1
$\begingroup$

I have just taken apart a broken USB microscope, and recognized that there is actually no magic going on inside.

So I thought how about building one of my own. The idea is to buy a regular (possibly used) optical microscope's objective lens and a USB webcam. All still low to mid cost, of course. I think I have learned that many objective lenses are optimized for a tube length of 160mm, so if I get one of those, I only need to remove the webcam's lens and put the bare CMOS sensor 160mm apart from the microscope lens (of course, in a pitch black tube with the suitable thread etc.).

But what about the aperture/sensor dimensions? The typical sensors I have seen are about 5mm width. Will that work or will only part of the CMOS sensor be illuminated then? If the sensor only covers part of the image, that wouldn't be a major problem, I guess, because that would mean additional magnification for free...

Or is the aperture of the objective a parameter I need to consider before buying? Or is there something like a "standard aperture"?

$\endgroup$

2 Answers 2

2
$\begingroup$

I don't think the objective will limit you. The focal plane dimensions are typically limited by the eyepiece or sensor. In this table, they are all much larger than 5mm.

$\endgroup$
2
  • $\begingroup$ So "reticle" means the diameter of the image circle (the one that the eyepiece "expects", in case of your linked website) that is cast by the objective, right? $\endgroup$
    – oliver
    Mar 21, 2023 at 21:42
  • $\begingroup$ A reticle is a ruled piece of glass you put at the focal plane. It gives you a way to make measurements of the sizes of objects. Its size normally matches what the eyepiece can capture. The objective will generally have a much larger focal plane. $\endgroup$
    – John Doty
    Mar 21, 2023 at 21:46
1
$\begingroup$

I would like to add my personal experience about this experiment, so everyone who wants to learn the hard way is getting some help here.

I have bought myself a low-cost (supposedly chinese) microscope objective on Amazon (those can of course be bought elsewhere too):

enter image description here

I have built a tube for my F-Mount camera (distance rings + 52mm diameter aluminum tube + 3d-printed part for mounting the objective). The mechanical tube length of the objective is the common 160 mm. Since there is no eye-piece in my arrangement but the camera sensor should be located directly in the image plane of the objective, I had to determine the optical tube length instead. It seems to be a standard (DIN 58887) that the optical tube length is 10 mm less, i.e. actually 150 mm, which then is the distance between the objective flange (in the image: "where the chromium and the brass meet") and the image plane. Subtracting the distance between camera sensor and camera flange (46.5 mm for F-Mount) from the optical tube length gives the "camera tube length" (103.5 mm for F-Mount and the given objective type), i.e. the distance between camera flange and the objective flange.

To my surprise, the image circle was actually large enough so as to illuminate my whole full-frame (35 mm) sensor (with very slight vignetting, to be honest). So the image circle is even bigger than what I had expected based on the accepted answer (my original plan was a ~5mm sensor, but the best/quickest and cheapest solution was to use my DSLR). The image quality is pretty decent, sharp with no noticable color fringes. This is a picture of the surface of my wooden desk taken at about 30 degrees.

enter image description here

Obviously it is not an objective with planar correction (understandable due to the low price) because the focal surface seems to be warped. One can guess the roughness of the wood, but the depth of field is extraodinarily tiny, which I had not expected based on my experience with my broken USB microscope. Obviously the objective used in the latter had a much smaller aperture. I can probably cure it on my DIY construction by adding a diaphragm close to the objective, but at the cost of lower light (and possibly some diffraction).

Summary: image circle is no problem, even with a full-frame sensor. Small depth-of-field is the point to worry about. Of course, for traditional transmitted light microscopy (flat specimen, 90 degrees orientation) this is not much of an issue, but since I targeted at a replacement of my USB microscope, it limits the usability of the arrangement.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.