Microstar IV Camera Setup

Here you can discuss everything related to taking light micrographs and videos.
Post Reply
Message
Author
BrianBurnes
Posts: 11
Joined: Sat Feb 08, 2020 5:05 pm

Microstar IV Camera Setup

#1 Post by BrianBurnes » Mon Jun 22, 2020 2:48 am

Hi all,

I've been trying to build a decent quality DSLR setup for my trinocular Reichert Microstar IV, and I could use some advice from the resident microscope experts on how to fix the problems I'm seeing (or - maybe these issues are expected with this microscope). I'm at a point now where I can capture images, but I'm disappointed with the quality I am getting: No matter what I do, the images seem blurry and contain chromatic aberrations.

I have a Reichert #1943 trinocular head, but I haven't been able to find much information about the intended camera adapter that fits onto the trinocular port (it is a 40mm male dovetail) and whether it contained any optics, so I am mostly guessing here. The head itself has a slider for a 100/0 split, where the slider moves a prism out of the way so that there are no optics between the trinocular port and the tube lens (other than a mysterious glass cylinder). The port itself is shaped like this:

Image

I am using a Canon EOS 7D, and I've been experimenting with two different setups for image capture: A direct projection setup, with the bare camera sensor located at the intermediate image plane and no optics between camera and tube lens; and an afocal setup, with a third eyepiece mounted in focus atop the trinocular port and the camera looking through the eyepiece (with a Canon 50mm f/1.8 lens) focused at infinity.

Pictures of the two setups are below: Direct projection (left), Afocal (right)
ImageImage

To mount the camera I'm using a DYI adapter tube, fixed to the microscope using a 3D printed adapter that replaces the original port:

Image

The adapter holds an M42 ring that leads into M42 macro extenders and a helicoid to fine-adjust the focus. For the afocal setup, I am mounting an eyepiece inside the macro extender tube closely in front of the camera lens. I designed a 3D printed eyepiece holder, but it was destroyed in the mail, and so for now I am testing using a makeshift wedge made out of cardstock that holds the eyepiece in the tube:

Image

In the end, the mechanical setup goes something like:
3D printed adapter -> M42 thread -> macro extension tubes -> helicoid -> M42-to-EF adapter -> camera (for direct projection)
3D printed adapter -> M42 thread -> extension tubes -> helicoid -> extension tubes -> eyepiece -> step-up ring -> Canon 50mm f/1.8 lens -> camera (for afocal)

I've been doing some comparisons to see whether either setup gives me satisfactory results, and I have been testing using a reticule from an eyepiece that I was able to remove.

Below I'm showing results from both direct projection and the afocal setup, at 40x/100x/400x magnification both at horizontal and vertical orientation of the reticule. There is no cover glass, just the glass reticule sitting on top of a microscope slide (so not optimal). Images were shot by focusing through the eyepieces, then adjusting the helicoid until the camera was parfocal with the eyepieces. For the afocal setup, I adjusted the helicoid without the camera attached until the image looked sharp through the eyepiece in the adapter tube. I then attached the camera, focused at infinity.
Results below:

Direct projection: 40x, 100x, 400x (click images for full-res)
ImageImageImage
ImageImageImage

Afocal: 40x, 100x, 400x (click images for full-res)
ImageImageImage
ImageImageImage

While I don't know what the quality ceiling is for this scope/camera combination, I can't say I'm fully satisfied with either of these. The direct projection setup is acceptable, but especially at lower magnification, the image looks a bit hazy. There is also a lot of chromatic aberration towards the edges, which is not great. These aberrations are not present when looking through the eyepieces, as far as I can tell.

The afocal setup is really not very good. The images are hazy and the chromatic aberrations are extreme. Aberrations are not symmetrical which is mysterious.

For now it seems the direct projection setup is just barely usable. Images shot under more realistic conditions look something like this (click for full res):
Image

The quality is disappointing however - even though the image looks very sharp in the eyepieces, the image from the camera just feels blurry even when shooting at the shortest exposure time. The 7D that I'm using has a low-pass filter which may be contributing, but overall it feels like the quality I'm getting out is less than it could be. For reference, here is an image of diatoms that I shot with a hand-held cell phone through the eyepieces (click for full-res):

Image

Even with a much cheaper phone camera held much less robustly (compared to the DSLR setup), the image is sharp and free of aberrations as I would expect.

I'm curious if anyone has any ideas on where these issues stem from and how I could possibly fix them. Maybe there's a glaring mistake I'm making?

User avatar
wporter
Posts: 341
Joined: Wed Oct 21, 2015 10:18 pm
Location: United States

Re: Microstar IV Camera Setup

#2 Post by wporter » Mon Jun 22, 2020 5:24 am

Maybe there's a glaring mistake I'm making?
Yes, perhaps. You are missing the shutter/photocell housing which goes on top of the trino head, and one of the various lens systems (1762, 1763, 1764, or 1768) that goes on top of the shutter housing. Not sure how much compensation or correction these latter would provide, but I suspect some. The shutter housing of course can be gutted and used just as a spacer, or you could make an equivalent.

See the manual "Reference Manual Photostar Automatic Camera System" off the internet, for details:

http://doclibrary.com/MSC167/DOC/PhotoS ... al0956.pdf

MichaelG.
Posts: 2061
Joined: Mon Apr 10, 2017 8:24 am
Location: NorthWest England

Re: Microstar IV Camera Setup

#3 Post by MichaelG. » Mon Jun 22, 2020 6:40 am

The afocal images are quite revealing

Chromatic Aberration is greater :
  • to the right side of the picture
and
  • to the top of the picture
Edit: and the direct-projection images [although less severe] are the other way

This would seem to indicate that the attachment is not axially aligned.
... Philosophical Question: Is 3D printing really up to this job ?

MichaelG.
Too many 'projects'

viktor j nilsson
Posts: 161
Joined: Thu Aug 23, 2018 10:12 pm
Location: Lund, Sweden

Re: Microstar IV Camera Setup

#4 Post by viktor j nilsson » Mon Jun 22, 2020 7:25 am

In fact, if I read the post right, the 3D printed part was destroyed in the mail, so the eyepiece is held by cardboard.

So there is one issue: the eyepiece may not be properly aligned.

However, I am still worried about the height of your afocal setup. Afocal setup are almost always more compact that setups with projection eyepiece. You said that it needed to be this high to be parfocal. This, however, to me suggests that you are missing some optics in your trinocular port. You really shouldn't need that amount of extension. So while your setup is parfocal, I highly doubt that it is working as intended with regards to optical corrections.

photomicro
Posts: 167
Joined: Thu Aug 03, 2017 10:28 am
Location: UK

Re: Microstar IV Camera Setup

#5 Post by photomicro » Mon Jun 22, 2020 8:45 am

I can't find any reference to *what* eyepiece is being used in the afocal method...

MichaelG.
Posts: 2061
Joined: Mon Apr 10, 2017 8:24 am
Location: NorthWest England

Re: Microstar IV Camera Setup

#6 Post by MichaelG. » Mon Jun 22, 2020 9:03 am

viktor j nilsson wrote:
Mon Jun 22, 2020 7:25 am
In fact, if I read the post right, the 3D printed part was destroyed in the mail, so the eyepiece is held by cardboard.
Ah ... but:
.
To mount the camera I'm using a DYI adapter tube, fixed to the microscope using a 3D printed adapter that replaces the original port:
.
For the afocal setup, I am mounting an eyepiece inside the macro extender tube closely in front of the camera lens. I designed a 3D printed eyepiece holder, but it was destroyed in the mail, and so for now I am testing using a makeshift wedge made out of cardstock that holds the eyepiece in the tube:
.

So we seem to have both 3D printed and Cardboard components involved.

MichaelG.
Too many 'projects'

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#7 Post by apochronaut » Mon Jun 22, 2020 11:44 am

There are three main problems with your set ups, if I am reading you correctly.
1) Chromatic aberration. It is worse with the afocal set up.
2) The chromatic aberration is asymetrical.
3) The images suffer from a lack of resolution.

Solution.
----------------

1). Your direct projection to sensor system in theory makes sense, only if the sensor size is smaller than the image circle size of the objectives. You need to install a relay lens in the system . I use a #145 or 145P for projection to an APS-C sensor , which provides aberration free images corner to corner. Probably a 483 eyepiece would do as well or possibly better. You will have to increase your extension considerably to a full frame sensor. Likely close to 250 mm above the telan lens.
The shutter box from a Photostar system will work as wporter suggested. The original photo lenses provided in those do not project as much field to the sensor as one sees. I removed mine and use a 145 eyepiece instead.

The CA with your afocal set up is being caused by your camera lens. There is an assumption often , that camera lenses are really good lenses. Usually they are adequate for photographs but they are not usually apos, so can easily contradict what the eyepiece is projecting, always at the periphery where they struggle to compensate for conditions caused by extremely bent light. A different , perhaps shorter focal length lens might solve the problem , or a different photo eyepiece that magically introduces the opposite types of distortion to that of your camera lens.
2) This is possibly caused by some asymetry in your extension system. Any cardboard pieces would be suspect. Alternately, perhaps the objective is not perfectly aligned. You might not see asymetry with your eyes but the sensor would record it.
3) Large bulky cameras are notorious for creating vibration due to mirror slap. Your set up, with many disparate components may be weak enough to allow that to affect it and be recorded. Mirrorless bodies are more user friendly for lighter bodied tubes than DSLRs but you should be able to find any weakness in your system. It is possible that there is an optical cause but the source of that isn't obvious to me right now.
Last edited by apochronaut on Mon Jun 22, 2020 3:16 pm, edited 1 time in total.

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#8 Post by apochronaut » Mon Jun 22, 2020 11:46 am

Regarding 3) above, if you haven't already you might just check the insides of your trino. I found a nice thick spider web inside one once. It didn't exactly perform as intended.

BrianBurnes
Posts: 11
Joined: Sat Feb 08, 2020 5:05 pm

Re: Microstar IV Camera Setup

#9 Post by BrianBurnes » Tue Jun 23, 2020 4:55 am

Thank you all for your replies!

I'm very much aware of the ad hoc-ness of my setup, and for sure would prefer to do without cardstock or 3D printing. With no access to a machine shop they sadly happen to be what can be easily accessed at the moment (and within budget!). I have been on the lookout for the photostar camera system but have not had much luck on ebay. There is an adapter tube that very rarely pops up but I'm unclear what eyepiece is supposed to go with it.

From your replies there seem to be three problems with my current setup: Centering of the eyepiece (for afocal), alignment of the optical tube, and mechanical stability. The eyepiece is visibly not centered at the moment, but it may be a while before a more reliable way of mounting it can be produced. That being said, it seems less desirable to add more glass between the objective and the sensor (the lens does add its own CA as apochronaut points out), and I prefer the direct projection setup, so the eyepiece may become less of an issue. Afocal was tried only because of the aberrations in the direct projection setup, which I assumed was because the microscope used corrective eyepieces (is this the case for this model?). I used the same #181 eyepiece as the head to (hopefully) achieve the same corrections.

As far as alignment goes, I am not sure whether the tube itself is responsible. Loosening its alignment screws and moving the tube side to side (and even tilting it) seems to have little effect on the chromatic aberrations. However, through a Bertrand lens I did notice something that I found unusual in the microscope itself: Even when the condenser is adjusted so that the field diaphragm is centered within the view, the condenser aperture does not appear centered when using the Bertrand lens to view the objective rear focal plane. I don't know enough to tell whether this is benign, or perhaps something is out of alignment in the rest of the scope (but what? the objective turret?) and contributes to asymmetric aberrations.

Finally, mechanical stability could certainly be an issue - my setup is far from solid - but what leads me to believe other factors are at play is that even in live view (with no shutter movement) the images still appear blurry to me (I have also been using EFCS to reduce vibrations in the posted images, though that may not bee enough).

That being said, I don't have enough experience to know what the quality ceiling of this microscope/camera combination is. With achromats I'm sure some CA is expected, and the image through the eye pieces appearing sharper may just be my eyes playing tricks on me (they do have higher dynamic range after all), so perhaps these images are about what can be expected?
I use a #145 or 145P for projection to an APS-C sensor , which provides aberration free images corner to corner. Probably a 483 eyepiece would do as well or possibly better.
This is very helpful, thank you! The #145 seem to be easy to come by, the #145P less so. Is there a practical difference from one to the other? I have also seen the #437 referenced elsewhere, would that be usable as well?

Zuul
Posts: 73
Joined: Fri May 01, 2020 9:01 pm
Location: California

Re: Microstar IV Camera Setup

#10 Post by Zuul » Tue Jun 23, 2020 3:58 pm

With achromats I'm sure some CA is expected
True, but the Neoplans should be better in this regard.

I’m following with great interest having recently received a Microstar IV of my own. Unfortunately, my plans have been slowed by a cracked lamp socket. It was working when I received the scope, but during my inspection and cleaning effort, it gave up the ghost.

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#11 Post by apochronaut » Tue Jun 23, 2020 4:36 pm

My computer has failed recently, so I have to use a tablet, which I find cumbersome. I tried answering your questions individually with quotes but by the time I was done, I was logged out, so I will just ramble.

The adapters that I have seen: several photostar set ups and Diagnostic Instruments are integrated with an optic.

Direct projection is problematic. The reference length for the system is around 180mm, so you might try that telan lens to sensor distance. I tried a direct to sensor approach and was unimpressed with the results. I think it was because, if your tube is too short, you take in too much of the telan lens diameter, straying into peripheral aberrations that the lens system as an entirety is designed to vignette out. Every lens system has peripheral aberrations, even apochromats: that's why lenses are always much bigger than the beam that passes through them. There was also the possibility that I was picking up the lmage in the wrong image plane.It was a long time ago and I don't remember all of the problems now but I remember that the background was uneven. I abandoned the idea in favour of using a relay lens.

A relay optic picks up the perfectly corrected central part of the image circle and passes a well corrected image circle to the sensor at a correct distance.Both the 181 eyepiece and the 145 or 145p eyepiece fulfill this role. The 145p has just a focusable photo framing reticle installed, so I remove that. The 437 is an older focusing ocular, made for the 34mm parfocal system, which required some further eyepiece corrections. In the 400 series system, it creates a little off axis ca.

Regarding alignment. I think you have two issues. Misalignmrnt of your relay optic and misalgnment of the illumination system. Misalignment of the relay optic will cause existing ca to be assymetrical. It won't create it, unless it is very severe. Misalignment of the illumination system is probably a consequence of someone frigging with the nospiece. That will not create ca in the image again, unless it is really off. Align the nospiece to the field diaphragm by manipulating the 3 allen screws that orbitally adjust it. Then bring the condenser in with tweaks to the nosepiece.

My sensor is an APS-C, and in order to achieve full field coverage with parfocality(while using a relay lens), my lens is about 150mm above the telan lens and the sensor another 100mm above the lens. I use a helicoid focuser for fine trimming of the focus. With a full frame sensor, you will need to go out to around 300mm.

There is no ceiling quality to your microscope/camera combination that is unique. Each of the two systems is what they is. A mismatch is due to the owner creating it.
I'm sure you know your camera well. The microscope makes an image as good as any other in it's class and era; better than most actually. Put planapos in it and it provides fine planapo imaging across a 20mm field. Put planachros in it and it provides fine planachro imaging. The ca is consistent with planachromat imaging; slight but uniform across the field. Skewed or increasing ca off axis is caused by mechanical and optical problems.

It is important to note that very few photomicrographs are representative of what the eyepiece sees. Most excellent images have been treated to a degree, sometimes high of post processing.
Last edited by apochronaut on Tue Jun 23, 2020 7:43 pm, edited 1 time in total.

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#12 Post by apochronaut » Tue Jun 23, 2020 4:43 pm

Zuul wrote:
Tue Jun 23, 2020 3:58 pm
With achromats I'm sure some CA is expected
True, but the Neoplans should be better in this regard.

I’m following with great interest having recently received a Microstar IV of my own. Unfortunately, my plans have been slowed by a cracked lamp socket. It was working when I received the scope, but during my inspection and cleaning effort, it gave up the ghost.
The opposite. The Neoplan objectives were a more budget friendly 45mm adaption of some of the 34mm designs. They were pitched at institutional use; schools etc.
They are not quite plan at 20mm, somewhere between semi-plan and plan.

Zuul
Posts: 73
Joined: Fri May 01, 2020 9:01 pm
Location: California

Re: Microstar IV Camera Setup

#13 Post by Zuul » Tue Jun 23, 2020 6:12 pm

apochronaut wrote:
Tue Jun 23, 2020 4:43 pm
The opposite. The Neoplan objectives were a more budget friendly 45mm adaption of some of the 34mm designs. They were pitched at institutional use; schools etc.
They are not quite plan at 20mm, somewhere between semi-plan and plan.
Wouldn't a simple "achromat" imply the objective is not particularly plan at all? Planachro would be a step above the Neoplan, but that wasn't what was said.

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#14 Post by apochronaut » Tue Jun 23, 2020 7:42 pm

There were no achromat objectives for that series. I think the reference to achromats was in a generic way, referring to a class of colour correction, which Neoplans also partake of, although with less correction for all aberrations and distortion at the periphery.

BrianBurnes
Posts: 11
Joined: Sat Feb 08, 2020 5:05 pm

Re: Microstar IV Camera Setup

#15 Post by BrianBurnes » Wed Jun 24, 2020 4:20 am

Thank you for taking the time to write such a detailed response apochronaut, this is very useful information. I've been able to realign the nosepiece and the quality seems improved.

In your setup, are you using a camera lens at all, or does the eyepiece project directly onto the sensor? If it is the latter, I did not realize that was possible, and I look forward to trying that. If you don't mind my continued questions, would I see an improvement using the 145 over the Microstar 181, or do they have roughly the same properties?

apochronaut
Posts: 3245
Joined: Fri May 15, 2015 12:15 am

Re: Microstar IV Camera Setup

#16 Post by apochronaut » Wed Jun 24, 2020 2:11 pm

No. A camera lens would be used if an afocal set up was desired. It would not matter whether the camera was mounted over an eyepiece or over the trinoc. Each would be afocal. Typically, a lens mounted on the camera of a shorter focal length than normal, works best. Camera lenses tend to introduce distortion or aberrations and may require cropping. This why phone cameras often produce better pictures than one would expect. They innately have greater depth of focus and some have software that automatically corrects for defects. Some even can stack. There is still the problem of the tiny sensor though.

Most set ups on the trinocular use a camera relay lens, that converts the convergent beam to a divergent beam. This can be anything that works. With some systems a reversed copy lens works, with some an enlarger lens, with some an eyepiece. Anything is possible as long as the lens produces a flat projected image which is free from distortion and aberration as far as the corners of the sensor. Different photo eyepieces can introduce different aberrations because in the system they were made for they are required to correct for aberrations innate to the system. That's why using a photo eyepiece other than that made for your system is trial and error.

In the AO/Reichert 400 series system, the telan lens located in the bottom of the head converts the infinity bundle tof rays to a convergent bundle focused at 180mm( approx.), where the eyepiece picks up the image, magnifying it 10X and reversing it to a divergent bundle. Projecting it. From the telan lens north, the image is fully corrected. If a planachromat image leaves the telan lens then a planachromat image projects from the eyepiece. No further corrections take place. Installing a 181 eyepiece in the photo tube works as a projection lens. Placing it at the correct distance from the telan lens and your sensor the correct distance from it, will give you as perfectly a corrected image to your sensor, as you get from the eyepieces.
I would recommend shooting for 178mm, glass surface to glass surface, for the eyepiece location. Mine sits at around 150mm, only because the photo star housing provided that resting place but I can still get parfocality. The original lens in the photostar was further up the line. I removed mine. With a full frame sensor you will need to place it another 100mm above the eyepiece. , probably less. Bear in mind that the working distance of your objectives will be progressively lessened inverse to magnification with the sensor sitting that high. Shimming the 2.5X and 4X are an option.

Something else you may want to try is to use a teaching bridge as a photo tube. This will pass the infinity bundle over to another head but instead of the head and telan lens, you attach a camera lens set at infinity as a convergent lens. For a full frame camera you will need probably a 300mm. You may even need a teleconverter. The quality will be determined by the lens quality. I have used an adapter I had Raf camera make up that went from the 2" dovetail to 52mm filter thread on a Nikkor Q 200mm prime lens. This worked pretty good with an APS-C sensor and covered a very wide f.o.v. , into parts of the image circle with correction defects. Cropping was therefore required but the field capture was exceptional. A zoom lens may also work but due to the moving elements and often physical restrictions in zooms, some of them will vignette.
Just a thought, if you wanted to stick to camera lenses. A 300mm apo might be great.

Post Reply