Chip tilt test jig.

[tooth_dr]

I did both my cameras earlier and the ZWO was significantly better than the QHY before adjusting.  It was clear tonight and a quick test shows good results, and finally I have decent enough stars at F2.8

[ollypenrice]

On 08/09/2021 at 20:49, wimvb said:

A bit of reflection on the subject (poor pun) leads me to the conclusion that the reflection spot on the screen closest to the incident laser beam is from the surface closest to the laser. The reflection furthest from the incident laser beam is from the deepest reflecting surface (= sensor). Unless there are multiple reflections, ie the beam reflecting back and forth between two surfaces. And this is something which I hope, will never occur.

A late reply but I wonder if this may only apply to a well collimated camera. Perhaps a significant tilt on the chip would send the chip-cover reflection back closer to the source laser when rotated to a certain orientation?

Olly

[wimvb]

@ollypenrice: yes, of course.

[GalaxyGael]

My jig is a little smaller but robust and cobbled together with some of the spare pieces from my optical bench. I have a tiltable prism to more precisely I put the laser beam but with a fixed laser position it works great. 

[Corpze]

Nice to se that more people have found this amazingly easy design ( and the results )
I have been imaging with my ZWO ASI 2600MM now for a couple of weeks and the stars are perfect in all corners, unlike with my Atik 460 which I tried to adjust during the imaging at night - to no sucess.

This rig got it right the first try ( but the ZWO was very good straight out of the box )

I made a new video regarding the items i use, and adress some questions I have got on my Youtube channel, more specific - what to use as a "barrel" for the nosepieces to the cameras - but as you guys say - a hole in a wooden plank and two supports seems good enough.

I have also read some questions about how to make sure everything is parallel - and to be honest - i haven't made sure. But, I use aluminium rig components which is quite parallel if mounted properly.
Not sure it needs to be parallel though. 

Anyway - here is the video:
 

 

[wimvb]

1 hour ago, Corpze said:

Not sure it needs to be parallel though. 

the sleeve you use to hold the camera needs to be reasoably flat, so you don’t get any wobble there. Having the camera on a flat surface pointing down may actually help with that because you eliminate any distortions due to gravity. Fibre board from furniture is quite flat and won’t change over time, so works just fine.

[tooth_dr]

@Corpze sorry I didnt thank you sooner - thank you for sharing this with the wider community.  And thank you to @ollypenrice for creating a DIY version for us all to emulate - the idea of being to test the camera for tilt on a jig in the house has been a game changer for me.  I just didnt have the capacity or incentive to do this on the fly at night using software and measuring FMWH etc.  Although my stars arent totallyperfect yet, they are at a very acceptable level for me, and I'm starting to enjoy my Epsilon after 2 years of ownership.

[Corpze]

On 29/09/2021 at 11:17, tooth_dr said:

@Corpze sorry I didnt thank you sooner - thank you for sharing this with the wider community.  And thank you to @ollypenrice for creating a DIY version for us all to emulate - the idea of being to test the camera for tilt on a jig in the house has been a game changer for me.  I just didnt have the capacity or incentive to do this on the fly at night using software and measuring FMWH etc.  Although my stars arent totallyperfect yet, they are at a very acceptable level for me, and I'm starting to enjoy my Epsilon after 2 years of ownership.

This is the very reason i do these videos - to help out fellow astronomers  
And the reward is you saying they are helpful - entirely my pleasure!

/Daniel

[GalaxyGael]

V2. 0, cobbled with a bucket of Thorlabs parts from my disassembled optical bench. 

Rotation stage house the camera, even a dovetail makes an appearance although imperial threading on sharpstar vs metric threading for everything else is a perennial peeve. The right angle prism sends the beam from the laser to the chip. 

Works wonderfully thankfully. 

Now I know star issues on the fast scope are the focuser and that is not straightforward. Or I could attach the camera to the f/5 quad Apo and enjoy the next clear night 😜. Decisions.... 

The Thorlabs pieces are also being used to form an autocollimator, much like those used for ultra precise lathing, angle measurements etc. Might be possible to replace the retro reflecting mirror part with the scope. New thread on that if it works. 

[gorann]

I am sure it works but that thing looks like something at least I could accidentally tip over🥴

[GalaxyGael]

17 minutes ago, gorann said:

I am sure it works but that thing looks like something at least I could accidentally tip over🥴

Ha. You would need to swing a tennis racquet at it. The red base is 3kg🎾. And the steel bar is 1kg etc. Quite rigid. The asi2600 is like a feather on it as it is bolted to a Losmandy. 

[Starflyer]

Late to the party here but thinking of building a jig to test my cameras.  Does it matter if a sensor isn't central to the housing, how does this affect the pattern when rotating?

I have a ZWO camera that is a country mile out of being central.  I was tinkering with it for secondary collimation using the circles in sharpcap, on rotating the camera the circles shifted against the secondary by a good few mm.

[symmetal]

5 hours ago, Starflyer said:

Late to the party here but thinking of building a jig to test my cameras.  Does it matter if a sensor isn't central to the housing, how does this affect the pattern when rotating?

I have a ZWO camera that is a country mile out of being central.  I was tinkering with it for secondary collimation using the circles in sharpcap, on rotating the camera the circles shifted against the secondary by a good few mm.

No it shouldn't matter. The pattern caused by the pixels may well rotate about a different axis to the camera body, but you're only interested in the reflection from the cover slip on the sensor front. The cover moving side to side while rotating won't matter as it just reflects off a different part of the cover. As long as the cover is orthogonal, the visual reflection from the cover won't move during rotation, which is all you're checking for.

Alan

[ollypenrice]

On 17/10/2021 at 20:46, symmetal said:

No it shouldn't matter. The pattern caused by the pixels may well rotate about a different axis to the camera body, but you're only interested in the reflection from the cover slip on the sensor front. The cover moving side to side while rotating won't matter as it just reflects off a different part of the cover. As long as the cover is orthogonal, the visual reflection from the cover won't move during rotation, which is all you're checking for.

Alan

I agree. Assuming the glass is flat, any part of the chip cover will do to send a reflection down to the screen. It doesn't have to be the same spot on the glass. If the glass isn't flat you're doomed anyway!

Olly

[Laurin Dave]

This topic has recently been discussed on the SX forum as a number of folk on there have been building their own jigs..  Terry Platts advice was 

..... Generally speaking, you need to look for the bright spots that often are not aligned with the regular grid pattern... These offset spots are from the glass windows – the front camera window and the glass cover slip on the CCD package. As the cover slip is usually very nearly parallel to the CCD chip, it will give a bright spot, which is very close to one of the diffraction spots, so locate this and then align the front plate so that the diffraction spot closest to the cover slip spot doesn’t move when you rotate the camera.

I'll remember this next time I do it..

Dave

[gorann]

For those that experience tilt problems, I just found these and ordered some (they ship to Europe for about 4 dollars). Should work with my RASA8 but I may have to cut off much of the handle since it will probably mess with the stars by sticking into the light path.

CS, Göran

https://astrodymium.com/collections/frontpage/products/rotating-tilt-adjusters

[tomato]

Thanks for this, I might need some for my Risingcam IMX571, when it finally arrives…

[ollypenrice]

22 hours ago, gorann said:

For those that experience tilt problems, I just found these and ordered some (they ship to Europe for about 4 dollars). Should work with my RASA8 but I may have to cut off much of the handle since it will probably mess with the stars by sticking into the light path.

CS, Göran

https://astrodymium.com/collections/frontpage/products/rotating-tilt-adjusters

That's a crafty idea.

Olly

[tooth_dr]

Looking some advice, I noticed tilt on an image taken with my QHY268M and when checked it with the rig it was quite a bit out, with the reflection of the laser pen tracing a large 10-15mm diameter circle.  I've tried adjusting it but it's still out.  In fact it hasnt really improved and I had the faceplate 1mm off at one side.  This sounds like a mechanical issue with a spacer or adapter?

[symmetal]

Can you post an image of it set up in the jig.  Are the filter wheel and/or adapters still attached? On another thread you had to buy some screws to fit in the QHY tilt adapter plate to make it function. If the tilt plate has a 1mm gap on just one side that should have made some difference. Did you loosen off the tilt screws on the opposite side a little before making the 1mm tilt adjustment and then retightening them. Depending on how stiff the tilt plate and the camera front are, adjusting just one set without loosening off the others, may just end up distorting the tilt plate/camera connection rather than giving just tilt.

Alan  

[tooth_dr]

24 minutes ago, symmetal said:

Can you post an image of it set up in the jig.  Are the filter wheel and/or adapters still attached? On another thread you had to buy some screws to fit in the QHY tilt adapter plate to make it function. If the tilt plate has a 1mm gap on just one side that should have made some difference. Did you loosen off the tilt screws on the opposite side a little before making the 1mm tilt adjustment and then retightening them. Depending on how stiff the tilt plate and the camera front are, adjusting just one set without loosening off the others, may just end up distorting the tilt plate/camera connection rather than giving just tilt.

Alan  

I’m working late, but will post a photo later of it on the rig.  I have as much of my imaging train removed in one piece, so right down to and including the base plate on the end of the focuser tube (circled blue).  I keep the filter wheel and adapters attached.

 

I loosened all the screws enough to allow movement of the plate.

 

[tooth_dr]

@symmetal I’ve attached a photo below.
 

Last night I checked the 2600 for tilt. When rotated in the holder the laser pointer dot remaining stationary.  I then swapped over the focuser base plates.  On retesting the 2600, the laser dot drew a large circle when the camera was rotated.  I swapped back the base plate to the original one and the laser dot once again remained stationary. This leads me to assume the base plate is the issue!

The base plate has an M48 male thread in it, it looks glued in.  I compared this to the other good base plate, and the other M48 part sits perfectly flat  

Here is a close up of what the tilted one looks like.  Unless APM has set it up like this on purpose to match the scope it, but that seems unlikely.

Where do I go from here to sort this out?  Would Goräns suggestion above work?

 

 

Edited December 3, 2021 by tooth_dr

[symmetal]

I'm not sure where the rotation occurs when you rotate the camera to change image orientation when it's mounted on the scope. If it's between the base plate and the camera then it's a problem, as tilt will change when you change camera orientation.

If the image train is fixed as you have it in the jig, then you should be able to offset the camera tilt to counteract the base plate tilt. It will act as in the earlier question where the sensor was not in the centre of the camera body. When the camera tilt is correctly set, all the laser dots will rotate in a circle apart from the reflection off the sensor cover slip which should be stationary. This should be the brightest reflected spot though may not be so easy to determine.

Doing this of course means that the image train must all be kept together in one piece so cameras cant be swapped over without tilt reappearing. This isn't ideal so it's best to correct the obvious tilt source as close to that source as possible.  Using one of the tilt rings as gorann posted fitted over the base plate thread in the correct orientation should help, assuming there is enough 'play' in the thread when loosened to allow it to achieve this degree of tilt. Check and adjust the camera on its own in the jig for no tilt, and then try the whole train in the jig with one of the tilt rings next to the base plate in the correct orientation, to see if you can again achieve no tilt indicated on the jig.

The alternative if the above doesn't work, is to fit another tilt adapter to the base plate and use that to correct the base plate tilt. As I've mentioned before, the Gerd Neumann tilt adjusters are very easy to adjust, though a little pricey. FLO don't stock them though.

Alan

[tooth_dr]

3 minutes ago, symmetal said:

I'm not sure where the rotation occurs when you rotate the camera to change image orientation when it's mounted on the scope. If it's between the base plate and the camera then it's a problem, as tilt will change when you change camera orientation.

If the image train is fixed as you have it in the jig, then you should be able to offset the camera tilt to counteract the base plate tilt. It will act as in the earlier question where the sensor was not in the centre of the camera body. When the camera tilt is correctly set, all the laser dots will rotate in a circle apart from the reflection off the sensor cover slip which should be stationary. This should be the brightest reflected spot though may not be so easy to determine.

Doing this of course means that the image train must all be kept together in one piece so cameras cant be swapped over without tilt reappearing. This isn't ideal so it's best to correct the obvious tilt source as close to that source as possible.  Using one of the tilt rings as gorann posted fitted over the base plate thread in the correct orientation should help, assuming there is enough 'play' in the thread when loosened to allow it to achieve this degree of tilt. Check and adjust the camera on its own in the jig for no tilt, and then try the whole train in the jig with one of the tilt rings next to the base plate in the correct orientation, to see if you can again achieve no tilt indicated on the jig.

The alternative if the above doesn't work, is to fit another tilt adapter to the base plate and use that to correct the base plate tilt. As I've mentioned before, the Gerd Neumann tilt adjusters are very easy to adjust, though a little pricey. FLO don't stock them though.

Alan

Thanks Alan.  I would hope once this is correctly adjusted, I wont be removing or touching the camera again.  I think there is some play in the threads, but I dont think enough to remove all the tilt.

The only other issue that I have now thought about is the corrector is threaded via M63 male thread on the reverse of this M48 thread.  How big an issue this is I dont know, but the corrector will be tilted relative to the image plane.

 

[GalaxyGael]

The tilt adapters Goran suggested might be the best option if they fit, and keeps your 2600 at least formally tilt free for any other use and for trouble shooting down the line. Based on your pictures, that base looks to be the source. tilt offsetting the 2600 would work if when you screw it on, one of the tilt screw is juts at the right place making the adjustment straightforward. Otherwise you might have a three-legged stool problem trying to raise the tilt plate in one particular place using three screws in different orientations.

An alternative and permanent workaround could be a very small 1-2 mm optically thick adapter with a sanded side that screws onto and compensates the higher side of the baseplate, onto which everything else is screwed. Not ideal of course. What I did with a Vixen VC200L with a similar tilt problem from a mal-threaded focuser attachment was to glue a thin metal shim to compensate. I used a segment of an aluminium 0.3 mm space ring to do that.