Is there any reason why a flexure is preferred over screw slides like 3D printers/CNC machines for the xy plane? A NEMA11 (200 step/rev with a micro stepping driver [up to 1/32]) and a 1mm pitch screw gives ~0.16um/microstep. That’s around the magnitude Openflexure gives ~0.08um while giving a much larger range of motion.
Flexure based systems are generally prefered over sliding systems for very fine motion. They do not suffer from stick-slip as nothing is sliding. If you look at traditional translation stages, they run on sliding for large motion and reasonable precision, then are flexures for a few mm range at very high precision. Both of those systems are formed from many separate parts.
The beauty of a 3D printed flexure mechanism is that it can be printed in a single piece, and gives that high precision movement. One piece helps with stability as well as build time. The range of motion is enough for high magnification imaging (100×) and the precision and stability are required. That was the starting specification for the OpenFlexure Microscope.
Other applications do need a much larger range, and for that a system based on metal sliding parts makes sense. In traditional systems you will often see stages mounted together to give large overall motion, with fine motion on top of that, which is used once the system has arrived at a region of interest.
The Enderscope uses 3D printer mechanics for a microscope using relatively low magnification.
I see…I mainly find it troublesome to move slides out of the way to calibrate for the grid illumination (whenever I change colors or light up different LEDs), then trying to find the same object all over again. Anyone has any tips to make the workflow easier?
Using the Delta Stage on a sliding system will be difficult due to the placement of the illumination. Assuming the optics module is ABOVE the slide, the illumination has to be at a fixed position relatively close to the slide; i.e. the illumination will be in the middle of the Delta stage, colliding with the stage when any long movement is attempted.
I think it’s also worth saying that we have thought for many years that the OpenFlexure Z mechanism combined with a printer-style XY stage would be a fantastic microscope. We’ve just never quite found the time to do it (and/or a suitable excuse to justify the time). @biodotpe has posted some really cool stuff using the “flat top” stage with a (manual, I think) XY stage which is going in that direction. I suspect @B.Diederich has also done some very relevant things with UC2.
I would be very interested to see how the out-of-plane motion compares between the two. My gut feeling is that a stage with linear bearings would have to be quite good to achieve the same stability as the flexure stage in the focus direction - but I have learned not to trust my gut too much in the face of hard data!
Thinking right back to the start of the project, printability was a big part of my initial design: I wanted to see how much could be printed, to eliminate custom/complicated parts. Printing the flexures was possible - printing nice linear bearings is harder. Sourcing and maintaining parts is also a consideration: nice linear bearings aren’t universally available, and they need to be kept clean and lubricated. Given that plenty people are interested in OpenFlexure because we’ve minimised the bill of materials, and/or because they are working somewhere a bit challenging (=dusty/humid/whatever), there’s an advantage for flexures there - but I think there are certainly plenty places where a more conventional XY stage would be a really useful option.