January 29, 2025

Three Mitutoyo lenses: 5x, 10x, and 20x

I have all three (M-plan APO). I began asking whether my 10x was in some way damaged or inferior. How would I know? I don't have another 10x at hand of known pristine condition to compare it to. So I got the idea of comparing the three lenses I do have.

The subject is a microcontroller chip I have used for such things in the past. It has two virtues. It is flat, and it has lots of fine detail. The chip is an ancient Intel 8748 -- an 8 bit controller that found extensive use in keyboards in times long past.

I am shooting with my Canon R5, which is a full frame mirrorless camera. I am using a Raynox 150 as a relay lens. I am using electronic flash (my Canon 580ex speedlight) for illumination.

The Mitutoyo 5x

First we have the full image captured by the camera. The field is about 8mm from top to bottom. I ran a stack of 50 exposures and selected number 35 as having the best focus. It turns out that I don't have the chip entirely level, so the entire thing is not in optimal focus.

The R5 sensor is 8192 by 5465 pixels, so if I want to extract a section 1200 pixels wide, I will sort of need to eyeball 1/8 or so of the sensor width and then crop. Doing that gives what I show next, where I pick a part of the image that shows the best focus.

Adobe lightroom makes it a total pain in the ass to do this, so this is not the exact 1200 pixels I might like, but it is verbatim camera pixels. The image is 989x650.

The Mitutoyo 10x

The same game agin, now with the 10x lens. Now the field is about 4mm top to bottom. Here I shot 41 images (with 3 micron steps) and decided that image 18 was near best. This lens shows a signficant amount of curvature in the "plane" of focus which you can see especially on the left side. You can't really tell on the right because the wires and such are drastically higher than the chip, but you an still see it starting at the edge of the chip.

Deciding on what section is in best focus (and which image is in best focus) is not easy, and I may want to revisit this someday. I managed to crop out the 1084x722 pixels below from what looked to me like the part of the image with best focus.

The Mitutoyo 20x

Here we decided to photograph the upper right corner of the chip. The field now is about 2mm top to bottom. I shot 28 images and liked the focus best in number 25. Here the right side of the chip goes out of focus, but the effect does not look "spherical" as it does for the 10x lens. It just looks like the chip was not level.

Here we end up with a 1562x1042 pixel crop. Close scrutiny shows diffraction limiting the resolution, which is what we hope for really.
This image is 0.38 mm top to bottom.

Although the image you see here is bigger than the others, it is still 1:1 pixels from camera to the cropped image with no scaling applied. I just cropped more.

As an amusing calculation. If the field is 2mm and the camera sensor has 5464 pixels, we are putting a 0.366 micron subject distance on a camera pixel.

Conclusions

The idea here was to get an idea of how well these Mitutoyo M-plan APO lenses perform. I am pleased and impressed with the 5x and the 20x, but I really think the 10x is flawed.
I want to get some other experts to look at these images and give an opinion.

Prices for these right now (1-2025) at Edmund Optics are: 

5x  0.14 NA -- $770   ($191)
10x 0.28 NA -- $1040  ( ?? )
20x 0.42 NA -- $2440  ($524)
The number in parenthesis is the price I paid on Ebay for these. My 10x was bought at least 5 years ago and I have entirely forgotten what I paid, probably around $500. You could argue that the money I saved on the 20x easily pays for another new 10x on Edmund.

Feedback? Questions? Drop me a line!

Tom's Computer Info / tom@mmto.org