Whats causing this drift?

[david_taurus83]

Whats your first thoughts? Pretty much consistent direction in Dec. This is 22 x 900s subs from last night so a period of over 5 hours. Guiding was ok (or so it seemed) with some blips in dec every now and again. Polar aligned in Sharpcap so pretty good.

 

 

[PeterCPC]

Have you checked the balance?

[david_taurus83]

2 minutes ago, PeterCPC said:

Have you checked the balance?

Yes its not bad at all. Its on the CEM25 so very difficult to balance perfectly without adding off axis weights etc. I've not noticed anything like this before. How would balance cause it btw?

[PeterCPC]

If it's not balanced, the motor is fighting against the imbalance and may not be able to stop it drifting with the imbalance. Just a thought but probably not that then.

[symmetal]

Are you using 'random' dither in PHD2. I've found it's not very random and tends to one direction. When cropping the edges after stacking only 2 adjacent edges need cropping off. I thought that 'circular' dither may be better but the final result is not much different.

Alan

[david_taurus83]

No dither at all!

[Seelive]

I would say it's drift due to polar alignment. It appears to be about 3 arcmin total over the total exposure time of 5.5 hours (assuming the scaling shown is correct). That drift could be achieved with a polar alignment error of less than 5 arcmin at that declination. At the end of the day, that's what stacking software is for.

[kens]

If the guiding was OK (a guide log would be helpful), then it would be differential flexure of some sort. The "blips in dec" may offer a clue.

[david_taurus83]

1 hour ago, kens said:

If the guiding was OK (a guide log would be helpful), then it would be differential flexure of some sort. The "blips in dec" may offer a clue.

There you go. Looks like my polar alignment was out after all, over 3 minues error. It was excellent in Sharpcap but I must have disturbed it somehow. I tried to rotate the imaging train by 90° but after testing, the filterwheel would have likely hit the tripod near the meridian so I rotated back. Pretty sure I done this after polar aligning, I think. I done 10 minute subs a few nights ago and the PA error was 0.5 minutes and the stars were pretty good. Hopefully its an easy fix.  Rebalance, careful PA and recalibrate PHD next night out. The guidescope is the William Optics slide base type btw so I hope theres no flexure either! Feels pretty solid.

PHD2_GuideLog_2020-10-09_194811.txt

[kens]

3 minutes PA error is not a problem. PHD2 was able to keep the guide star centred in the guide camera but the image camera shows a steady drift. So  one camera is moving in relation to the other. My first guess would be that the dec blips are a result of the guide scope moving. Given the dec movement I would look at whether it is rotating slightly where the guide scope is attached to the main scope. We are talking microns here.

Irrespective of that, you also appear to have backlash or stiction issues.

[david_taurus83]

20 minutes ago, kens said:

3 minutes PA error is not a problem. PHD2 was able to keep the guide star centred in the guide camera but the image camera shows a steady drift. So  one camera is moving in relation to the other. My first guess would be that the dec blips are a result of the guide scope moving. Given the dec movement I would look at whether it is rotating slightly where the guide scope is attached to the main scope. We are talking microns here.

Irrespective of that, you also appear to have backlash or stiction issues.

I had the cables routed under the carry handle. Probably the cause? I'm going to put a longer dovetail on the bottom, the short one on the top and then the carry handle to that. Should add more rigidity. Will try to route cables under the scope. I like to have them fixed along the dec axis so they aren't hanging off the back of the imaging train. Didn't think routing them through the handle would be an issue!

Yes, it has some dec backlash but I've reduced it as much as I can. It calibrates in North almost instantly and returns South in 3 or 4 steps but if I run the GA it can never seem to measure the backlash. Not enough movement detected or so.

[david_taurus83]

Well, I think that'll do it. Slightly concerning to say the least how I missed this.

 

[newbie alert]

Oops!

[tomato]

2 hours ago, david_taurus83 said:

Well, I think that'll do it. Slightly concerning to say the least how I missed this.

 

Glad you found the problem and hey,  I’ve left the counterweights off on more than one occasion, but you tend to get more instant feedback on that omission.

[david_taurus83]

Well I tightened everything up again. Drift is still present so flexure as suggested. I'm hazarding a guess and saying its focuser sag as the drawtube on this setup is around 60mm out at focus. I've measured the drift at approx 1 pixel (2") per 2 x 600s subs. It shifted 20" over 3 hours last night. Not an issue doing 10 minute subs but if I want to go longer it will be. OAG it is then! A shame as guiding has been spot on as well! Need to go tweaking again!

 

 

[Brujo]

Have you solved this problem ?
I am having the same problem and just can't get out of it.
Polar alignment is ok, guiding is ok, no differential flexure, no heavy cables.

And still having this drift of 1 pixel per minute on a 6" RC with ASI533 camera.

 

[david_taurus83]

2 hours ago, Brujo said:

Have you solved this problem ?
I am having the same problem and just can't get out of it.
Polar alignment is ok, guiding is ok, no differential flexure, no heavy cables.

And still having this drift of 1 pixel per minute on a 6" RC with ASI533 camera.

 

Definitely flexure. I've moved to a mini guidescope on the focuser so imaging camera and guide camera are on the same tube, as opposed to being separated by the scope rings and handlebar. There is still some small drift over the evening but not to the degree seen above. I can still do 15 minute subs without any noticeable drift in each sub. That may be because I'm guiding on a star off centre of the main imaging camera. OAG guiding completely eliminates it btw but it puts my guide camera in an awkward place and it gets very close to crashing into the tripod near the meridian.