Diagnosing differential flexure

[cgarry]

I am working through some teething troubles with my updated DSO imaging kit and need to get to the bottom of some flexure I am seeing.

The setup:

*  Mount: CGEM DX

* Scope: SW Esprit ED 120 PRO Triplet

* Guider: QHY5 + 9x50 finderscope

* Imaging train: Atik 460ex, EFW2 filter wheel, Adapter (4 mm), Spacer (16 mm), x0.79 Reducer/Corrector

Last night I captured 8 x 30 minute Ha subs back to back.

This is the first sub (with curves applied) to show frame orientation:

This is a repeating video of a 400 x 400 crop from the 8 x 30 minute subs:

The first sub started at 00:39 (2014-03-10) and the last sub started at 04:12.

The first 6 subs seem to show a steady drift of flex but had round stars in the main.  The last 2 subs had a more drastic movement in a slightly different direction, I do wonder if this par twas caused by cables getting tight.

So, I want to figure out if the finder guider (connected using just its foot) is the problem or if the scope's focuser or some other part of the imaging train is the problem.  I am wary that the focuser needs to be wound out a long way (7 cm) to reach focus and I am sure I can see movement if I apply a little left/right or up/down pressure to the end of the camera.

Are there any techniques to figure out where the flexing is coming from that doesn't take multiple nights worth of testing?  I would like to get this system working with the finder guider if possible rather than trying to squeeze an OAG in where the 16 mm spacer is currently.

Cheers,

Chris

[ollypenrice]

The guide rings are small and close together while the guidescope is quite long relative to the ring spacing. This cannot be good practice and since you shouldn't need to hunt for guide stars I would bolt that guidescope down hard on something home made and ditch the rings. I'd run a plate to span the main scope tube rings and attach it to that. The little finder shoe isn't the beefiest piece of metal, either, and the cables can pull with some leverage as it stands.

If you ran well past the meridian I think that you can expect guiding to go off a bit. On all the mounts I use I see this happen, though it usually remains acceptable. I'm not sure why this is. Balance, probably.

Olly

[AstroAdam]

I agree with the securing of the guidescope without rings.  A small finder guidescope like that shouldn't really often have any issues locating a guide star...  Mine is solidly attached...

[IanL]

Avoid a single big loop of cable from the cameras.  Instead:

- Make a very short loop from the cameras to the first tube ring of the imaging OTA. Fix the cables there with a tie, velcro or whatever you prefer; you can add intermediate fixings if required. Make the loop as short as possible to reduce the weight on the cameras, but not so short that it pulls on the connectors or cameras.

- From the tube ring make another loop to the main body of the GEM (i.e. the part that rotates in RA), with sufficient slack that the OTAs can rotate fully in Dec. Fix again.

- From there make a final loop to the mount base/tripod head with sufficient slack to allow full rotation in RA.

- Run back to the laptop/computer from the final fixing.

You should end up with cabling that minimises the drag on the cameras/OTAs and also has less chance of snagging.

The three point finder rings are not a good idea and I concur with the others that you need something more solid.  You don't need to worry too much about aligning the imager and the guider. At the kind of focal lengths you are using field rotation is not going to be an issue for any realistic length of sub provided the two are pointing in roughly the same direction.

[ChrisLX200]

Good points made so far. I would add that you may want to use a longer adapter for the camera so the focuser doesn't need to be extended so far, it may be more stable further in. Even a thou or so movement here will cause eggy stars. With a separate guidescope absolutely everthing must lock up tightly to avoid flexure and if you have doubts or the equipment isn't up to it then go with an OAG.

ChrisH

[cgarry]

Thanks for all the responses, excellent advice that I shall be putting into action.  I shall come back to this thread with an update when I have made the changes and done some more tests.

Cheers,

Chris