Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Misaligned filters in the EFW?

  • Please log in to reply
63 replies to this topic

#26 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 03:54 AM

I prefer images on Abin....

 

There are some serious shifts with your filters. As I understand, you are using the 1.25" so these are screw on filters, so no way the filters are moving in the FW slots. 

Either there is something not moving properly with your wheel, or the FW is off axis somehow or your reducer is causing this.

Maybe try to remove the reducer and see what you get.

 

Are all these connections threaded? Or are you using thumb screws to tight something down? 


  • jonnybravo0311 likes this

#27 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 07:51 AM

I prefer images on Abin....

 

There are some serious shifts with your filters. As I understand, you are using the 1.25" so these are screw on filters, so no way the filters are moving in the FW slots. 

Either there is something not moving properly with your wheel, or the FW is off axis somehow or your reducer is causing this.

Maybe try to remove the reducer and see what you get.

 

Are all these connections threaded? Or are you using thumb screws to tight something down? 

Correct, these are 1.25" mounted (i.e. threaded) filters in the EFW. I pulled the reducer off yesterday and attached just the camera, EFW and a single spacer ring (so the 2" compression screws had something to lock down). Same result.

 

These connections are all threaded. The only thumb screws I had were when I pulled my reducer and attached the camera/EFW to the scope. With my "normal" imaging train of camera -> EFW -> extenders -> reducer -> scope it's all threaded.



#28 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 07:55 AM

Correct, these are 1.25" mounted (i.e. threaded) filters in the EFW. I pulled the reducer off yesterday and attached just the camera, EFW and a single spacer ring (so the 2" compression screws had something to lock down). Same result.

 

These connections are all threaded. The only thumb screws I had were when I pulled my reducer and attached the camera/EFW to the scope. With my "normal" imaging train of camera -> EFW -> extenders -> reducer -> scope it's all threaded.

Do something silly. Take out all the filters and place the EFW back in the imaging train and redo the flats. Just do one flat per, now and open slot. So you should have 8 flats. Put back the reducer first so everything is threaded.


  • jonnybravo0311 likes this

#29 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 07:57 AM

That was going to be my test this morning. Just need to get a coffee and breakfast in my system before I attempt handling tiny screws and delicate filters :D



#30 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 08:01 AM

Breakfast?! Pffff, amatuer..... fingertap.gif


  • jonnybravo0311 likes this

#31 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 10:02 AM

So... it's the filters, not the wheel. I took every filter out, put the imaging train back together and took flats. Every flat comes out exactly as you'd expect. Here's an example of one:

 

med_gallery_347158_16175_1814668.png

 

OK, root cause identified. There's really nothing I can do here, is there - other than contacting the vendor/manufacturer and getting a new set? I mean, it's not really possible I somehow inserted them incorrectly, is it? Just to ensure I'm not somehow screwing up threading a filter... here's what I did:

  1. Unscrew the 8 screws on the back cover of the EFW
  2. Looking at the exposed filter wheel, I see 8 positions numbered 1 - 8
  3. Into each position 1-7, I insert a filter
  4. I screw each filter clockwise until it meets resistance, then give just a touch more torque to tighten

That IS how to screw filters into the EFW, right?



#32 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 11:06 AM

So... it's the filters, not the wheel. I took every filter out, put the imaging train back together and took flats. Every flat comes out exactly as you'd expect. Here's an example of one:

 

med_gallery_347158_16175_1814668.png

 

OK, root cause identified. There's really nothing I can do here, is there - other than contacting the vendor/manufacturer and getting a new set? I mean, it's not really possible I somehow inserted them incorrectly, is it? Just to ensure I'm not somehow screwing up threading a filter... here's what I did:

  1. Unscrew the 8 screws on the back cover of the EFW
  2. Looking at the exposed filter wheel, I see 8 positions numbered 1 - 8
  3. Into each position 1-7, I insert a filter
  4. I screw each filter clockwise until it meets resistance, then give just a touch more torque to tighten

That IS how to screw filters into the EFW, right?

Well, I am not sure about root cause identified really.

The filters are barely the size to go with this camera but still should work. They work for me. I suggested to take them out to see if the is any vignetting regardeless. Well there might very well be some off axis alignment here but not noticeable without the filters in which is a good thing. If you had some APS-C size chip to test that would have been better but don't worry about that now.

 

Going back to the filters, what they show is not necessarily that they are vignetting, but rather that the is unven rotation of the wheel. Since they are screwed, their positions cannot shift. So, it is the wheel itself that cannot calibrate the positions well. This is way beyond the mechanical error tolerance I would say.

 

What are the filter cells thickness? from the edge of cell to beginning of threads? Do you screw them all the way until no thread left?

I believe if the cells are more that 8mm thick then you might run into a problem when the cells rub against the back cover of the FW. Do you hear any grinding noises while you rotate?



#33 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 11:15 AM

Well, I am not sure about root cause identified really.

The filters are barely the size to go with this camera but still should work. They work for me. I suggested to take them out to see if the is any vignetting regardeless. Well there might very well be some off axis alignment here but not noticeable without the filters in which is a good thing. If you had some APS-C size chip to test that would have been better but don't worry about that now.

 

Going back to the filters, what they show is not necessarily that they are vignetting, but rather that the is unven rotation of the wheel. Since they are screwed, their positions cannot shift. So, it is the wheel itself that cannot calibrate the positions well. This is way beyond the mechanical error tolerance I would say.

 

What are the filter cells thickness? from the edge of cell to beginning of threads? Do you screw them all the way until no thread left?

I believe if the cells are more that 8mm thick then you might run into a problem when the cells rub against the back cover of the FW. Do you hear any grinding noises while you rotate?

I don't have calipers, so I can't tell you the exact thickness of my filters. Yes, I screw them all the way in. No, there is no grinding/scraping noise at all. The filter wheel turns freely with the filters in it. I've tested this manually turning the filter wheel by hand and mechanically by running the calibration routine.

 

If I understand your first paragraph, you're stating that the EFW might be out of tolerance, which is why the edge of the filter housing is apparent in the image circle? OK, I can buy that theory. I'll reserve judgement on it being the filters for now... if I had another set of filters, I could compare and see. Unfortunately, I don't have a backup set lying around anywhere :D



#34 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 903
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 07 March 2021 - 02:10 PM

No calipers?  I thought everyone in the world had that exact same set of digital calipers...

 

So, you can see through the little circle that each filter screws into, but when filters are actually installed, the edges of the filter housing are cutting off the edges?

 

Which seems to be the case, as your most centered image showed cutoff in all 4 corners, correct?

 

To really prove that out, as you have no larger sensor dimensions available, I would probably take apart one of the filters and then just install the blank housing into the EFW.  But that's just me lol, and probably not necessary.

 

I looked up the WO 81, and it seems to say it is compatible with FF and APS-C sensors.  So, big image circle.

 

The question becomes, where are you cutting off the light cone, and what does the light cone look like?

 

I don't know a whole lot about light cones (except on the other end where I engineered a correcting mod on my 100ED), but I assume that somewhere it comes to a point and flips itself.  You have tried a FR already and that is not straightening out the light path sufficiently.

 

So, can you stick the EFW in a position where the light cone is thinner?  I mean, it's just a filter, and I presume flat across the entire diameter.  In the four scenarios you previously posted, you had the EFW right up against the astrocam.  If the light cone flipping point is somewhere to the front, maybe even inside the drawtube, then at your EFW location the cone is expanding once again.  No?  So, move the filter wheel forward?  Is there any reason the EFW can't be right up against the WO81, or in various other places in the train?


  • jonnybravo0311 likes this

#35 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 02:21 PM

The filter needs to be close to the sensor. As you move further away, you need larger filters to cover the image circle. See here: http://astronomy.too...ccd_filter_size. Put in a camera and scope, then change the distance from filter to sensor. As that gets larger, so too does the required minimum size of the filter.


  • Ken Sturrock and Mike in Rancho like this

#36 dswtan

dswtan

    Viking 1

  • *****
  • Posts: 622
  • Joined: 29 Oct 2006
  • Loc: Morgan Hill, CA

Posted 07 March 2021 - 02:24 PM

If we're still on the corner cut-off, I'm still in the "mechanical/optical tolerances" camp. I'm close to betting with imtl on this. ;-)

 

OP: have you shown yet whether it's the L filter *in any position in the EFW* that is the problem, or if it's *any filter that's in one bad position*. I.e., is it one filter with the problem, or one position in the wheel -- or multiple?

 

If it's just the L filter -- get another L. Ideally a bit bigger (31mm unmounted) to avoid this sort of issue. :-)

 

If it's the position in the wheel, or multiple positions -- get another wheel, or repair the wheel.

 

If it's multiple filters, get another wheel -- and to play safe, get new filters (again a bit bigger ideally).


Edited by dswtan, 07 March 2021 - 02:26 PM.

  • jonnybravo0311 likes this

#37 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 02:35 PM

WO stating the the scope is suitable for FF and APS-C is absolutely meaningless statement that all mass production manufacturers/retailers put on their website.
  • Ken Sturrock likes this

#38 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 02:36 PM

Jonny, put the filters back in a different order. i.e. L in another slot. Take a flat. Do you get the same result for L?
  • jonnybravo0311 likes this

#39 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 903
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 07 March 2021 - 02:37 PM

The filter needs to be close to the sensor. As you move further away, you need larger filters to cover the image circle. See here: http://astronomy.too...ccd_filter_size. Put in a camera and scope, then change the distance from filter to sensor. As that gets larger, so too does the required minimum size of the filter.

Ah, you are correct.  I must be thinking of diagonals or EPs.  I stuck my camera on the scope, and of course everything is straight up.

 

So, you'd actually need closer.  I imagine you are as close as you can get, with no thinner adapter of any kind?

 

And 2 inch filters are probably a bit more spendy, aren't they?



#40 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 02:39 PM

If we're still on the corner cut-off, I'm still in the "mechanical/tolerances" camp. I'm close to betting with imtl on this. ;-)

 

OP: have you shown yet whether it's the L filter *in any position in the EFW* that is the problem, or if it's *any filter that's in one bad position*. I.e., is it one filter with the problem, or one position in the wheel -- or multiple?

 

If it's just the L filter -- get another L. Ideally a bit bigger (31mm unmounted) to avoid this sort of issue. :-)

 

If it's the position in the wheel, or multiple positions -- get another wheel, or repair the wheel.

 

If it's multiple filters, get another wheel -- and to play safe, get new filters (again a bit bigger ideally).

I have not run the "put the L in every position and take a flat" test. Sounds like a good test, so thanks for that suggestion! Both the filters and the wheel are brand new - and the entire mono setup is completely new territory for me.

 

I did not go with the larger filters for a few reasons. First was cost. I was able to get a mid-to-high level set of 1.25" filters, whereas I would have had to drop down to a lower tier set had I gone for the 36mm, 50mm, etc. I also would have had to purchase the larger wheel - also adding to the cost. The second reason was that the 1.25" filters should be more than enough to cover my entire image circle. Finally, the 1.25" filters and the EFW were in stock.


  • Mike in Rancho and AstroVagabond like this

#41 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 903
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 07 March 2021 - 02:53 PM

Have you tried flipping the EFW around yet?  When I google it, I've seen both ways - bump out, or bump in.

 

In your post #21, you seem to indicate that the the side shown in the picture is the side attached to the camera.  I think - couldn't figure out the north and home stuff you mentioned.

 

But, the screws for the wheel are on the other side, right?  And that's what you open up to then screw the filters in.

 

Meaning the filters are actually closer to the side with the screws.  So...to get them closer to the sensor, could you try flipping around the EFW the other way?  Bump side away from the camera?

 

I may have embarrassed myself again, but...just brainstorming lol.



#42 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 02:56 PM

Mike, the side with the bump should be towards the camera and then the filters sit roughly 10mm from this sensor

#43 RogerM

RogerM

    Ranger 4

  • *****
  • Posts: 361
  • Joined: 13 Aug 2020
  • Loc: California

Posted 07 March 2021 - 04:16 PM

Mike, the side with the bump should be towards the camera and then the filters sit roughly 10mm from this sensor

On my EFW I measured from filter surface to outer case, 10mm camera/bump side and 12mm on the OTA side.

 

Edit:  And yes I accounted for that gap between the swab handle and the case wink.gif

Attached Thumbnails

  • Filter_to_housing_front.jpg
  • Filter_to_housing_rear.jpg

Edited by RogerM, 07 March 2021 - 04:17 PM.

  • Mike in Rancho likes this

#44 RogerM

RogerM

    Ranger 4

  • *****
  • Posts: 361
  • Joined: 13 Aug 2020
  • Loc: California

Posted 07 March 2021 - 04:40 PM

So Jonny had observed the EFW correctly indexing each filter position relative to the EFW's entrance/exit aperture. This by itself can be two fold causality--the rotating disk has each filter mount location correctly positioned (radially and equidistant along a circular path) and the EFW's electronics is correctly indexing each position. This also means that the indexing holes that the opto detectors used on the control board are properly located (same parameters as for each filter hole location) on the rotating disk as well.  One can possibly adjust the positioning of the electronics board relative to the case in order to 'shift' the position of the disk to the housing but as prior mentioned, the indexing appears to be correctly indexing each of the filters to the housing aperture.

 

So along the lines of having a properly configured EFW with mounted filter elements and making general assumptions about the rest of the optical train going forward towards the OTA, what about the orientation of the actual camera's sensor?  Could the sensor's clocking relative to the EFW have an impact here?  Maybe Jonny can add a thin shim to change the clocking between camera and EFW to see if that improves the L filter field?

 

Another thought about the filters removed and no visible vignetting image circle, offset or otherwise.  This proves that the larger (blank) aperture doesn't impinge upon the light path (we all knew that would be the case but its good to have that verified.)  In this instance I am curious about the sizing of each of the filter's thread-in holders, not thickness (diametral) per sey but I have noticed differences in the heights of the holders amongst a few lessor brand types.  Differences on the order of almost a full millimeter and then some.  With this height bringing the filter holder's 'aperture' that much more forward that it could potentially impinge upon the light path enough to produce this cut-off region that we're seeing?

Attached Thumbnails

  • Screen Shot 2021-03-07 at 1.25.15 PM.png
  • Filter_height.jpg

  • imtl and jonnybravo0311 like this

#45 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 07 March 2021 - 05:26 PM

I think Roger's suggestion might be worth while checking. Take the EFW + camera out and look under a light source how is the camera sensor oriented with the filters. Maybe you will see this cutoff. If so, you could try and add a very thin shim and change the orientation of the sensor compared to the EFW T2 hole. Even though the sensor should be enclosed fully, who knows if there is some weird mismatch there. Worth a check.

 

I would do that and also the changing the L filter between different slots in the FW to see if there is any difference.


  • dswtan, RogerM and jonnybravo0311 like this

#46 Mike in Rancho

Mike in Rancho

    Viking 1

  • -----
  • Posts: 903
  • Joined: 15 Oct 2020
  • Loc: Alta Loma, CA

Posted 07 March 2021 - 09:55 PM

On my EFW I measured from filter surface to outer case, 10mm camera/bump side and 12mm on the OTA side.

 

Edit:  And yes I accounted for that gap between the swab handle and the case wink.gif

Thanks for those measurements and photos.  I am learning a lot in this thread in case I ever step up to this equipment.  At Jonny's expense.  Sorry Jonny!

 

I wasn't so much worried about the surface of the glass in the filter, more of the filter frame that sets the aperture that he needs to get through, preferably cleanly.  But 2mm seems too much to make up since the rack is that much closer to the bump side, even if the filter isn't centered in its frame and threads.


Edited by Mike in Rancho, 07 March 2021 - 09:56 PM.

  • RogerM and jonnybravo0311 like this

#47 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 07 March 2021 - 10:44 PM

Thanks for all of the help everyone! I decided to put the tests on hold until tomorrow so I could actually image tonight. I can deal with cropping and work on the tests during the daylight hours. Before I set my rig up for running tonight, I did run the L in a few different positions. I tried it in position 8, 1, 2, 3. The results were as follows:

 

In position 8:

 

med_gallery_347158_16175_2914384.png

 

In position 1:

 

med_gallery_347158_16175_2015224.png

 

In position 2:

 

med_gallery_347158_16175_6276036.png

 

In position 3:

 

med_gallery_347158_16175_4626966.png

 

Hope that helps provide some insight.


  • dswtan likes this

#48 imtl

imtl

    Gemini

  • *****
  • Posts: 3,491
  • Joined: 07 Jun 2016
  • Loc: Down in a hole

Posted 08 March 2021 - 01:17 AM

Okay. It's the FW being uneven in rotation positioning. Might be tolerance, might be mechanical defect.

I would contact ZWO with those set of 4 pictures above explaining this and ask for their advice. They answer pretty quickly. Since you just bought the FW you are under warranty.

 

For now, place the L in the best possible place (3 but you haven't tried all of them) and then check out the rest of the filters, one by one, with changing their positioning in different slots until you get best results for each filter. Going to be a bit tedious but that is what it is for now. 

 

When you upscale your imaging system, compexity grows and problems arise. But you grow as well when you investigate and learn. Problem solving in this hobby is contantly happening. With time you gain experience and will know faster and more efficient how to approach things and solve them.

 

Let us know how this works out and how the imaging session tonight worked.

And show some pictures besides flats! I have never spent so many hours staring at flats in my life! Well, actually it's my fault, I need to break from CN.... :p


  • dswtan, RogerM and jonnybravo0311 like this

#49 jonnybravo0311

jonnybravo0311

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1,719
  • Joined: 05 Nov 2020
  • Loc: NJ, US

Posted 08 March 2021 - 09:34 AM

Okay. It's the FW being uneven in rotation positioning. Might be tolerance, might be mechanical defect.

I would contact ZWO with those set of 4 pictures above explaining this and ask for their advice. They answer pretty quickly. Since you just bought the FW you are under warranty.

 

For now, place the L in the best possible place (3 but you haven't tried all of them) and then check out the rest of the filters, one by one, with changing their positioning in different slots until you get best results for each filter. Going to be a bit tedious but that is what it is for now. 

 

When you upscale your imaging system, compexity grows and problems arise. But you grow as well when you investigate and learn. Problem solving in this hobby is contantly happening. With time you gain experience and will know faster and more efficient how to approach things and solve them.

 

Let us know how this works out and how the imaging session tonight worked.

And show some pictures besides flats! I have never spent so many hours staring at flats in my life! Well, actually it's my fault, I need to break from CN.... tongue2.gif

Thanks for the confirmation on what the issue is. I'll ping ZWO to get their input.

 

Since you are jonesing for something other than a flat...

 

med_gallery_347158_16159_5479343.jpg

 

Straight HOO from the data I've collected so far.


  • dswtan, moonrider, 42itous1 and 5 others like this

#50 LYAstro

LYAstro

    Vendor - Antlia Filters

  • -----
  • Vendors
  • Posts: 158
  • Joined: 16 Oct 2018
  • Loc: China

Posted 10 March 2021 - 10:05 AM

This situation looks like an occlusion. One possible reason is that the filter wheel is not centered in its rotation. The another reason is that using the 1.25 inch filter is at the very limits of providing proper coverage of the large ASI294mm sensor. Due to the combination of a focal reducer,

 

the Antlia 1.25" filter cannot be completely covered. 1.25 inch  filter's effective aperture > 26mm, filter thickness is 2mm.


  • jonnybravo0311 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics