 steps north - when moving south it does start moving immediately but only by about 0.5 - 0.75px at a time and never catches up.

Maybe it's a balance issue, pretty difficult as the ED80 is small and light and the AVX dec shaft is very stiff (seems they all are).

I don’t know if I should care too much, my guiding is around 1" RMS with my finderguider and 6min exposure looks pretty good with slight egging at the corners.

Any thoughts?

Rob.

That is the challenge I find. The desire to optimize something even though it isn't image affecting!

[knobby]

Seem to recall something about setting calibration steps to 1500 or 2000 for lack of movement ...