Sunday, August 30, 2020

Demo and teardown of an X-ray fluorescence gun (measures chemical composition)

I show how an X-ray fluorescence gun works, do a teardown and failure analysis, and discuss a new kind of XRF tech now "available" on eBay.

The XRF data can be transferred from the gun to a PC, and analyzed with I did this for several materials, but ran out of time in this video to show the software usage, which is quite involved. Good overall description of XRF: Electron shell shielding described via roller derby ?! XRF supplies: MightyOhm geiger counter: The two XRF guns in this video were generously donated by one of my viewers. Thank you! The Tribogenics tape-peel XRF gun on eBay: Applied Science on Patreon:


  1. hi Ben, can you share which brand-model of xray tube with us?
    thnx ;)

    1. I wish that I knew. The tube has no manufacturer listed, only handwritten serial numbers or quality control numbers. I've done an extensive image search and couldn't find anything similar either.

  2. Hi Ben

    Ive been thinking about an idea well beyond my ability to give a spin, but i figured it might be something for your backlog of projects if you share my interest in the topic.

    Ive always been fascinated by carbon-carbon and related ceramics, but ive always considered them well out of reach of DIY potential. However, reading a bit on the subject recently, I wondered how hard it would really be. Carbon-SiC actually does seem like it might be managable with none too exotic materials and processes.

    In particular; what if youd mix silicon nanoparticles (these seem quite affordable) into garden-variety PDMS, and wet out carbon tow with that mixture? If you wind such a tow onto a mandrel, say a simple aluminium tube as a proof of concept (or some other sacrificial material), how hard would we need to bake this, and what material properties would be in reach?

    A slow heat to around 500c should bake off the volatiles (oxygen and hydrogen; apparently baking out PDMS volatizes away some silicon too, and there are more suitable polymers known. ‘PCS’; essentially pre-baked PDMS, but at first glance I havnt found that very sourceable). In any case a low temp baking should leave behind amorphous SiC in a probably quite friable green state. The next question would be how hot youd need to get that for the silicon to start getting mobile and forming SiC phases. Id imagine youd ‘only’ need 1300C or so for that . Then you might be able to go higher for ‘proper’ SiC sintering temperatures, but I don't know to what extent that would be necessary. Perhaps mixing in some carbon black or other finely divided carbon source might help in getting the correct matrix ratios.

    Probably this does not give you the most dense ceramic (youd bake off a good percentage of your PDMS); and perhaps in terms of properties it wont be the best. But how far can you get, using reasonably prices materials and tools, is an interesting question I think. With a SiC matrix this stuff should be very chemically resistant as well also to oxidative stress. Could bake on some glaze if porosity is a concern. Would be great for DIY furnace components, rocket nozzles, or other crazy things (im thinking DIY ramjet; maybe one day…).

    I have not seen this exact process described anywhere; this particular recipe is born from DIY practicality concerns rather than a knowledge of how to, or expectation of reaching record-setting material properties. But it seems well without reach for a guy with your skills and equipment, and while I suppose there are many things that can go wrong, id give it a good chance of yielding a very unique and useful set of material properties.

    In any case; I thought id spam that random thought here, with apologies for the off topic rambling.


  3. Hi,

    Can you do a video about AR Coating for optics?

    1. AR coating process is very expensive and complex, so you need:
      Ultra high vacuum chamber + turbo pump with e-beam deposition aperture (plashma vapour also uses but not ideal for this) because you must to use metal fluorides not only metal eg depends how reflective and which wavelenght... ;)