Build summary
I just finished building my v7 OpenFlexure microscope with all parts purchased from IO Rodeo. I think it is hardware version v7.0.0-beta1, and I used a Sangaboard v0.5. The build took me about 2 hours and the kit had everything I needed except a prepared slide for testing.
Build feedback
The instructions from the OpenFlexure wiki were very thorough and easy to follow. I particularly appreciated the design of the O-Ring mounting tool, which made what would have been an otherwise extremely frustrating step very easy. The 2 screw reinforcements to hold the motor gear onto the keyed stepper shaft were also a great touch. I do some hobby 3d design and printing myself, so I very much respect the expertise that went into this design. Overall, OpenFlexure as an open source project is extremely cost-effective and well-designed and documented. Hats off to Mr. Bowman and the rest of the OpenFlexure team. I don’t have much other design feedback besides maybe using herringbone gears for less backlash, but I think I already saw a discussion about that elsewhere on the forum.
Build Questions
The one thing I didn’t find in the build guide or on the forum was a how to set up the brightfield lighting system. I intuitively figured out that the adjustment nuts for the lighting system should center the light cone on the objective lens area. How should I determine my ideal z-height along the height dovetail system?
It seems like the condenser lens and LED matrix PCB system produce a bullseye-profile (see picture) light cone, at least when I back the z height of the system to the top of the adjustment dovetail. I would naively expect the ideal lighting system to produce a uniform, diffuse flux of light across the field of view of the objective lens. Have I somehow mis-assembled this system? Does the lack of uniform brightfield even matter since I can use the “Auto Flat Field Correction” feature to correct for it?
Intended Applications
The main application I have in mind for this microscope is doing imagery of local pollen samples, eventually working my way up to semi-automated analysis of pollen grains from honey samples. I plan to use the image tiling capture feature to capture a wide field of view, then use openflexure-stitching or some other program to stitch the images together. I may even try some focus-stacking as well, we’ll see how it goes. I’ve also got parts on the way to upgrade the modified raspi camera 2 lens to the recommended 40x planar objective lens, and I’ll do a build report on the upgrade process as well.
1 Like
I, being a new user, am only allowed to upload one image per post. So here are the rest of my images:
2 Likes
Detail of “bullseye-shaped” light cone
1 Like
Here is one of my first raw camera images of a prepared pine stem specimen. I’m going to do some focus-stacking and try to enhance it. The lack of planar optics seems like an obvious limitation here, and I look forward to trying again with the 40x planar objective lens instead.
1 Like
Awesome, Thank you for the detailed build report.
On the topic of the the bulls-eye focus even with a lens in perfect focus you will get a slight bulls eye known as an Airy disk. However, this gets much worse for lenses with spherical abberation. As the condenser lens is a low cost plastic lens I would guess that it is spherical, and this is the cause. It is normal and isn’t anything to worry about. For really pushing the resolution limit of the microscope with a 100x oil immersion objective it becomes much more important to start considering the illumination lighting.
This a very good point. I think we are getting the main assembly instructions to a point where they are really clear, but once it is assembled. The “first use” instructions or setup instructions are lacking, we are also lacking a good complete user manual, these are on the roadmap!
As to your specific question, The best thing to do is to put a piece of paper where the slide will go as you have done. And then move the condenser into the sharpest focus (smallest spot). You can then remove the paper and looking at the display on the microscope try to get the spot as central as possible. Once you have done this, you can run the flat field correction again.
Thanks again for the build report and the images. We look forward to seeing your pollen images!
I’m always happy to contribute to a worthy project. I’ll keep track of my first-use notes and maybe I can contribute to that. I poked around the details of the official OpenFlexure team, and wow are you guys highly-credentialed! Your contribution with GitBuilding seems very useful to solve what you term “reinventing the wheel”. Great work. 
In my opinion, documentation and repeatability are really underappreciated parts of making progress in physics science.
2 Likes
Thanks 
Any notes on your first use would be great. I think it is difficult to document first use when you are “too-close” to a project. We do have this handbook which is very out of date, incomplete, and ill formatted. We will probably start again but pull in some of the content.
I finally got my upgraded optics parts in the mail, and the difference is huge! The upgrade was really simple. I just followed the standard build instructions, no pitfalls or trouble along the way.
Both of the following images are the same 0.01mm-division calibration slide. As everyone here probably knows, the X and Y axes of the microscope are “diagonal” to what one might expect.
The low-quality optics build has a field of view of about 0.8 x 0.8 mm corner-to-corner
And using the 40x plan lens from the high-quality build, the field of view is about 0.2 x 0.2mm, roughly 1/16 of the visible area:
Here is a colorful part of the edge of a pine stem I’ve been using for test photos. The better optics capture the details so much better than the low-cost one. Focus-stacked images to follow.
4 Likes
