MozFest 2017

I’m excited to announce that we’ll be showing off the OpenFlexure Microscope at MozFest 2017, in the Youth Zone.  Come and find Richard there on Saturday morning to build, hack, and play with an open source digital microscope.  An hour isn’t very long, so it will be a pretty hectic session – but it’s always fun to share the joys of microscopy with new people, and to get folk tinkering with some code that controls the hardware!

The OpenFlexure Microscope started as a curiosity project, built because I wasn’t sure whether it was possible.  Several years later, it has a bunch of people working on it, it’s been replicated around the world, and I’m showing it off at MozFest!  One of my dreams is that this sort of hardware will be used in schools – printed and assembled in tech studies, programmed in computing, and then used for physics and biology experiments.  Testing out the microscope (and just as importantly, the build instructions) at MozFest will be an interesting learning experience for me – that’s how the assembly process has been refined so far.

So, I’ll be heading to London this weekend for some open-source fun – pop in if you’re in town.  I know I’m in the Youth Zone, but I’m sure adults are welcome too!

Goodbye bent paperclip!

Goodbye bent paperclip!


There are a couple of steps that people consistently struggle with when assembling the microscope, for good reason; they're pretty tricky.  Number one on the list has to be hooking in the elastic band - I've spent many hours myself jiggling around the bent paperclip tool in an effort to get the bands in, although I've assembled a lot now and probably have "the knack".

It's always easier to spot the problem here than to find a solution, but the latest development branch (v5.16) now has some progress - a different design of actuator where the elastic band gets hooked in from the bottom.  While I was at it, I changed the way the nut goes in, so it's inserted from the side rather than the bottom.  That means there's no more need to get the push-fit to work (which required strong fingers, and sometimes didn't work at all) in order to fit the actuating screw.  We've still got to test it before it's ready for the kits, but there's more details on the OpenFlexure Wiki and you can grab the STL files from the development branch on GitHub.

Once we've checked it works nicely, you can look forward to updated kits soon!

What magnification is it?

What magnification is it?

The number one question I get asked when I show people the microscope for the first time is "what's the magnification?" - and as anyone who's asked me this will tell you, the answer's not entirely straightforward.  Microscopes are often characterised by a magnification, a number telling you how many times larger they make things appear.  Typical laboratory microscope objectives range from 4x to 100x magnification, often used in conjunction with eyepieces that magnify the image further - often 10x but sometimes considerably more.  Sometimes (particularly in microscopes for home or school use) the product of eyepiece and objective magnification is quoted - this can be a high number, perhaps 1000x or even more.  Our microscope is digital - that means you don't use an eyepiece, and instead the second magnification step is performed by acquiring the image on a small sensor (usually a few mm across) and displaying it on a much larger screen.  Because the pixels on the sensors we use are really small, this can magnify the image 100x or more without losing quality.  If I'm pushed to give a magnification number, I will usually tell people that the resolution (i.e. the smallest thing you can see) of our basic kit is equivalent to a typical 20x microscope objective, though the magnification is closer to 50x.  The important quantities are the resolution (the smallest thing you can see) and the field of view (the area of the sample you can observe at once); for the basic Raspberry Pi kit these sizes are around 1 micron and around 280 microns respectively.  There's now a couple of pages describing this on the OpenFlexure Microscope Wiki.  I hope they make things clearer - suggestions and improvements are welcome!


Using “proper” microscope optics with the OpenFlexure Microscope

Using "proper" microscope optics with the OpenFlexure Microscope

A commonly-asked question when I speak to people about the microscope is "can it do X?" for various different values of X.  Often my answer is that, while the basic model I'm showing you here can't, I have another version that can.  This blog post is about the other versions, and how to get hold of them.  First though, I'll list a few of the most commonly asked-about features:

  • Better resolution/higher magnification
  • Wider field of view
  • Use of a "proper" microscope objective
  • Imaging modes other than bright-field (e.g. dark field or phase contrast)
  • Fluorescence imaging
  • Motorised control of the stage position and focus
  • Upright geometry (looking down on a sample rather than looking up through the coverslip)

All of these things have been done with the microscope, pretty successfully.  however, not all of them are available easily.

Writing good instructions is hard, and despite putting a lot of effort into our instructions, it's still very easy to confuse people.  This is particularly problematic when there is more than one way to do something; experience has taught us that it's easier for everyone to pick a method and stick with it.  However, this makes it hard to include optional features, which usually affect one of the assembly steps in some way.  Consequently, I usually encourage people to get in touch if there's a feature they want, so I can point them in the right direction.  I'll save most of those features for future blog posts, but I'll add the instructions here in case anyone wants to use a real microscope objective instead of the Raspberry Pi Camera module's lens.

Better resolution, higher magnification, and wider field of view are all possible if you use a "proper" microscope objective.  There are designs on GitHub for a modified optics module that takes a 40mm focal length planoconvex lens and a conventional RMS-threaded microscope objective.  The "Large Stage" version of the microscope is needed to accommodate this (the small stage is not big enough to fit the larger objective into), and it in turn requires a different illumination arm.  The bigger microscope is denoted by an "LS65" in its filename/model number, meaning a wide stage that is 65mm high in total. I try to keep the STL up to date for those who don't have OpenSCAD, but the most up-to-date version will always be the OpenSCAD file; in microscope_parameters.scad you need to set "big_stage=true" to get the large stage version.  Then, open and compile main_body.scad to generate the microscope and illumination_and_rear_foot.scad to make the illumination.

We've settled on a 40mm focal length, 16mm diameter lens for the inside of the tube - this is necessary as otherwise we'd have to use the microscope objective at the wrong working distance.  I get ours from Comar - the 40 PQ 16 works nicely.  You should be able to push-fit the lens, and essentially push fit the objective too, though we usually screw them in to avoid crushing anything.

If you've tried this recipe, or if you'd like us to sell you a custom kit for it, do get in touch!