... differential flexure?

[lukebl]

Sorry if I'm repeating stuff which has been said before, but can anyone tell me what's the cause of the drifting of these images on this animation which is made up of 5 separate subs.

I was imaging M57 last night using the following kit:

HEQ-5 pro

Canon 400D attached to Skywatcher 200

Guided with PHD + webcam + 90mm scope attached to main scope

Is this an example of diff flexure and, if so, how can it be happening if the guide scope is fixed firmly to the main scope with two rings, and how can I stop it? Or is it something to do with bad polar alignment?

The guide star remained stationary on the PHD screen throughout, and I used a 3x barlow on the guide scope so it had a considerably greater focal length than the imaging scope.

Incidentally, I made this animation online through a website called Gickr.com which automatically turns jpegs into animations. Pretty easy and handy.

Many thanks,

Luke

[barkis]

I find it difficult to see where your problem lies. What is the animation trying to show here.?

If your guide star is being faithfully tracked, then any star trail will appear to rotate around the guide star. That is usually indicative of poor polar alignment.

Ron.

[lukebl]

What is the animation trying to show here.?

The point is that if it was guiding correctly, each frame would be identical and the stars wouldn't move at all.

The whole point of guiding is that the object of one's desire doesn't move!

L

[barkis]

So, an Image of all the subs stacked will show bad trailing. These are frames of each of the subs.

I understand what you say about the frame movement indicating something wrong, but the final stack should show the sum of the trailing, which might give a clue to the possible cause.

Ron.

[blinky]

It does look like diff flexture, I have the same problem. I need to wait about 6-8 weeks before I can start again as it's not dark in Edinburgh at the moment. Anyway I digress...... What you need to do (and what I shall be doing) is to tighten everything up. So make sure the guide scope is tight in the rings, if it has a rotating focuser make sure that is nice and tight... etc etc ALso make sure that any cables from the camera's dont tug slightly on the scopes either.

Sombody worked out somewhere how much flexture will cause a few pixels drift and it was less than 1/10th of a mm so you can see how everything needs to be really locked down.

[lukebl]

It's just occurred to me that the obvious cause of the problem is the distance from the guide star to the object being imaged.

In this case, as I'm only using a webcam to guide, I had to pick on a bright star and used Vega which is obviously a few degrees away. If I'd used beta Lyrae, which is clearly very much closer, then the drifting wouldn't have been so great.

Sorted. Apologies to those with greater experience than I for not realising the obvious!

Luke

[Psychobilly]

Whats the logic behind that Luke...?

I quite often use guide stars "away" from the target without problems...

Looking at the anim no star says in the same place...

Peter...

[Milamber]

The point is that a star closer to the pole will move less for a given drift than one further away. If you guide on Polaris for example (and you were polar aligned) and imaged (say) M16 then any movement of the target would not be accurately mirrored in the guide star. Hence why it's a good idea select your guide star as close to your target as possible.

Arthur

PS - having said all that, the snip you posted just looks to me like your guiding is not working, plain and simple.

[themos]

I had the same kind of drift with a piggyback setup (450D@135mm, guided by ToUCam@900mm) and my guide star was right in the middle of the frame. I'm pretty sure it was flexure.

[themos]

Backyard Astronomy contains useful tips.

[lukebl]

Whats the logic behind that Luke...?

Oh dear, maybe I'm wrong then.

What sort of angular distance to your guide star is cosidered acceptable? I just assumed that the drifting was due to the fact that the guide scope was pointing at Vega, the imaging scope was pointing at M57, and the relatively large angular difference betyween them was resulting in this effect. I assume that I will need to get a better guide cam so that I can lock on to fainter stars much closer to the object being imaged.

The drift obviously limits potential exposure times. The images on the animation were around 3-4 minutes each, with about a 5 minute gap between each, and the individual images show quite a bit of trailing although not as much as without guiding.

Regards,

Luke

[lukebl]

PS - having said all that, the snip you posted just looks to me like your guiding is not working, plain and simple.

Some sort of guiding is going on, at the guide star (Vega) stays perfectly stationary in the guidescope's FOV.

Confused!

L

[themos]

I hope that PHD adds a feature to counter flexure. In my miniscule experience, flexure seems to be a pretty uniform drift that can be estimated within 10 mins or so. We just need the direction and magnitude and PHD could guide it out.

[SteveL]

I hope that PHD adds a feature to counter flexure. In my miniscule experience, flexure seems to be a pretty uniform drift that can be estimated within 10 mins or so. We just need the direction and magnitude and PHD could guide it out.

That would require PHD to not only see the images from the guide camera, but also know how long the images are, and be able to process and calculate the difference between that and the subs coming out of the imaging camera. Its certainly doable, but I doubt it will happen any time soon.

You also have to take into account that flexure is not a constant value/rate in any direction, it depends on where the mount/scope is pointing, and the forces/angles/leverage involved with the scopes, dovetails, SCT mirror flop, mounting equipment, rings, screws, mount. You also have to take temperature into account of you want to be picky, where materials flex less when they get colder. At no point in the sky is the amount of flex the same as any other spot.

This leads us to the fact that any flex adjustment in PHD would only be possible after an image, and would only adjust/guess the flex amount and direction in the next image by using the data from the last image.

Interesting idea though

[Psychobilly]

I thought Luke mean t a star in the same "region" as the target but not neccessarily right next to it...

My guide and imaging scopes are all co-aligned and i tend not to slew then around with respect to each other...

I tend to select a guide star from anyhwere in the FOV of the guidecam not neccesarily a central one...

Peter...

[Milamber]

He did, Vega and M57 not *that* far apart, but it's easier to make a point at extremes, makes visualisation simpler.

[themos]

SteveL, you are right in all of these but my first experience of flexure shows that the first order effect was a uniform creep over the course of an hour or two and my interest was aroused after seeing more examples of such an effect by others on SGL. It would be easy for PHD to accept user input of a uniform creep (heading + arcseconds per minute) to correct for. It would be hard(er) for software to detect that creep (astrometry.net could do it with two timestamped exposures) and I wasn't envisaging PHD doing that (it's supposed to do one job well).

[RobH]

It does look like flexure.

With my guidescope, a WO ZS80, I need to have the drawtube near the camera supported with a third ring to stop the small amount of sag that happens. Before I did this, even with everything tightened up with molegrips, I got flexure.

Using Vega as a guidestar and imaging M57 should be fine, they are easily close enough.

Cheers

Rob

[lukebl]

Hi,

Thanks for all the responses. So it could be differential flexure.

Here's a picture of my setup. Can anyone tell me if this configuration is more or less prone to differential flexure? And, if so, what should I do to sort it. It all feels pretty firm and solid to me.

I've read a lot of postings about this topic, but can't find a definitive answer!

Cheers,

Luke

[Milamber]

That looks pretty solid to me - unless your main scope rings are loose or the cameras are not clamped tightly I can't see what would be flexing there.

Arthur

[RobH]

Your guidescope looks like a skywatcher Mak to me?

I haven't had a play with one of these, but looking at a few pics, it looks like you focus by moving the primary mirror....am I correct?

If so, the cause of your problems could well be mirror flop, where the main mirror itself moves a little as the scope changes orientation.

Mirror flop is a common problem with many CAT type scopes.

Try using a small refractor as a guidescope...no mirror flop with a 'frac!

Cheers

Rob

[lukebl]

If so, the cause of your problems could well be mirror flop

That's a new one on me. Anyway I bought the Mak to use as a guidescope, so it would be hard to justify mothballing it! I'll try really tightening everything up and see if that helps.

So the thinking is that using a guide star as far from the object as, say, vega is from M57, is OK?

Cheers,

Luke

[RobH]

That's a new one on me. Anyway I bought the Mak to use as a guidescope, so it would be hard to justify mothballing it! I'll try really tightening everything up and see if that helps.

So the thinking is that using a guide star as far from the object as, say, vega is from M57, is OK?

Yes, Vega will be fine as a guide star for M57, as long as it isn't too big on the PHD screen. If it is, PHD will find it hard to determine the centre.

Go for something around magnitude 3 if you can.

If mirror flop is the problem, then no amount of tightening things up will make any difference I'm afraid.

It's why most of us stay away from small Maks as guidescopes.

See of you can borrow a small refractor and run some tests to determine if it is mirror flop or not.

Cheers

Rob

[Tim]

I had pics just like these, and diff flex was diagnosed, by the author of phd among others.

Easy answer, get an off axis guider, and sensitive camera, or use shorter subs.

Although, I found that my worst flexed pics of M57, with quite bad trails when stacked, actually resulted in a very highly detailed M57 when they were pulled and pushed back into round with PS, and the details matched closely to a pro pic.

If you need any help re Off axis guiding, just shout.

[SteveP]

Luke

It's a little difficult to diagnose the problem. What you seem to have is 5 frames from the webcam. The direction of the drift looks to be roughly in the E-W direction. It looks fromthe 5 frames that the drift is fairly regular. I'm not convinced that it is necessarily differential flexing. There isn't enough information to go by. It looks from just the 5 frames that it is guiding while the frame is exposing ie there isn't much evidence of star trailing and then not guiding between images. Any star trailing is nothing like the change in position between images and yet the times are not dissimilar ie 3-4mins vs 5 mins. Is there anything in your software (I'm not familiar withPHD) which allows the mount to stop guiding between frames Secondly, are you positive that the imaging star isn't moving in the guidescope? Sometimes it can be difficult to determine whether there is any drift or not.

Steve