143 replies to this topic

[Jinux]

I designed 3D printer model to use Sony E-mount lens 35mm F/1.8 on ASI533MC to take picture of Bernard's loop. FOV is just right.

But found that the lens doesn't allow me to manually focus once it's detached from Alpha camera. So I wasn't be able to actually test it with star, while it seems to be able to reach focus in some distance. 

It could be useful if somebody has a lens that allows manual focus without camera body. So posting here.

 

https://www.thingive...m/thing:4655644

 

If anybody interested in to be a beta tester of it, please print it and give me a feedback if any modification is needed.

Please read note on thingiverse description for printing instruction and tightness.

 

Hope this helps!

 

Attached Thumbnails

• 

Edited by Jinux, 16 November 2020 - 01:03 PM.

[Linwood]

I'd be very interested in how this works out.  I have a big Sony telephoto I'd like to mount an ASI camera to. 

 

Is there space (I realize there's very little space) to split the adapter in half, and have a threaded connection that you could screw and unscrew to focus.  Then set the lens to infinity in the camera (will it stick?) and then you can fine tune on the mount by screwing?   There's no real space for a real focuser in there of course.

 

The other alternative, which is not pretty, is to wire power and signals to allow the focus to work. I have no idea if that would even be possible to figure out, much less how you can print it.  But it sure would be handy!

[Jinux]

I'd be very interested in how this works out.  I have a big Sony telephoto I'd like to mount an ASI camera to. 

 

Is there space (I realize there's very little space) to split the adapter in half, and have a threaded connection that you could screw and unscrew to focus.  Then set the lens to infinity in the camera (will it stick?) and then you can fine tune on the mount by screwing?   There's no real space for a real focuser in there of course.

 

The other alternative, which is not pretty, is to wire power and signals to allow the focus to work. I have no idea if that would even be possible to figure out, much less how you can print it.  But it sure would be handy!

Thread for focusing is an interesting idea. 

Total length of this adapter is only 11.5mm where top 5mm is used by bayonet and bottom 4mm is by M42/0.75. So there is only 2.5mm left.

But if I use outer part of ring to create thread and design a way of locking mechanism, I might be able to make a focus using it.

Let me think about it.

[Jinux]

BTW, I created another model for small ASI camera that has T-thread in front.

 

https://www.thingive...m/thing:4664592

 

More than 30 folks downloaded but didn't get any feedback whether it's working or not. Interesting.

I sent these to one of CN'er who promised to give me feedback. I'll report later.

 

Attached Thumbnails

• 

[Linwood]

Did you have any luck with alternatives for focus where the lens is electronic only?

[Jinux]

Unfortunately, no.

Thinking about the idea and reality, it was long shot.

Let's assume, focus is kept when lens is detached from camera. Now my adapter should make exactly parfocal from lens to sensor of Sony camera and lens to ASI sensor. Even I design it to exact 18mm spec, I guess there is still a room to wiggle to make critical focus.

Design it as exact 18mm is easy but printing it as exactly 18mm would be tough as plastic shrinks after cooldown a bit.

[Linwood]

Yeah, I mean like threaded adjustable focus.  But I think that's even a more long shot, I have no idea how far it would have to move.

[cucubits]

That doesn't look like it's big enough to get correct backfocus.

 

I've also designed adapters for the ASI planetary cameras and also for the DSO cooled ones. These are for Canon lenses but I can upload the files for anyone to download and measure exact distance from flange to lens and use them to design for other mounts. If I recall, there is 5mm difference in the adapter between the planetary cameras and DSO cooled cameras, the same adapter won't work with both.

 

I will upload tomorrow and post details.

 

 

 

This is example of the one for ASI DSO cooled cameras:

 

Edited by cucubits, 02 December 2020 - 02:38 AM.

[Linwood]

Footnote for those following along at home who might not be into regular photography... mirrorless cameras generally have a much shorter backfocus distance (aka flange depth) than DSLR's.   For example, Nikon's regular DSLR is 46.5mm but their current mirrorless is only 16mm.  Canon EF is 44mm but RF is 20mm.

[Jinux]

That doesn't look like it's big enough to get correct backfocus.

 

I've also designed adapters for the ASI planetary cameras and also for the DSO cooled ones. These are for Canon lenses but I can upload the files for anyone to download and measure exact distance from flange to lens and use them to design for other mounts. If I recall, there is 5mm difference in the adapter between the planetary cameras and DSO cooled cameras, the same adapter won't work with both.

 

I will upload tomorrow and post details.

 

lens.jpg

 

lens2.jpg

 

This is example of the one for ASI DSO cooled cameras:

 

Dso.jpg

Canon EF mount has flange to focus distance of 44mm, so the adapter would be 26mm longer than this. 

You can find all flange to focal distances from Wikipedia like below.

 

https://en.wikipedia..._focal_distance

 

BTW, is your adapter designed with OpenSCAD? I'm trying to collect/create all (or common) mount design files and create a parametric generator for them. 

[cucubits]

Canon EF mount has flange to focus distance of 44mm, so the adapter would be 26mm longer than this. 

You can find all flange to focal distances from Wikipedia like below.

 

https://en.wikipedia..._focal_distance

 

BTW, is your adapter designed with OpenSCAD? I'm trying to collect/create all (or common) mount design files and create a parametric generator for them. 

Fusion 360. The one for the ASI 533MC camera I designed from scratch and the one for the planetary camera, I used parts of an existing design from Thingiverse. Can't find it now though. I can upload both sources but I think they wouldn't work for everyone. I had to make minor resize tweaks to have the threads work with my particular printer/material. 

[erictheastrojunkie]

So I'm the one that Jinux mailed a few copies of the adapter to to test, I'm using a Voigtlander 50mm f2 APO lens that I've wanted to try with my ASI533mc to see how it would perform for ultra-wide field stuff (I'm planning a big Milky Way mosaic for next year). I had the 533mc on my telescope last night so I put the ASI183mm I had on the adapter with the Voigtlander lens, first impressions were that the lens and camera fit very snug on the adapter, no movement, a nice tight fit and very lightweight (total weight of the lens, camera, adapter, and lens collar were ~2 pounds). With a few strips of tape I got this lens collar to wrap nice and tightly around the adapter:

 

https://www.amazon.c...0?ie=UTF8&psc=1

 

I was having problems with my table that I'm using to control my portable mount, a Sky Watcher AZ-GTi, I honestly think it was just too cold and the tablet kept freezing up and shutting down. I did managed to grab a few test shots, I had the lens at f2.5 (I originally intended to do some shots at f2 through f2.8, but the tablet problems and lack of feeling in my fingers prevented that), focusing was relatively easy even though the lens is a macro lens and has a very short focus throw around the infinity area. I had no problem reaching infinity with a little leeway beyond it, which is a good thing for temperature variations. Here is a single, 60 second exposure I did with the 183mm at gain 0/offset 11 (just going to post the link to the full resolution image):

 

https://i.imgur.com/GTSxaCh.jpg

 

No image corrections were made, just opened in pixinsight, did a soft stretch and saved as jpg. I ran the image through PI's FWHMeccentricity script, the FWHM contour plot shows some slight tilt:

 

The aberration inspector mosaic shows some mediocre stars in the corners:

 

Jinux sent me a second adapter to try made from a different plastic, I'm going to throw that on and actually use my 533mc to do some testing tonight if it stays clear. The tiny pixels of the 183mm are more revealing of lens tilt imperfections, I've used this lens on full frame with better results so this wasn't the best result, but I think overall it seemed to work ok. Given that the stars are ever so slightly radiating out away from the middle I'd say the adapter needs to be slightly thicker, I measured it at ~11.8mm with my digital calipers and I wouldn't be surprised if that missing 0.2mm is contributing slightly to the star deformation. As Jinux mentioned, it's difficult to account for that as the plastic shrinks when it cools so it'll probably be a little bit of trial and error to get the actual 12mm thickness after being made/cooled. I'll post back more results when I test the 533mc and this lens. 

Edited by erictheastrojunkie, 08 December 2020 - 10:17 AM.

[Jinux]

Thanks for great review! Even if my lens allowed me to, the only test I would have performed be focus to star test! 

 

I guess the tilt is coming from bayonet mount. It's snug fit but I can see it's squeezing in on bottom opening of bayonet with lens, so bayonet and lens are not exactly tight fitting. I'll update design to correct it. Probably make bottom opening of bayonet a bit larger and increase wedge angle a bit more to make bayonet connection stronger.

In general Z-precision of 3D printer is pretty good, but if my print bed leveling is off, then the print would have came out tilted regardless of the fitting. I think it's quite level but considering the sensitivity of astro stuff, I'll double check on it.

 

Outer radius doesn't have specific requirement and I can increase it to fit in the lens holder, would 1mm in diameter be good enough? Or would you measure ID of the holder and let me know?

 

Confusing thing is, E-mount flange to focus spec is 18mm where camera consumes 6.5mm out of it. So adapter should be 11.5mm where designed as 11.8mm by mistake. So to meet the right spec, it should be 'thinner' while your think it should be thicker. I'll make both and send it to you.

 

Again, thanks for the feedback!

 

-Jinux

[erictheastrojunkie]

No problem, thanks for the adapter(s).

 

The tilt from the bayonet mount is certainly possible and makes sense, I suppose it could also come from the print bed of the printer not being perfectly level, honestly there's a TON of things that could cause which you may not be able to control, but I'll certainly try out anything you can think of. It's nearly impossible to correct all sources of tilt when you are imaging at ~f2-f2.8 apertures. I'd be ok with the current adapter when I factor into overlap between frames I'll be doing for the mosaics I plan. 

 

The inner diameter of the lens collar is roughly 65mm (it's a cheap plastic one so it is not perfectly round inside, it ranges from 65-66mm in some places), the only thing that is a challenge is the tiny little "button" on the edge, I just filed it off to make the outside of the adapter perfectly round without it then put some tape around the outside of the adapter which allowed for a nice tight fit in the collar. 

 

I guess it could be slightly too thick, to my eyes it looks like the stars are radiating out from the middle, but perhaps that's from the slight tilt. I can measure the adapters in different places and see what the actual thickness is, maybe it varies ever so slightly in different spots. 

 

Thanks for making these, they definitely open more possibilities. The weather is looking absolutely awful for the foreseeable future so I'm not sure when I'll be able to test again after tonight, maybe end of the month if I'm lucky. 

[Jinux]

Critical focus zone for F2.0 optics are too thin, it's hard to make 'precision' adapter without tilt. My bed level was 'reasonably' flat but I haven't tested the tolerance yet in that level and I'm not sure whether I can achieve it. I'll measure with digital caliper after printing.

Layer thickness was 100um on last printing and I can reduce this to 65um at the cost of 2X printing time.

 

The protruding button is for align marker and I can make it as notch to keep it while not preventing from mounting on round collar. No problem.

 

Thickness is still puzzling but I can print several thickness version of it. Plastic is cheap for print. It's printing time that matters. I guess electricity for keeping the printer cost more than plastic for this small but precision printing thing. And shipping cost totally ruins it.

[bemo47]

Thanks for great review! Even if my lens allowed me to, the only test I would have performed be focus to star test! 

 

I guess the tilt is coming from bayonet mount. It's snug fit but I can see it's squeezing in on bottom opening of bayonet with lens, so bayonet and lens are not exactly tight fitting. I'll update design to correct it. Probably make bottom opening of bayonet a bit larger and increase wedge angle a bit more to make bayonet connection stronger.

In general Z-precision of 3D printer is pretty good, but if my print bed leveling is off, then the print would have came out tilted regardless of the fitting. I think it's quite level but considering the sensitivity of astro stuff, I'll double check on it.

 

Outer radius doesn't have specific requirement and I can increase it to fit in the lens holder, would 1mm in diameter be good enough? Or would you measure ID of the holder and let me know?

 

Confusing thing is, E-mount flange to focus spec is 18mm where camera consumes 6.5mm out of it. So adapter should be 11.5mm where designed as 11.8mm by mistake. So to meet the right spec, it should be 'thinner' while your think it should be thicker. I'll make both and send it to you.

 

Again, thanks for the feedback!

 

-Jinux

i am currently printing the "merongi" model, the goal is to attach it to a 1600MC Pro and a FishEye Meike 6.5mm f2.0 in order to build an AllSky, so thanks a lot for these designs and all tests. I will keep u aware of my results.

 

Just a remark about BF : for these cooled asi cams the backfocus as defined by ZWO is 6.5+11 = 17.5mm

Attached Thumbnails

• 

[AstroCatinfo]

Thanks for sharing this models.

 

I was wondering if there would be any space for a filter there? My setup would be as follows:

 

ASI2600MC + L Extreme Optlong filter + Sigma Art 50mm.

 

Any clues on how to attach it?

 

Thanks so muc,

 

Aleix

[erictheastrojunkie]

Thanks for sharing this models.

 

I was wondering if there would be any space for a filter there? My setup would be as follows:

 

ASI2600MC + L Extreme Optlong filter + Sigma Art 50mm.

 

Any clues on how to attach it?

 

Thanks so muc,

 

Aleix

Nope, there's no way to get a filter in there, just not enough back focus space. With the 2600mc the tilt adapter is going to have to be removed and a replacement adapter with the e-mount built in used in its place, there's BARELY enough space for a replacement adapter with the built in e-mount lens mount. I've been working with him to have one made up and tested with my 2600mc, tilt is almost certainly going to be an issue. 

[Jinux]

Thanks for sharing this models.

 

I was wondering if there would be any space for a filter there? My setup would be as follows:

 

ASI2600MC + L Extreme Optlong filter + Sigma Art 50mm.

 

Any clues on how to attach it?

 

Thanks so muc,

 

Aleix

Actually there is EASY solution. Put filter "in front of" lens like we used to put UV-IR or Sky filter for lens. 

I'm not sure how typical star halo, the signature of L-Extreme filter, changes in this configuration though. 

How to attach filter to front of camera? Design and 3D print a ring that has two threads of 48mm/0.75 female thread and the male thread that screws into lens front, which I just looked up and seems like 77mm. I guess it would use 0.75 pitch. About 30min job for design and a couple of hours to print. 

50mm/F1.4 lens has clear aperture of 35.7mm in theory, so 48mm filter in front may not cause light path obstruction. Or in worst case use a bit higher F to step down.

 

-Jinux

 

ps. Due to some 'urgent' projects from work and home requested by CFO, I wasn't able to work on this even the idea is there. Both projects are about to finish. Will be back SOON!

[AstroCatinfo]

Actually there is EASY solution. Put filter "in front of" lens like we used to put UV-IR or Sky filter for lens. 

I'm not sure how typical star halo, the signature of L-Extreme filter, changes in this configuration though. 

How to attach filter to front of camera? Design and 3D print a ring that has two threads of 48mm/0.75 female thread and the male thread that screws into lens front, which I just looked up and seems like 77mm. I guess it would use 0.75 pitch. About 30min job for design and a couple of hours to print. 

50mm/F1.4 lens has clear aperture of 35.7mm in theory, so 48mm filter in front may not cause light path obstruction. Or in worst case use a bit higher F to step down.

 

-Jinux

 

ps. Due to some 'urgent' projects from work and home requested by CFO, I wasn't able to work on this even the idea is there. Both projects are about to finish. Will be back SOON!

 Ops! A better option could somehow be the possibility to make space for it between the ASI2600MC and the Sigma 50mm... therefore I could use the solution for other lenses. I do not own a printer myself, so I asked a friend to print a solution for me. When I get it I'll try to figure out how to do it.

 

IMO my case it's not an strange situation: use Sony lenses with ZWO cameras... the problem comes when I want to add a filter :/

 

Thanks so much for your answers!!

Edited by AstroCatinfo, 20 January 2021 - 02:06 PM.

[erictheastrojunkie]

 Ops! A better option could somehow be the possibility to make space for it between the ASI2600MC and the Sigma 50mm... therefore I could use the solution for other lenses. I do not own a printer myself, so I asked a friend to print a solution for me. When I get it I'll try to figure out how to do it.

 

IMO my case it's not an strange situation: use Sony lenses with ZWO cameras... the problem comes when I want to add a filter :/

 

Thanks so much for your answers!!

The thickness of the needed adapter between the 2600mc and the sony lens is roughly 5.5-5.8mm in order to fit within the back focus distance, the depth of the lens bayonet itself is nearly all of that thickness, I don't see how you could incorporate a 1-2mm thick piece of glass in the 3D printed adapter (you'd have to pull the glass from its holder, you couldn't get a mounted filter to fit). You already have to remove the metal tilt adapter (which is 5mm thick) on the 2600mc in order to attach your own 3D printed adapter in order to meet the back focus requirements. 

[AstroCatinfo]

Thanks Eric for the raw numbers. Clearly it won’t fit... :/

[Jinux]

I have a thought. I'm writing loud.

By the famous lens equation

 

1/o + 1/i = 1/f, where o = object distance, i = image distance, f = focal length

 

In camera lens system with given focal length of f, the only way to make a focus to specific object distance of o, is to move lens element distance relative to sensor and make change of i. So, amount of allowable lens movement would define minimum focus distance. 

So, say, there is a 50mm prime lens with minimum focal distance of 0.38m ( Sony SEL 50mm F1.8, for example ), image distance at minimum focal distance would be

 

i = of / (o + f), (o is typically negative, so use positive numbers when move to right term for simplicity) = 0.05 * 0.38 / (0.05 + 0.38 ) = 0.044m

 

and o at infinity, i = f = 0.050m, i.e. there should be about 6mm lens movement. 

 

of course this is 'simple lens' equation and camera lens is not but the idea is that lens has to move by in the order of 6mm. 

 

So, what if, we design an adapter that has 6mm longer than E-mount's 18mm backfocus requirement? 

Now, if my thought is correct and simple lens equation applies to the same way to camera lens, the lens will focus at infinity when lens is moved to minimum focal distance because the lens is trying to make image distance to 44mm while actual sensor distance becomes 6mm longer, making 50mm distance which is to focus at infinity. 

Because we as astronomer only need focus at infinity with this adapter, we can repurpose this 6mm movement slack for other purpose like a room for filter (typically 5mm in thickness) or not to have to remove sensor tilt adapter for ASI2600. 

 

I already designed and published the design in Thingyverse https://www.thingive...m/thing:4664592 that connects small ASI cameras to E-mount lens that has nothing but E-mount bayonet and M42/0.75 thread. If above hypothesis is right, we can use this adapter on ASI2600 with E-mount lens at close to minimum focal distance of the lens to the star.

 

Ah.. I really wish I had a manual focusable E-mount lens to prove this. 

 

Eric, have I sent the adapter to you as bundle? I don't remember, if I had, can you try?

[AstroCatinfo]

I have a thought. I'm writing loud.

By the famous lens equation

 

1/o + 1/i = 1/f, where o = object distance, i = image distance, f = focal length

 

In camera lens system with given focal length of f, the only way to make a focus to specific object distance of o, is to move lens element distance relative to sensor and make change of i. So, amount of allowable lens movement would define minimum focus distance. 

So, say, there is a 50mm prime lens with minimum focal distance of 0.38m ( Sony SEL 50mm F1.8, for example ), image distance at minimum focal distance would be

 

i = of / (o + f), (o is typically negative, so use positive numbers when move to right term for simplicity) = 0.05 * 0.38 / (0.05 + 0.38 ) = 0.044m

 

and o at infinity, i = f = 0.050m, i.e. there should be about 6mm lens movement. 

 

of course this is 'simple lens' equation and camera lens is not but the idea is that lens has to move by in the order of 6mm. 

 

So, what if, we design an adapter that has 6mm longer than E-mount's 18mm backfocus requirement? 

Now, if my thought is correct and simple lens equation applies to the same way to camera lens, the lens will focus at infinity when lens is moved to minimum focal distance because the lens is trying to make image distance to 44mm while actual sensor distance becomes 6mm longer, making 50mm distance which is to focus at infinity. 

Because we as astronomer only need focus at infinity with this adapter, we can repurpose this 6mm movement slack for other purpose like a room for filter (typically 5mm in thickness) or not to have to remove sensor tilt adapter for ASI2600. 

 

I already designed and published the design in Thingyverse https://www.thingive...m/thing:4664592 that connects small ASI cameras to E-mount lens that has nothing but E-mount bayonet and M42/0.75 thread. If above hypothesis is right, we can use this adapter on ASI2600 with E-mount lens at close to minimum focal distance of the lens to the star.

 

Ah.. I really wish I had a manual focusable E-mount lens to prove this. 

 

Eric, have I sent the adapter to you as bundle? I don't remember, if I had, can you try?

I’ll print this (ask a friend to print it) and do some tests. I do not have the ASI2600 yet because I was just wondering if I could solve this issue. Maybe I can try it with one of my other ASI’s.

 

Thanks for sharing your thoughts. Regards,

 

Aleix

[Linwood]

I skipped any optics class at university so not sure if this is meaningful, but ability to focus the center is not the same as producing a flat, in-focus field across the whole sensor.  Consider an SCT -- it has a VERY wide range over which you can focus (say with an eyepiece), but only one precise point at which the image is best across the whole field, otherwise you get boomerangs at the edges.