Instructables

How to Restore, Improve, and Digitize an old microscope

FeaturedContest Winner
Picture of How to Restore, Improve, and Digitize an old microscope
This microscope was discarded because the lighting mechanism it came with had stopped working entirely.  Changing the light bulb did not fix the problem, and because the whole electrical mechanism had been epoxied into a solid slab during the manufacturing process, it was impossible to see which individual component had gone bad.  The rest of it, however, worked just fine.  The optics were in good shape, all of the dimensional stages worked, just no light. Because the microscope was manufactured in the 80’s in West Germany, it was unlikely that finding a replacement to the whole lighting unit would be possible or cost effective.  Not to mention that, when it had worked, it used a ton of power and expensive, extremely hot, bulbs that tended to burn out anytime the microscope was left on overnight. So I got rid of the whole lighting mechanism, which was archaic, inefficient, wired and heavy.  I replaced it with a simple housing that allowed me to plug in an LED light and power it from a cheap 3.3V coin cell battery I had lying around. 

With the lighting situation fixed, I turned to improving the current setup.  The optics worked just fine, but I wanted to be able to take pictures of what I was seeing, and while I’m doing that, why not just turn it into a digital microscope and not even bother straining your eyes looking through the eyepiece.  I had a high quality webcam lying around and liked that it has some of the best light-correction I have seen on a webcam.  But as anyone who has ever tried to line up a camera with a lens before can tell you, even the slightest misalignment produces blurred, unusable pictures.  So I designed and and printed an adapter to solve this issue.
 
Remove these adsRemove these ads by Signing Up
Light_Lab9 months ago
Wonderful job, I am particularly impressed how well the 3D printed camera adapter came out, very good finish.
You can improve the depth of field by using image stacking, there is a free program available:
http://myrmecos.net/2009/12/05/combinezp-stack-your-images-for-free/

We have used CombineZ in the light-lab and it compares very well with expensive commercial systems. It does require care to avoid X,Y movement and considerable patience to take enough photos at different focal points.
Branebot (author)  Light_Lab9 months ago
Thanks Light_Lab! I really appreciate your feedback and the info. I knew there was expensive image stacking software out there, but had no idea there was a freeware version. I will absolutely be incorporating this into my project. THIS is the beauty of instructables, everyone teaching each other and the ability to draw from a community pool of knowledge. Love it!
Great Project! Gives me some ideas to try out. Concerning the biology though, I would consider it very unlikely that a fly caries around its larvae. They usually lay the eggs in some food containing substance. Much rather I would think these were parasites eating their host from within. Even closer to the Alien movie than you thought...
Branebot (author)  bluegreenberry8 months ago
Thanks! and thank you for your comments. It is true that most flies deposit their eggs on whatever "meal" their young are to be eating as new larvae.  The eggs kinda look like tiny 1mm diameter caviar.  However there is a family of flies known as Sarcophagidae that have the interesting feature of ovolarviparous, which is essentially letting the eggs hatch while still inside the mother and then hoping they find something delicious to deposit them on shortly thereafter...
I would only consider myself an amateur entomologist at this point (and even that is rather generous) but from what I could tell of the pictures I have found online of both the fly and larvae, it does appear to be from the Sarcophagidae family, who bear the unfortunate colloquial name of "flesh fly." 
Most parasites that are laid into flies, such as the larvae of parasitic wasps, only emerge when they are adults or very nearly so, and look quite different.
So I'm still pretty sure those were the larvae of the fly. If you find any evidence to the contrary please post for my, and future readers, enlightenment.
alexkhvorov9 months ago
What an exellent idea!
Is it possible to plug second camera in order to get a 3D imagery - to 3D screen&glasses or Occulus Rift?
Branebot (author)  alexkhvorov9 months ago
Thanks! Unfortunately adding an extra camera to the other eyepiece would not result in a parallax, which is what is required for 3d imaging between two cameras. This is because the picture is actually split from a single light path coming through the objective, and thus the extra camera would see the exact same picture and no depth data could be computed from it. Splitting the image is done so it would be more comfortable for human eyes to stare through the eyepieces for a long time.
It does appear to be possible to generate 3D imagery using image stacking the way Light_lab mentioned, and I fully intend to experiment with this as soon as I can get some time to do so.  If and when I get it up and running, I will post an update.   This will also coincide with my rebuilding the lighting system, as I want to have more control and options (i.e. RGB).  If it turns out being a huge deal, or very complex, I'll make a separate instructable on 3D imaging through the microscope.
Nobin9 months ago
Uma maravilha, beleza de projeto
Wonderful , nice job Thanks
A word of caution about converting a transmitted-light microscope to LED illumination. If originally fitted with a halogen bulb, there may be a filter (or a place for one) to remove infra-red light. White LEDs often have a large blue spike (hint: it's a very good idea to investigate the LED of interest first). Blue light is the most energetic part of the visible light spectrum. Use of blue or blue light for bright-field microscopy, without a filter to tone-down the blue-end is potentially hazardous to eyesight. Dark-field microscopy is probably much less hazardous. Dichroic filters are available to remove selected portions of the spectrum, normally expensive, but bargains may be had from eBay. I recommend that anyone considering this should investigate 'blue light hazard' and decide for themselves. LED technology is changing, high CRI White LEDs (CRI is color rendering index) are increasingly available and are beneficial to photomicrography and eye-safety. If one wishes to drive the LED from the mains, LED drivers are available.

Remember LEDs are DC devices and need a constant current drive for maximum life.
Enjoy your microscopes!
The led blue problem is all about brightness.
Is it uncomfortably bright to look at, at low mags?
It no worse than sun light.

I suggest putting a variable resistor / pot in series with the led
for a brightness control / dimmer.
Thisll also prolong your battery life when not needing full brightness.
The led blue problem is all about brightness.
I'm afraid that is incorrect. The blue light hazard relates to the short wavelength = high energy combined with the intensity of blue light emitted.

White LEDs typically have a large blue spike (this is often hinted at by the blueish tint to the light), visible in the example spectrum posted previously. High CRI types (e.g. CRI>90) are characterised by a much reduced blue output and are considered much less hazardous to eyesight.

There is extensive research about this, it is a very real hazard.
Cree is a well-respected manufacturer and responsibly  issue this warning, but I am confident it applies equally to all  manufacturers of LED products. I am not singling-out CREE for criticism, indeed, I posses a number of Cree products and find they function well. The blue light hazard is due to the technology and is not unique to Cree. I anticipate that most if not all LED product manufacturers make products which present a blue light hazard.

"....Cree has engaged an independent lab to conduct photobiological testing, also known as eye safety testing, on its blue, royal blue and select white LED components. The results of this testing (explained below in further detail) show significant health risks from some of Cree’s visible light LED components when viewed without diffusers or secondary optical devices. These risks warrant an advisory notice to indicate the potential for eye injury caused by prolonged viewing of blue light from these devices. To date, the testing shows that Cree’s blue and royal blue LED components (450-485 nm dominant wavelengths) pose a higher potential eye safety hazard than its white LED components. Other colors of LED components, such as green and red LED components, do not pose as significant of an eye safety risk. Regardless of LED color, Cree advises users to not look directly at any operating LED component...." Source

BTW, one should never stare at any bright light.
Branebot (author)  totally_screwed10 months ago
These are very good points, TS, and I appreciate you bringing them up. I'll do a little poking around on these subjects and include a cautionary word at the beginning about using LED conversion for direct visualization. I didn't really think too much about it at the time because I was converting it to digital and wouldn't actually be looking through the eyepiece personally. However, for those considering doing this kind of conversion for their own scope that they might be looking through, such caution is warranted. Thanks for your input!
I have a compound microscope which I have yet to convert from Halogen to LED illumination. It was during the research process that I serendipitously discovered the blue light hazard. I must really get round to it!
This spectrum is for a CREE XP-E P4 Emitter high CRI LED and shows the blue spike for an 'outdoor white' LED versus an 85 or 90 CRI white LED. It's easy to see the huge spike in the blue region of the regular / low CRI LED. In this case, not only has the spike been substantially reduced, but what remains has been shifted to longer wavelengths, meaning less energy content. I am not endorsing CREE LEDs, this is just an illustration for comparison purposes only.
CRI-90 spectrum.jpeg
Light_Lab Branebot10 months ago
I was just about to raise the blue issue and I noticed totally_screwed got in before me.
In fact some older microscopes have a blue filter to 'whiten' the light from tungsten or even halogen blubs. This can really boost that blue peak in white LED spectra to a dangerous or useless level.
Even if you are going to go digital you should be aware that white LEDs produce white by providing a small number of spectral peaks that visually average to white light. Digital cameras break an image into 3 or 4 color channels using color band pass filters over the CCD elements. A lot depends on how the spectral peaks match up with the filters. It is possible to get big disparities between what is seen visually and what is seen by the digital camera when using white LED illumination (similarly fluorescent lighting).
See http://en.wikipedia.org/wiki/Metamerism_%28color%29 
I am very glad totally_screwed mentioned that CRI LEDs are becoming more really available, I would love to know where I have looked in the past. Lately I have been checking microscope suppliers for LED systems to fix my old scope and I was about to buy an expensive system. Now I am going to look around for CRI LEDs again.
Machine9 months ago
Very nice job, very professional. Thanks for showing us.
foobear10 months ago
I picked up a lovely old microscope from Goodwill. However, one of the lenses got wet and I'm told it is ruined. It would be nice to be able to fix it though.
foobear,
I recommend you join some microscopy forums. If you post the details of your microscope and the problem, I suspect somebody might well be able to offer help or point you in the right direction.
Good luck!
ElectroFrank9 months ago
Just a thought: a flat top or strawhat LED might provide a more even illumination.

(Or you could file the top off a standard one.)   The comment about RGB LEDs is good, and a UV LED would cause some things to fluoresce.   Perhaps a carrier that would allow different LEDs to be slid across into position ?   And a brightness control ?
bZirk10 months ago
Thank you and this is wonderful timing. I have an old microscope that one of my kids could use, but the light no longer works. I was going to fix it but not this well. This has inspired me! Thanks again.
Branebot (author)  bZirk10 months ago
Certainly! And I'm glad you found it inspiring. Best of luck with your own refurb. I you end up making it an instructable, please send it my way.
Well done! Your Instructable is excellent! There are many reasonably priced microscopes available that could be put to use as you've shown here. Thanks for your post.
Branebot (author)  Perspective Image10 months ago
Thank you! And these reasonably priced microscopes don't have to be new or fancy either. As long as their optics are in good condition, the rest will fall into place. I think it is a great way to revive and modernize old lab equipment. On top of that, it makes documenting things SO much easier because anything that you see on the screen that you like, you just click the picture button.
jmw1910 months ago
Excellent project! The printed camera adapter is inspired.
This feels like a good application for a high-powered RGB led, something in the 1-watt total power range. With a good controller/power supply, you'd be able to vary the spectrum from pure monochromatic to probably a richer white than the 5mm led. Pure blue may even cause some fluorescence, depending on the led.
Branebot (author)  jmw1910 months ago
Thanks! Adding additional lighting options is the next upgrade I'd like to tackle. Would you happen to know a good supplier of this type of high powered RGB LED? I agree that it would be preferable to the little white one I have in there now.
kintekobo10 months ago
I like the 3D printing idea. It took me months of keeping a look-out in everyone's junk boxes till I found something that I could use to do a similar job. Maybe I have the excuse to invest in a 3D printer after all :-)


Just one caveat which you are probably aware of, but others might no be. If you are planning to use UV light to do fluorescence microscopy always be aware that you must use a UV filter in the optical path from the object to the eyepiece since certain types of UV are extremely harmful and damage the retina.

Branebot (author)  kintekobo10 months ago
I kind of underplayed the 3D printing aspect in the instructable, but to be honest, that was the one aspect of this project that made it viable. I have had plenty of experience in the past trying to line up microscopes, spotting scopes, telescopes, etc with a digital camera, and it was always very difficult. Back then, I didn't have the option to 3D print the adapter and the fabrication was difficult and always sub-par. Having a 3D printer has really opened up a whole new set of possibilities for me. For a maker, I would say that it is an indispensable tool.
You have a good point about the UV light and optical path. Part of the point of my adding the digital camera was to get my eyeballs out of the optical path. So I don't think it would be too much of a problem as long as I use it as a digital microscope.
Graham Leach10 months ago
Superb
padbravo10 months ago
great!... thks for sharing...
destroyer201210 months ago
Very cool project! I may attempt this with my microscope as well, but I have a different webcam that is square in shape...
Branebot (author)  destroyer201210 months ago
Thanks! If you do end up attempting this, please let me know how it goes and if you find any tips or tricks in the course of your journey.
bob303010 months ago
Great job not just repairing but improving the microscope! You are a true "Maker". The "creepy" bug images were icing on the cake! Thanks for posting.
Branebot (author)  bob303010 months ago
And thank you for commenting , sir! I appreciate the complement. I have imaged nearly everything that creeps or crawls around my house and it is astonishing to see some of it so "up close and personal."
ChrysN10 months ago
Nicely done, those old Zeiss microscopes have great optics.
Branebot (author)  ChrysN10 months ago
Thanks!
Indeed they do. Fantastic German engineering all around. I also dismantled the optics housing (not advised) just to see what it looked like, and was very impressed with the precision.
lostbord99910 months ago
Very well dun sir!
Branebot (author)  lostbord99910 months ago
Thank you sir!