Hey!
We’re looking into making the Open Flexure Microscope here at Glia, with a possibility of R&D work into the device in the future. We plan on selling the Microscope to nearby schools, later on setting up an ISO-13485 facility to produce them for lab use. We will also be making them for humanitarian aid purposes, specifically to Guatemala, Gaza and Ukraine. But to do all that, we need to make the first one, so here it is:
The goal of this first unit was to experiment with the existing project as much as possible. The main body is made out of PETG, with other parts being mostly ABS. For illumination, I found cheap LEDs that are meant to be sewn into clothing - they have a tiny pcb with a resistor on it that has holes perfectly spaced for the screws in the illumination stl model. I looped the cables around the screws, and now only need to solder a cable with the crimped end to fit onto the sangaboard. In the future, I’ll need to crimp a round connector with a hole through it. I had trouble sourcing a polypropylene sheet, so I bought a cheapo diffuser for photography and cut that instead. I’ll update this thread once I get the illumination module all connected and working.
I’m using a 60x objective.
For now I’m using the microscope by shining a desk lamp on top of it. I’ll post the photos in the comments, since as a new user I cannot post more than one pic per thread.
Having some trouble with camera calibrating the stage though, it tries to go for ~8k steps, I see the movement in the preview, but the software doesn’t see that the microscope moved. Slide scanning is also unavailable, but I’ll be poking around it seeing if I can get it going
Had two moments where I thought I had everything ready, but got stopped by small details - I didn’t have the card reader, so I was running everything off of USB. Sadly the OFM alpha hardware doesn’t run on USB yet. I got a card reader today and everything went smooth though. The second thing is a failure of logistics on my part - I couldn’t mount the microscope objective because I didn’t have a long alan key here at home. Lesson learned, get alan and torx wrenches to every location you work at, a toolkit isn’t enough ^^
The HQ camera support is very experimental at present. The camera-stage mapping looks at image correlations to check the image movement when the stage moves. The HQ camera sensor extends beyond the image circle of the lens, which gives black corners. @JohemianKnapsody suggests that it could be those dark corners are dominating the image correlations, and they do not move.
One thing to note is that you have the v7.0.0-beta4 body. We made a big change in the v7.0.0-beta5 release last week. It is late in the release cycle for a major update, but we think it is worth it to be able to remove the M4 screws screwed into plastic to hold the motors.
It is something that was always in the long term plan, but we had not worked out how to do it neatly until @ffesti showed the way.
This was exactly what I was thinking about when the calibration part failed - that it might need the image cropped, since it might be looking at the dark area that doesn’t move at all. I tried to mess with the background settings, but I don’t think I understand them enough to know what that does. I’ll check out the new main body, though I might be out of PETG to print it. Regarding the M4 screws, I had to loosen the Z axis ones, because the motor would jam otherwise, but I now have it where it seems to work alright
The z axis clearance is quite small, but you should not need to leave the motor loose. Make sure that the small gear is fully seated on the motor shaft, and that the screws are in as far as possible.
Some small addons, I edited the illumination mount and added a piece of plastic to mount a 7” touchscreen on it. I used a phone mount from a phone holder arm (i.e. like an Ikea Tertial lamp). It has a balljoint on the back and a hinge which you can lock with a screw. The screen that I’m using comes in two variants - just the screen, and the screen + raspberry pi enclosure with a fan. I bought the enclosure version, then hacked off the plastic meant to house the RPi. In the future, I’ll probably model up an enclosure to fit the screen as flat as possible, with the balljoint and hinge modelled up as well and ready for printing. Sadly, I’m being filtered off by OpenSCAD, so I might do it in FreeCAD instead. The biggest issue with this screen / setup is the cable management and the two cables running on the opposite sides of one another.
Edit: This is how the microscope looks when you take the screen as far forward as possible. Moving the mount up would allow using it from the front or the back of the microscope.
