How to get proper camera image from microscope photo tube?

[Aimbit]

Hi all,

I just got an Amscope SM-4TP for soldering and I am trying to hookup a camera to it. The problem is not getting the camera connected, but rather that when connected, the displayed image is so zoomed in its basically unusable. I'm using a 0.5x barlow lense on the microscope, 10x on the eyepieces. The camera is connected to a 0.5x reduction lense before being put into the photo port.

Basically my question is, is there a way to 'zoom it out' so that it is much closer to what I see in the regular eyepieces?

It is this microscope, https://amscope.com/products/c-sm-4tp-af?variant=41421264289967 and this camera, https://www.amazon.com/Industry-Inspection-Microscope-Detection-Soldering/dp/B07WS2X1NR/ref=sr_1_5?crid=2L1AV3AG454C1&keywords=hayear+16mp&qid=1704658726&sprefix=hayear+16mp%2Caps%2C231&sr=8-5&ufe=app_do%3Aamzn1.fos.f5122f16-c3e8-4386-bf32-63e904010ad0

Do I need a different magnification lense? I see another Amscope camera designed for a '23 mm tube' like mine, comes with a 1x compensating adapter.

[jfiresto]

Have you made everything parfocal: first the eyepieces and then the camera?

That will place the entrance pupil of the relay lens + camera at the photo tube's intermediate image. If the lens is designed for the 'tube and camera, that should match the sensor view to the microscope's image – with perhaps a little cropping or vignetting (a matter of taste).

Reading that back, that may be a little too pithy and I could smooth things out a little.

[Aimbit]

I understood a few words.

There is no way to adjust the focus on the camera tube as far as I'm aware. There is an attachment on it to adjust the height of the camera up/down, but that seems to be it.

I took two pictures... one is what the camera outputs to the monitor the other is what I see through the eyepieces. As you can see, the Camera is very zoomed in. This is the lowest magnification I can get. I also tried putting the camera into the right eye piece, the result is the same.

https://imgur.com/a/RzW9mRM

edit: I just read in the manual -- This feature allows the images through the microscope’s eyepieces, and those displayed on your
computer screen or television be viewed at the same time (although unless using 20x eyepieces to
match your camera, the image may differ in magnification)

So I guess the default is stuck at 20x in there... that stinks. I'll have to try a 0.37x reduction lens. Going to try adding a 0.3 barlow also instead of my .5... Should reduce the camera even further but we'll see how it works out.

[jfiresto]

There is no way to adjust the focus on the camera tube as far as I'm aware. There is an attachment on it to adjust the height of the camera up/down, but that seems to be it.

By that, you mean you can move the reduction lens and camera, up and down, as a body? If so, I would do that to make the camera parfocal.

I just read in the manual -- This feature allows the images through the microscope’s eyepieces, and those displayed on your computer screen or television be viewed at the same time (although unless using 20x eyepieces to match your camera, the image may differ in magnification)

If that is in the manual, Amscope is aggressively cropping the image by design. You can try to shrink the image with a lower power reduction lens, but I would be sorely tempted to substitute a camera with a larger sensor to get a brighter image. I have yet to see a complaint about having too much light. With your present 0.5X reduction lens, a four thirds sensor (what I decided on for a different microscope) has nearly four times the area, and would roughly capture the view through 10X eye pieces – assuming there is nothing in between to block or blur the image at the sides. Amscope may expect you to magnify the camera image so much and only view the middle, because only that part is good.

You can size the sensor you are after, by removing the camera and viewing the image projected by the reduction lens. I set the parfocal microscope at minimum zoom, view an upside naked-LED ring light and project the image on to tracing paper or a frosted microscope slide. I then view the projected image through an eyepiece or project the image on to a camera sensor: to check how the image holds up, and how far, approaching the sides.

[magic]

1/2.3" sensor size means about 6mm across. With 0.5x demagnification you see a 12x9mm crop of the primary image. 10x eyepieces may see 20mm or so.

[jfiresto]

Grrrrr. I misread the camera size, perhaps from seldom seeing fractions.

A camera with a 6mm diagonal would need something like an 0.3X reduction lens for the corners to touch the edges of the view through 10X/20mm eyepieces. That implies the camera tube has the same magnified, ca. 20mm diagonal view as the eyepiece tubes. But then the manual suggests using 20X eyepieces to match the camera which implies its tube image has more like a 12~13mm diagonal.

Assuming all three tubes have the the same views, before reduction, you would want a 1X relay lens/adapter for a one-inch sensor, and quite possibly a four thirds sensor as usually the intermediate images in both the camera and eyepiece tubes are somewhat larger.


EDIT: The first two sites I found, give 7.66 and 7.70mm diagonals for two 1/2.3" sensor with slightly different aspect ratios. From that I calculate a 0.38X reduction lens to match a 20mm camera tube intermediate image and 10X/20mm eyepieces – which confirms the 0.37X reduction lens you calculated. Sorry about the noise.

1/2.3" is a small sensor. Have you considered something larger?

[magic]

A small camera and reduction lens probably ends up cheaper than a 4/3" sensor.
Not sure about image quality of those lenses, but I suppose there should be some reviews around as it surely is a common issue, isn't it?

(NB: I'm not a stereo microscope user myself, but I have some interest in microscope cameras for other reasons.)

[jfiresto]

A small camera and reduction lens probably ends up cheaper than a 4/3" sensor.
Not sure about image quality of those lenses, ....

Yes, that seems to be the crux. There are companies that will cheerfully sell you a $35–45 dollar reduction lens, but the optical performance can be quite poor. When I was adding a camera to a different stereo microscope, good reduction lenses seemed to cost around 600-900 dollars/euros. Granted, they may have been marked up 100 or 200%, coming from low volume, Western vendors. But I could not find anything decent in between.

That persuaded me to seek a camera with a bigger sensor that could capture the image out of the photo tube – projected directly on to the sensor with no intermediate optics. It is hard to beat air for not degrading an image, and purely mechanical adapters can be quite cheap or 3-D printed. The photo tube and adapter, however, have to project the intermediate image somewhat less than 20mm above the top flange, or be made short enough that they do.

A few years ago, a used micro four thirds camera was the cheapest source of a suitably sized, high quality sensor, in Germany for my microscope (ca. 110-150 euros for a camera with fewer than 100 shutter actuations and often a one year guarantee). These days, if I were buying in the U.S. – where the OP is? – I might consider, say, the Nikon 1 J3 to see if it would answer. It has a one-inch sensor, an electronic shutter, live video out, a shorter flange focal distance, and as little as a $100 sell price on ebay.com.

Anyway, just an idea that may not work. I really do not know Nikon cameras, the OP's microscope or the optics of its photo tube.

[metebalci]

Yes, that seems to be the crux. There are companies that will cheerfully sell you a $35–45 dollar reduction lens, but the optical performance can be quite poor. When I was adding a camera to a different stereo microscope, good reduction lenses seemed to cost around 600-900 dollars/euros. Granted, they may have been marked up 100 or 200%, coming from low volume, Western vendors. But I could not find anything decent in between.

That persuaded me to seek a camera with a bigger sensor that could capture the image out of the photo tube – projected directly on to the sensor with no intermediate optics. It is hard to beat air for not degrading an image, and purely mechanical adapters can be quite cheap or 3-D printed. The photo tube and adapter, however, have to project the intermediate image somewhat less than 20mm above the top flange, or be made short enough that they do.

A few years ago, a used micro four thirds camera was the cheapest source of a suitably sized, high quality sensor, in Germany for my microscope (ca. 110-150 euros for a camera with fewer than 100 shutter actuations and often a one year guarantee). These days, if I were buying in the U.S. – where the OP is? – I might consider, say, the Nikon 1 J3 to see if it would answer. It has a one-inch sensor, an electronic shutter, live video out, a shorter flange focal distance, and as little as a $100 sell price on ebay.com.

Anyway, just an idea that may not work. I really do not know Nikon cameras, the OP's microscope or the optics of its photo tube.

I might be using a scope from the same brand (as @jfiresto), as I have the same experience.

I am not familiar with Amscope products and I dont understand quickly from the product page if this scope has an infinity corrected design. If it is infinity corrected and if there are no special needs, the best solution I found to this problem for my scopes is to get a camera with a sensor as big as possible but still staying within the image circle at the photo port so no vignetting will be visible. So the sensor diagonal should be smaller than the image circle diameter but still as big as possible. For my scopes (Zeiss) and among consumer cameras, micro four thirds work best (+ there is an MFT camera that can both output a 4K video via HDMI and also capture a 4K video, which was not a common feature a while ago). Thus, without using any optical elements, I can capture the maximum field of view (+ it is cheaper, as mentioned above, a proper extra optical element can be quite expensive). When a scope is used in a professional/research setting, there are other things to consider, so microscope specific cameras are a different topic.

If the scope is not infinity corrected, then there has to be an optical element in the tube anyway, which might change things, and I dont have any experience with them to say something more.

[jfiresto]

I did a little more Nikon camera research as there is a Nikon 1 J3 up on ebay.de at the moment for under 80 euros, delivered, with accessories, 4767 shutter actuations and a one year guarantee. Apparently, all the Nikon 1 models have fixed screens, and the J3 can only redirect the 3" LCD output, at its resolution, to an HDMI monitor. Together, those two limitations are a deal-breaker for me. Now, the Nikon 1 V3 looks interesting. It supports tethering at higher resolution, but it is a good deal more expensive. EDIT: Nikon also wants more money for the tethering software.

In any case, if I had the Amscope, I would first want to view and measure the camera tube's intermediate image to see what I had to work with. Things could turn into a little engineering project, but there might be a few odd characters, here, who enjoy such things.

[magic]

OK, I guess this can work if you find a camera which supports live view over HDMI. But you have to be careful with camera choice as you found out, since they differ in output resolution and not all of them have this feature in the first place. I'm 100% sure my Olympus PEN can only output HDMI in playback

Tethering sounds like USB and that doesn't have a lot of bandwidth. You will have the image quality of a USB 2.0 webcam at best. I wasn't very impressed with USB 2.0 cameras - it's either low resolution, very low framerate, or MJPEG at low quality factor. Maybe newer cameras can tether over USB 3.0, but they won't be so cheap anymore (compared to basic large sensor microscope cameras).


BTW, isn't the image in those phototubes projected 1cm below the top of the tube, like in eyepiece tubes? How are you connecting large cameras to them without projection eyepieces?

[jfiresto]

... I'm 100% sure my Olympus PEN can only output HDMI in playback

The later PEN models support limited tethering over WiFi, if video over USB 2 would be too fast.

[magic]

Yeah, it requires some Android app (or a desktop equivalent, I believe somebody wrote a replacement in Python) because the live view is a custom protocol based on UDP. My camera's WiFi is 802.11g so I'm sure quality would be much worse than USB, I didn't even bother trying.

However, I can trigger the camera wirelessly and download files from it (slowly). So it could be handy in some situations, maybe. Oh, and the camera can't join your network, it starts its own which you must join, sacrificing one WiFi adapter for this connection.