Strange star elongation issue

[angryowl]

14 minutes ago, steppenwolf said:

I may have missed something earlier in the thread but I notice that your RASA exposures are twice as long as those of the refractor - what was the reasoning behind not having exactly the same unguided image lengths?

Well, for one the clouds were starting to roll in when I mounted the refractor so I thought I'd get as many in as possible and two, I couldn't fit any filter inside the imaging train and stupidly thought that 600s exposures on the frac would oversaturate stars, making results harder to interpret. Only then I had realised that I was imaging at F11 which is an entirely different ballgame from F2.2. I had taken one 600s exposure on the frac I think but not kept it, but I can tell you the stars were as pinpoint as those in the 300s exposures.

On the RASA I only took 600s exposures as that’s probably the longest I’m ever going to go and my latest astro images have all had 600s subs.

[FLO]

28 minutes ago, angryowl said:

I have been told David Hinds would be the person to contact in such situations regarding Celestron issues within the UK. Is this correct?

We contacted them earlier this week (they are Celestron’s UK distributor / service centre). Sent them a link to this discussion asking if they think the telescope needs collecting / checking. 

Hope to have an answer for you soon. 

Steve 

[Davey-T]

If you read the Celestron blurb on manufacture of the RASA they claim the mirror slide mechanism is made to high precision tolerance so there should be no possibility of the mirror assembly  " rocking "

Next time we get a clear bit of sky I suggest aiming the scope due south at 0 Dec' and taking a series of exposures starting at 5 secs and then increasing exposure by 10 sec's at a time and taking 5 exposures.

Then aim the scope in the opposite direction somewhere towards Polaris and repeat the exposures.

If it's something moving in the optical system then presumably it will move then stop rather than gradually move more over time so the star trails should stay the same length regardless of exposure length.

The huge light grasp of the RASA means it can catch small star movements that may not be so noticeable in a longer focal length.

Dave

Edited March 1, 2019 by Davey-T

[souls33k3r]

50 minutes ago, Davey-T said:

The huge light grasp of the RASA means it can catch small star movements that may not be so noticeable in a longer focal length.

Well it did when he was using my EdgeHD 8 at 2m FL.

[angryowl]

13 hours ago, vlaiv said:

This really leaves mirror cell as being the main problem. If this is the case, it might be even possible to tell by inspecting back of the ota - see if back side of the scope is secured in place or if it has some slack - just use your hand and try to wiggle it a bit and see if you can feel it move (even tiny motion - looseness). If there are screws holding back cell on tube - see if any of them is loose and tighten them up with screwdriver if needed. It might be something simple as that and I think tightening loose screw would not void warranty.

Just tried this and the back aluminium mirror holder is rock solid with not even the tiniest movement detected when pulled and wiggled. Additionally, all screws are completely tightened and no play whatsoever there so I think this can safely be eliminated.

[steppenwolf]

2 hours ago, angryowl said:

Well, for one the clouds were starting to roll in when I mounted the refractor so I thought I'd get as many in as possible and two, I couldn't fit any filter inside the imaging train and stupidly thought that 600s exposures on the frac would oversaturate stars, making results harder to interpret. Only then I had realised that I was imaging at F11 which is an entirely different ballgame from F2.2. I had taken one 600s exposure on the frac I think but not kept it, but I can tell you the stars were as pinpoint as those in the 300s exposures.

 

On the RASA I only took 600s exposures as that’s probably the longest I’m ever going to go and my latest astro images have all had 600s subs.

Thank you for the explanation and confirmation that on the one 600sec refractor image that you have, the stars are good. When trying to assess these sorts of issues, I like a level playing field wherever it is possible and this was a wide deviation, now resolved.

[angryowl]

1 minute ago, steppenwolf said:

Thank you for the explanation and confirmation that on the one 600sec refractor image that you have, the stars are good. When trying to assess these sorts of issues, I like a level playing field wherever it is possible and this was a wide deviation, now resolved.

You’re absolutely right and many thanks for pointing that out.

This was a mistake on my part as I should have clearly mentioned that a 600s exposure also looked sharp on the refractor. I think I was a bit tired and frustrated when posting those tests, but still that’s no excuse.

[Astrobits]

Looking at your images, it strikes me that where there are long trails they are far too regular to be caused by some sloppy component. Surely, a sloppy component would slop from one position to another fairly quickly and then stop. A sloppy component would have a limit to it's movement so the movement should become less with time in one telescope position. Do repeat subs show the same amount of movement over the same time periods?

Nigel

[FLO]

8 hours ago, FLO said:

We contacted them earlier this week (they are Celestron’s UK distributor / service centre). Sent them a link to this discussion asking if they think the telescope needs collecting / checking. 

Hope to have an answer for you soon. 

Celestron U.K. (Hinds) have contacted Celestron in the US. for their opinion. Both are aware of this discussion. 

HTH, 

Steve 

[angryowl]

14 hours ago, FLO said:

Celestron U.K. (Hinds) have contacted Celestron in the US. for their opinion. Both are aware of this discussion. 

HTH, 

Steve 

Appreciate all the help on this Steve and look forward to hearing their opinions as to what may be causing this.

Edited March 2, 2019 by angryowl

[angryowl]

18 hours ago, Astrobits said:

Looking at your images, it strikes me that where there are long trails they are far too regular to be caused by some sloppy component. Surely, a sloppy component would slop from one position to another fairly quickly and then stop. A sloppy component would have a limit to it's movement so the movement should become less with time in one telescope position. Do repeat subs show the same amount of movement over the same time periods?

Nigel

This makes sense and logically thinking if we're talking about a sloping component it should eventually stop. Immediately after the telescope is pointed in a new position the trailing begins in the direction it wants as that depends on the position of the OTA. The trailing or slopping does indeed seem very regular and one would think that if something were sloping, over time, in a few minutes or an hour at most say, it should sort of stop. But through all my tests it didn’t.

[Astrobits]

If whatever it is that is moving is doing so consistently over a considerable period of time without stopping then logically whatever bit  is moving will fall over or out!! I assume that that hasn't happened. Also if something in the optical train is moving to shift the star positions then the collimation will be out at some point and should show up in your longer star trails as a changing of the width of the tracks in at least one corner. I see no evidence of this.

I think you should look again at the drive train of the mount.

Nigel

[angryowl]

Well for one I didn’t consistently monitor one particular object or position in the sky for hours on end. I never really spent more per night on an object than say three hours during which elongation always occurred.

Then two, if we’re entertaining the theory others have suggested here, that something is loose and moving inside the OTA like the mirror, then are we not talking about minute movements of the component here? Like arcseconds of movements in those elongations which in turn wouldn’t translate into a large measurable movement in the component itself? That way, at the RASA’s focal length, a millimetre of movement/slop in the mirror or any other component, could translate to several hours of movement visible in the stars? I could be wrong here, but this is how I see this play out.

I welcome any suggestions, like looking at the drive train of the mount again. But what exactly do you mean by that? As I’ve said previously, the mount is rock solid and I could not induce any movement anywhere including the front saddle. Balance has been checked so many times and always found to be spot on so the mount couldn’t be slipping in any direction. I now know the mount tracks well as it can produce pinpoint round stars at 900mm focal length at 600 second exposures.

[Astrobits]

While I agree that the movement in one individual exposure is only arc seconds, your post on the 18th:

Quote: "To illustrate the issue better, these are four 300 second exposures guided. Guiding was helpful in this case, but up to a point after which it couldn't keep up with the drift. I would have to stop guiding and exposing and re-centre the image every five subs otherwise it would slowly drift quite a ways". end quote.

clearly states that the movement continued for at least 5 consecutive subs at which time the total movement was sufficient to require re-centering the target. This amount of movement is much greater than a few arc seconds and should have caused an out of collimation situation if only one of the optical components was moving.  As this does not seem to be the case then the whole of the optical system must remain in their correct positions. The fact that the guiding set-up works at some points indicates that there is nothing moving within the RASA when the guiding is on.at that time. 

I cannot see this trail going anywhere but back to the mount and drive system.

Nigel

 

[angryowl]

From what I understand, I’m not sure the scope can come out of collimation so easily.

For instance, the backlash in the focusing mechanism is there due to not being able to have the tightest tolerances for the shaft that the main mirror slides on and this small gap is filled with grease. When focusing and looking at a star and going through the entire backlash I can definitely say that it shifts the star’s position more than the movement seen during those four 300 second exposures you’re referring to. Does that mean that the scope should be out of collimation after focusing or reversing focuser position?

If I’m completely wrong here and it’s something in the mount itself then how did the refractor worked so well on the mount prior to mounting on the RASA? I balanced both scopes equally well, I know the Mesu is more than capable of handling the weight of the OTA. I didn’t see anything out of the ordinary happening with the mount when the RASA was mounted after the refractor, I’m just at a loss as to what it could be if it’s something with the mount.

[Davey-T]

Can you piggy back the refractor on the RASA in case the movement is a product of more weight on the mount, then have a camera imaging on both scopes at the same time.

Dave

[angryowl]

Not able to do that Dave as I don't have the means to mount it properly on top. Could try a DIY solution of mounting it, but I fear it wouldn’t be sturdy enough and might introduce flexure which will affect the results.

[angryowl]

Dave, keeping a constant eye on the forecast and when I get a clear patch of sky, I’ll give the tests you mentioned a go.

The mount also has a feature that lets you see the amount of current motors on each axis are actively pulling from the controller. Next time I mount the heavier RASA on the mount I’ll keep a note of these values and compare with others I’ve seen that are considered normal.

Edited March 4, 2019 by angryowl

[Astrobits]

I don't think I saw a mention of alignment shift with focus/backlash previously in this post. Sorry if I missed it.

If that is the cause it still implies that there must be a limit to such movement while the telescope is in one position. You don't mention if you have an observatory but views from the house would indicate not. Pity, as you could leave the scope pointing in one direction for a day or two and then take some shots without moving the scope to another sky position. Any slow slop should have reached it's limit by then. Shots of the brick wall might not be representative due to the mirror being out of it's normal operating position.

I was thinking along the same lines as Dave with the extra mass of the RASA causing extra friction in some components or slippage in a clutch ( would not have thought so with the claimed load capacity but you never know).

Nigel

 

[angryowl]

I was simply referring to the common SCT mirror backlash/slop while turning the focuser in and out. This shows what I mean better. The mirror shift then is more than a few arc seconds I would image, yet the telescope is still in collimation. I was replying to what you said in regards to the movement visible in my four 300 second shots should have resulted in an out of collimation situation if only one component such as the mirror moved.

No observatory unfortunately, just a DIY tripod on top of which the Mesu is mounted every night. Just as a side note, there’s no chance of the tripod legs slipping into the ground as they’re 70 mm diameter steel plates.

Some good advice regarding the mount and food for thought. The Mesu is a friction drive mount so is always engaged but you are right, there could always be slippage somewhere even with this design I suppose.

[angryowl]

Was planning on having an imaging session last night as the forecast looked promising, but the conditions turned out to be very different so not had a chance to test anything else yet.

In addition to the tests and checks suggested in the thread, been thinking of a few more:

• With the refractor mounted, check the current draw on both RA and DEC servo motors in the SiTech controller. If the mount is struggling with imbalance or the weight once the RASA is mounted, hopefully this’ll show up. The mount has a very distinct sound to it whilst tracking and I’ll also keep an ear close to the mount during a few subs for any other sounds, like the servo struggling or slipping. As the elongation shows up in both RA and DEC then it seems logical that both axes would have to slip at the same time which I find hard to believe.
• With the mount tracking and scope pointing near the zenith, I’ll rack the focuser up and down and take exposures to try and measure how much the mirror shift is in arc seconds. Because the scope will be pointing up where the elongation never shows, I’ll be looking purely at how much the mirror is free to move in its carriage whilst racking the focuser up and down.
• Align the guide camera to the same orientation as the camera on the RASA and point scope in different positions in the sky. As I can’t take long exposures on the guide camera due to having no filter in, I’ll simply take some short exposures on both cameras every 10 minutes and check if they both drift in the same direction. I did this before, and although the cameras didn’t have the same orientation, I remember working out the drift was in the same direction on both. If this gets me nowhere, I’ll put together a metal mounting plate and piggy back Dave’s refractor on the RASA as he suggested earlier.
• Another test I’m thinking of is pointing the RASA in a position elongation occurs, start tracking unguided, take a 10 minute sub and stop tracking. I’ll let both mount and scope just sit there for an hour or so, then without touching anything, start tracking and take another exposure. If there’s anything moving inside the OTA, then as @Astrobits puts it, it should stop moving in that hour and the second exposure should show no elongation. Would an hour be enough or should I wait more?

I’ll also run another test to see exactly at which point stars start to elongate. I’ll start at 30s and work my way up 30s at a time.

All of the tests above will be done with the mirror locks fully engaged.

Looking at most images taken with the RASA on AstroBin, and I’ve found only a handful of images using 5 minute exposure times but the norm is 3 minutes. The 5 minute ones on AB show no sign of elongation whatsoever and although my tests were done at 10 minutes, I can definitely see elongation even in 5 minute subs. I can get away with 3 minute guided subs, but I’d like to go longer than that with narrowband due to the read noise of my Atik 414EX being 5-6 electrons.

[angryowl]

@FLO – Any response from Celestron? Is this considered normal behaviour in the RASA?

If not any suggestions for any tests I can perform or do they need more info?

If this is normal and just regular SCT behaviour, then I suppose the only option is to go the OAG route and try to find the smallest camera I can and figure out a way to mount it all so that it doesn’t cut into the light path. With a bit of DIY this just might be doable.

[vlaiv]

16 minutes ago, angryowl said:

@FLO – Any response from Celestron? Is this considered normal behaviour in the RASA?

If not any suggestions for any tests I can perform or do they need more info?

If this is normal and just regular SCT behaviour, then I suppose the only option is to go the OAG route and try to find the smallest camera I can and figure out a way to mount it all so that it doesn’t cut into the light path. With a bit of DIY this just might be doable.

Maybe look into getting cameras with integrated guide sensor? I've seen at least one such model - it has primary sensor and small guide sensor mounted right next to it - this qualifies as OAG although there is no pick off prism. Makes focusing easier though - both sensors come to focus at the same point.

Just found this online, it looks like SBIG has a patent to this sensor arrangement - so probably only option for this.

[FLO]

5 hours ago, angryowl said:

@FLO – Any response from Celestron? Is this considered normal behaviour in the RASA?

I received a communication from them 5th March saying they are watching your thread with interest and noted from your posts you were planning to perform some additional tests to your mount and the RASA. They said they were looking forward to the results. 

Currently, neither they nor we can see anything here that confirms your RASA is faulty. I.e. it could be the mount. But when I say that I am aware it might appear we are being evasive. FLO doesn't do evasive so please be assured if your tests confirm your telescope is faulty we will collect it then forward it to Celestron-UK for testing. 

HTH, 

Steve 

[angryowl]

1 hour ago, vlaiv said:

Maybe look into getting cameras with integrated guide sensor? I've seen at least one such model - it has primary sensor and small guide sensor mounted right next to it - this qualifies as OAG although there is no pick off prism. Makes focusing easier though - both sensors come to focus at the same point.

Just found this online, it looks like SBIG has a patent to this sensor arrangement - so probably only option for this.

Thanks for the suggestion and I did consider this at one point but because the dual chip self-guide models available are physically large they would end up obstructing more of the light path than I’d like.