27% Tilt on my Celestron Imaging Train

The biggest source of tilt in any longer focal length, slower systems systems such as a C8 are tilt adjusters.  Normal machining tolerances should be sufficient to hold the sensor to within the depth of focus over the entire field.  At F/10, the depth of focus is around 250 microns, which is a lot.  I’m not familiar with how ASTAP works and you didn’t show any results so I can’t comment on that.  Have you used the FWHMEccentricty tool in PI to look at the field?  If you measure a dense star field, the FWHM values shouldn’t vary by more than about .5 pixels and the contours of the eccentricity plot should be roughly centered with values less than 0.5.  

My suggestion is to start by zeroing the tilt on all of your adjusters (or simply removing them) and starting from scratch.  I never use tilt adjusters and I’ve never seen tilt on the 4x C14s or on the CDK20 that I’ve used.  Unless there is something really screwed up on your C8, it shouldn’t be much different.  Just be sure that EVERYTHING is properly seated and pulled tightly together in the image train when you assemble it and you shouldn’t see much sensor tilt.

John

Mike,
Ah…I'm sorry.  I am familiar with the NINA image sensor tilt algorithm–and it's pretty good.  I agree that getting the back working distance set correctly is critical.  There are ways to end up with a tilted image plane with respect to the reference datum on the rear of the telescope but that's only going to happen unless something is seriously out of whack.  Just be careful that everything is mounted flush and that nothing wobbles when you firmly grab it and try to move it.  Try what I've suggested and if you still can't get it right, let us know and we'll come up with plan B.  It might mean that the scope itself has something out of alignment.  Speaking of which:  Is the scope properly aligned to begin with?  I didn't originally take note of the fact that you are using an older, non-Edge system, which means that you'll have both coma and field curvature–most noticeable with a larger sensor.  That may be affecting the ASTAP output as well.  Having a small misalignment in the secondary will mimic sensor tilt because it will affect the symmetry of the aberration field.  So, it's important to start with a very well aligned secondary before you try to evaluate sensor tilt.

John

Collation is very close. Although, it wouldn't hurt to check it again. The telescope side of the reducer is a 2" barrel that inserts in the Baader Click Lock. Thr other end is 42mm male threads. Thank you Cesar.

I'm going to try something I heard. Turn the scope with the corrector down to the ground. Then insert the imaging train into the click lock and tighten securely.  That way the weight of the train is not pulling down at the time of tightening.  Just a fancy, maybe? 😜

Mike

Thank you for the suggestions Cesar.  I'll let you know the results.  


Mike

🤣 I would be thrilled to get it down to 4 or 5%. I'll post a picture of the imaging train as I have it right now. I did away with the OAG. Just couldn't get it to where I wanted it. Went back to a guide scope and the guiding numbers are great. Easier to find usable stars. You may have misunderstood. The thread coming off the Starizona corrector "is" 42mm male. I've achieved the 90.3 back focus. I'll recheck the collimation and try to better center the corrector in the Click Lock. Then, test again. I'll post my results. Very generous of you to offer your help. Thank you my friend 😁

Oh Bob I didn't see that at all. I thought the click was the best way to center and support the image train. When I tighten it, the image train is rock solid. I have a common Celestron visual back that uses one or two screws and the typical band of metal that tightens against the 2" barrel. So you're saying that is better? I'm certainly willing to try it. 


Mike

Andy 01:
This tilt/spacing adjusting gadget may be of interest - I'm looking into getting one myself. 

https://www.asgastronomy.com/photon-cage/

Cheers

Andy

I'm sorry, but you do not need that kind of kind of adjustment on a C8 and I've never needed anything like that with a Hyperstar based system either.  The depth of field on a F/10 system is 250 microns (+/- 125 microns) and that is well within normal machining tolerances for fixed spacers--either threaded or with dovetails.  I don't know anything about the Click-Lock that the OP is using but if it is something different than a simple machined spacer, don't use it.  With a properly configured and assembled camera package, you should be able to grab the entire image train at the end of the telescope and physically apply a fair amount of force back and forth on the camera and feel ZERO play--and you should see virtually zero sensor tilt.

Here are a few things to watch out for:
1) Flimsy components.  I once used a Pegasus filter wheel that was made from very thin metal that appeared to be made of architectural grade aluminum with a LOT of mechanical flexure.  I could easily see it flex when I applied even a light load so I tossed it out and went back to FLI.  In my opinion, that product was a piece of junk.

2) Improperly designed components.  I have an Optec Sagitta AOG on my refractor and it is not well designed to avoid mechanical interference.  When I first installed it, I wound up with a lot of tilt and I tore my hair out in an effort to find out why.  It turns out that the focusing stalk on the OAG is about 0.010" wider than the base and if the unit is placed flush with a FW, that 0.010" wouldn't allow the dovetail to be pulled uniformly flush when everything is tightened down.  It was such a small amount that it wasn't immediately obvious what was going on.  I finally found the problem by wiggling everything.  I found that in one direction, everything was solid, but in the other direction, I could detect just a tiny bit of play.  By rotating the OAG to avoid having the focusing stalk next to the FW, I could properly tighten everything to be perfectly flush, which eliminated all but the smallest amount of the tilt--over a 42 mm image circle.  In my opinion, the Sagitta OAG is a nice product but Jeff should redesign it to fix this problem.  And a fix would be very simple.

3) Normal dovetails that you find on astro-gear are not kinematic.  Furthermore, with just three set-screws, a single loose screw will cause a poor fit and possible tilt.  When I design a dovetail, I don't generally address the kinematic issues but I do design them with 6-screws.  That makes them much less susceptible to loosening during thermal cycles or vibration and much easier to snug flush.  Before assembly, make sure that all of the surfaces are as clean and damage free as possible and be extremely careful to check that everything is flush and that all three screws are snug after assembly.  If you can insert a "lock" set-screw on top of the three "holding" screws, that can help to prevent thermal or vibration induced loosening over time.

Eliminating tilt requires a bit of OCD to make sure that everything is correctly assembled and that eliminates the need for tilt adjusters.  As I've said, in most cases, sensor tilt is caused by tilt adjusters.  The exception is when you have a much faster optical system using a very large sensor--often in combination with a mechanical focuser that doesn't control tilt very well.

John

Bob, the 2" barrell I'm speaking of is the Starizona Corrector/ Reducer. It has a 2" barrell that inserts into a visual back and the other end is a 42mm male thread. There is no getting around the Starizona unit needing a visual back on the C8. 

Mike

Mike H - Sky View Observatory:
Bob, the 2" barrell I'm speaking of is the Starizona Corrector/ Reducer. It has a 2" barrell that inserts into a visual back and the other end is a 42mm male thread. There is no getting around the Starizona unit needing a visual back on the C8. 

Mike

Mike, which type of the Starizona Reducer/Corrector you have? can you post here in order so we can give further assistance?

That is correct. I was told I would get better centering and a better overall grip on the imaging train. When you crank down that click lock, is isn't going anywhere.

Your pictures would be great 👍🏻