A Vacuum Tube And Barbecue Lighter X-Ray Generator

A certain subset of readers will remember a time when common knowledge held that sitting too close to the TV put you in mortal peril. We were warned to stay at least six feet back to avoid the X-rays supposedly pouring forth from the screen. Nobody but our moms believed it, so there we sat, transfixed and mere inches from the Radiation King, working on our tans as we caught up on the latest cartoons. We all grew up mostly OK, so it must have been a hoax.

Or was it? It turns out that getting X-rays from vacuum tubes is possible, at least if this barbecue lighter turned X-ray machine is legit. [GH] built it after playing with some 6J1 rectifier tubes and a 20-kV power supply yanked from an old TV, specifically to generate X-rays. It turned out that applying current between the filament and the plate made a Geiger counter click, so to simplify the build, the big power supply was replaced with the piezoelectric guts from a lighter. That worked too, but not for long — the tube was acting as a capacitor, storing up charge each time the trigger on the lighter was pulled, eventually discharging through and destroying the crystal. A high-voltage diode from a microwave oven in series with the crystal as a snubber fixed the problem, and now X-rays are as easy as lighting a grill.

We have to say we’re a wee bit skeptical here, and would love to see a video of a test. But the principle is sound, and if it works it’d be a great way to test all those homebrew Geiger counters we’ve featured, like this tiny battery-powered one, or this one based on the venerable 555 timer chip.

55 thoughts on “A Vacuum Tube And Barbecue Lighter X-Ray Generator

    1. Hi – original author of the BBQ lighter post here. Unfortunately I don’t think X-ray imaging with this device is possible. The output is way too low (I can pick it up on counters consistently, but barely – it’s only a few photons reaching the inside of the detector tube per trigger pull). But if you hooked up this vacuum tube to an actual DC high voltage supply over 15kV (which I did with a TV power supply), you could definitely produce enough X-rays to expose film. Whether or not those 15-20 keV X-rays are penetrating enough to actually get through objects and work as an “X-ray image” is debatable.

  1. “so it must have been a hoax. Or was it?”

    No it wasn’t an hoax but there were no danger either because CRT 1 inch tick lead front glass was blocking most of it.
    X-ray are produce any time high velocity electons hit a metal target. In color TV with CRT the anode voltage was around 30KV enough kinetic energy to generate X-RAY when electrons where hitting the shadow mask.

    1. I should have mentionned that mammograph (x-ray machine used to detect breast cancer) use x-ray tubes with anode voltage around 30KV same voltage that was used in color TV CRT. This should tell you that yes those CRT were emetting X-RAY.

        1. Possibly because the corrupt, cause you cancer, then offer you life time therapy, drugs, medical industry of the west, like the irony of causing you the exact cancer they claim to be screening you for.

          For what it is worth though, CRT were clearly safe to use long term as I used to play video games on my old pentium 4 pc, sat infront a 17″ CRT for upto 12 hours a day (not just games but anime, film etc).
          I have had a mri scan a few years back over a neck injury and there was nothing wrong with me having tumours etc thank God.

          1. They are harder to shield because they interact less (have a lower probability of) with the shied…however the same goes for what you’re shielding.

        2. The operating voltage for mammography units ranges from 20-34 kV for conventional units using a molybdenum target. Newer mammography units using tungsten targets typically operate between 26-38 kV.

          Regular diagnostic x-ray imaging are operated anywhere from 50 kV (for extremity imaging i.e. hands) up to 140 kV for chest imaging.

        3. You can dial them up and down depending on the tissue type and thickness. If you increase energy the contrast will go down for stuff that is nearly identical – as opposed to broken bones or a kid who swallowed a marble.

      1. My understanding it was the high voltage rectifier that create soft X-rays, not the CRT. Though I suspect the electrons hitting the mask in a color TV could also create X-rays.

        1. Fairly high-velocity electrons hitting things are a source of X-rays. Old HV rectifiers were vacuum tubes, and thus had fairly high-velocity electrons hitting metal. So long as those rectifiers lacked the shielding of the CRT, they were likely to be a larger source of X-ray emissions than the CRT, even if the CRT produced more total X-rays. This could be mitigated by placing the diode-tubes inside the same envelope as the CRT, thus interchanging serviceability for emissions safety (note that a modern design would likely use HV semiconductors instead, which themselves seem to either survive to rectify the next cycle of the oscillation, or potentially emit plasma-generated X-rays, but not both).

          All of this assumes the presence of an HV rectifier though. With the use of the common flyback topology, the HV rectifier can, with the right circuit, be replaced with a LV high-current rectifier + the CRT’s own rectification abilities (though it is fairly common to include voltage doublers or triplers on the transformer output), leaving the lower-voltage line-rectifiers as the primary concern. Note that you’ll be looking at either a capacitive CRT (mostly oscilliscopes), or higher-amperage deflection coils.

          Regardless, whether any HV rectifiers are a meaningful X-ray source or not will be dependent on the individual circuit design.

        1. It is funny, but I’d always say attenuate, rather than absorb. Maybe I’ve been talking with too many hams, they have started to influence my thinking.

          You could have a 10 mile thick slab of lead shielding in front and still a few would make through, the number would be infinitesimally negligible and nearly impossible to measure. But there are statistically outliers. So I never say absorbed.

          1. Eh, both are correct. Attenuate focuses on what happens to the receiver, and absorb focuses (at least somewhat, since I assume there’s some reflection and refraction as well) on where the “missing” signal strength goes. 6 of one, half-dozen of the other.

        2. Most any type of glass tube, CRT or otherwise, has a good bit of lead mixed in with it. It lowers the melting point and makes it more gooey, thereby making it much easier to ‘blow’ and form and manipulate. Same is true of lead crystal glasses, and those little figurines, most glass bottles, laboratory glassware, and neon. But there is a massive difference in the thickness of a CRT face and a rectifier tube, so while a CRT blocks most of the X-rays, a radio tube will block very little.

      2. Difference between TVs and medical or scientific Xray generators is not the voltage but the current. Current is responsible for the amount of radiation emitted.

        1. Voltage determine the energy of each x-ray photon. E=hc/w where h is Plank constant, c is speed of light and w is photon wavelength. The current determine the number of x-ray photons emitted. Otherway said, more electrons hitting the anode means more x-ray photons emitted but the energy of each photon is related to the speed of electrons hitting the target as faster electrons have more kinetic energy. The speed of electrons depend itself on the difference in voltage between the cathode and anode.
          When taking radiographies the technologist adjust voltage and current according to the need.
          And there is many other difference between a CRT and an X-RAY tube. The anode of X-RAY tube is made of tungsten and rotate at 10,000 RPM otherwise the anode would melt.

    2. IME we were told not to arrange desks at work such that people were sat behind others’ big CRT monitors, as the backs were less protected, obviously not having glass.

      1. dcfusor2015 has a good rebutal below. In short: not in any credible amounts, for any affordable models, available for any extended period of time. It was a simple engineering problem, and so got caught and fixed quickly.

    1. I just did some checking online, because 10KeV X-rays are highly attenuated by most materials.

      I can only estimate the thickness of tube glass, but 10KeV X-rays are attenuated 99.9502% by 2mm of glass. This doesn’t account for attenuation by air between the tube and measuring device, which is also significant at the 10KeV range.

      I’d like to see some independent confirmation that this system is actually generating X-rays. At the present time, I’m highly skeptical.

      No, scratch that: I don’t believe it’s doing what the author claims it’s doing.

      Needs more proof.

      1. Hi – Original author of the BBQ lighter post here. Thanks for your healthy skepticism – I’d like to clear up a few things and provide a little more proof that this device is producing X-rays, as I should have done when I made the post in the
        First off, I’ve just now made a quick video demonstrating my CDV-700 geiger counter reacting to the produced X-rays (Youtube link: https://www.youtube.com/watch?v=Uq_j0TWAbZs&feature=youtu.be). As you can see in the clip, the meter moves just a tiny bit when the trigger is pulled – only a few photons are reaching the detector per trigger pull. I’ve also had success on my SBM-20 based geiger counter, but that one is a little weird so I won’t use it as proof – the CDV-700 is a much more reliable device. Unfortunately the video doesn’t show the final version of the device showcased in the post because I broke it today after excessive abuse of the trigger from me and others pulling it many times way too quickly. I took the tube/diode off and attached it to another lighter (will solder it back up tomorrow) to make the demonstration video, so the circuitry is identical.

        10 keV is a number I typically use as the threshold for what can make it through vacuum tube walls to the geiger counter. It is by a no means a scientific number – it’s just based on what I’ve seen online and in my own experiences, and could be inaccurate. Regardless, whatever maximum energy is produced by the lighter-tube assembly is enough to escape the glass, as I can detect it on multiple devices. I have not been able to measure the actual peak voltage produced by the lighter, but I do know that I can easily detect large amounts of x-rays from the TV power supply, which I have measured at 20,000 volts. Only a tiny fraction of the x-rays needs to get through the vacuum tube walls in order to explain the <5 photons per trigger pull I see on my detectors.

        As for electromagnetic interference, I have never once been able to set off any of my radiation detectors via electromagnetic effects (excluding gamma/x-rays of course), despite many attempts (HV arcs, spark gap coils, high frequency stuff, etc). For that reason I have no reason to believe that this is EM interference rather than X-rays, especially since I can explain why X-rays would be produced but not why a large electromagnetic radiation pulse would be.

        In short, as far as I know, the device is producing detectable X-rays that can make it to the inside of the geiger-muller tube. It's almost too few to even detect, but they seem to be there. Soon I will have access to a solid-state photodiode detector and I will again attempt to measure the X-rays produced here – if that goes off, I think electromagnetic interference can be safely ruled out for good. If you still believe that there is something other than X-ray radiation going on here, feel free to let me know and I'll look into it, as I have no intentions of publishing a device that doesn't work as intended.

        1. Used to do this ( when I was young 13 or so) with car ignition coils. Got some boron to sparkle. Had to push the voltage on the coils. Managed to expose film as well, but was so scattered that a good xray picture was not going to happen.
          The short of it is I guess is I beleave you could do this with a igniter. Wern’t the first xrays made useing a Whimshurst generator?

  2. I was going to ask what Name answered. Not sure if x-rays and gamma rays are close enough to trigger a geiger counter. The guy who built this must have some application in mind, other than scaring people with his x-ray gun.

    1. I can’t see 10kV getting to the Geiger tube. At higher energies the x and gamma naming is rather fluid. You can buy a 1MeV industrial X-Ray tube. Yet the 510kV photon from positron annihilation (from say, Na22) is invariably called a gamma ray.

      It is all here “The Scientific American Book of Projects for the Amateur Scientist” in the project to build an x-ray machine. Great book and you can download in several formats now here https://archive.org/details/TheAmateurScientist . Everyone in hacking should read this. It is nearly 100% hacks and maybe it will reduce modern snowflake response to vacuum, mercury, radiation, and essentially everything done behind a Mythbusters blast shield.

  3. Vacuum tubes for larger voltages (about 10 kV?) have warnings that they emit X-rays. I have some C1139/2 tubes (not sure if I remember correctly the type) and they have such warning. But I think I won’t be making an X-ray machine from them ;)

  4. Did something similar a while back using a spark igniter module from a gas oven (9V version) and a 5642. I found that a DAC32 put out a ridiculous amount of X-rays which actually made the GM counters click 2 feet away and a scintillator screen light up dimly through a black project box clearly showing objects between it and the tube. Yes the tube also glowed green when “flashed” but alas it didn’t last very long as the module failed eventually as it wasn’t designed to run from 20+ V.

    It would be interesting to try this experiment again but this time use a proper pulsed supply with calibrated output,

  5. Skeptical.
    1) X-rays are non-ionizing radiation, which is not detected by Geiger counters.
    2) Piezo igniters only produce a few kV, not enough for detectable X-rays, which need >> 10 kV.
    3) Microwave oven rectifiers are only good for about 2 kV, again well below X-ray energy.

    I think it’s most likely the Geiger counter was picking up an electromagnetic pulse from the igniter. The acid test is a fluorescent screen or piece of photographic film – these detect X-rays but not EMP.

    1. Sorry BBJ, you’re so far off I have to comment in case you lead others astray.
      https://en.wikipedia.org/wiki/X-ray Read this first.

      X rays ARE ionizing radiation, anything over a few tens of eV is. Hydrogen is fully ionized at 13.6 eV, roughly.
      Fluorescent screens DO detect electrons hitting them, as well as photons. How do you think TV’s even worked? Or scopes?
      Film ditto – see anything about early cosmic ray work. Or Rutherford’s stuff. Old news.

      See how many kV it takes to jump a spark a given length. Just because the gap in most piezo igniters is limited to ~ 1/4″ doesn’t mean it won’t make more – enough to sel-destruct was explicitly mentioned.

      This stuff isn’t for people who don’t get it. I’m against bubble wrapping the world, not everything is dangerous.
      But this can be, so you should know enough to calculate a risk if you’re going to take it.

      The reason we didn’t die from TV’s is it was a diddly amount of X rays. At most, a TV supply would make a single digit number of mA of current (no one seems to remember it’s the number of electrons that limits how many photons you can have – the volts jolt, but the mils kill). Now look at those x ray machines at the doctor’s and their current ratings.

      That said, due to low current and low duty cycle, this thing isn’t going to make enough X rays to be a worry.
      Sadly, they also won’t be point source and therefore lousy for making a film exposure even if you click it enough times.

  6. X-rays from TVs only became an issue with the larger color screens as the HV went to 35kv and beyond. Once noticed, manufacturers more or less instantly went to lead glass to attenuate them (which is the correct term, see Compton scattering in any Nuke physics 101 book).

    In the very early (maybe one or two production runs) super HV TVs, a shunt regulator triode, for example 6BK4 was used, but this was quickly replaced with more intelligent regulation at the drive end of the flyback transformer. That would have been the *only possible* source of X rays from the “rectifier section” as when forward biased, the 3A3 (or other similar) tube only had ~~ 1-200v across it, and when reverse biased, had next to no electrons flowing at all (see – rectifier).
    Even then, manufs like RCA didn’t quite have anal-cranial inversion – the plates of the tubes were made of lower atomic number metals (rather than say, tantalum) so as to be less efficient at making X rays (see, K,L,M lines and so on again in any nuke book).

    Very few (on the scale of things) TVs made noticeable X rays at all. There’s a ton of urban myth of course, but a small amount of critical thinking combined with a little knowledge of how things work – something one might wish to assume in this forum, will tell you that all that myth was…myth. Only a few early adopters were exposed. Normal income families (one job, one house, two cars) as I grew up in – not an issue. A big screen color TV with that problem that new on the market was far out of reach in those times for most. My parents didn’t even get us a TV, a B/W one, till somewhat before Apollo 11 as they already figured out being raised by TV was going to be stupid. By then I was a teen and thinking of the mating ritual more than the soaps.

    Yes, you can likely get X rays out of this thing, probably not a super amount, depending on your definition of big.\
    Lower energy ones are easier to stop by far, so yes, it’s the energy (in increments above chemical bond strength) you absorb that messes up DNA (cell phones are sub eV so not enough to ionize anything and break bonds, for example).
    Energy that goes on through has no effect, and cross section matters (again look it up instead of spouting myth, this stuff is well known and not disputed), but higher energy photons do also leave some energy and damage behind too.

    You might guess from my handle I do some work in the area.

  7. Creative, but I suspect the geiger counter is responding to spurious fields from the wires and not x-rays. I expect x-rays are not being produced as commercial x-ray tubes require a heated cathode for electrons to be emitted. (The tube should be glowing.). Also, a positively charged anode is used. I see neither here.

    1. Hello – original poster here. As I mentioned in another comment, I have never been able to set off any of my detectors using electromagnetic interference (other than x-ray themselves of course as these are electromagnetic), so I have no reason to believe that is what is happening here.

      As for the cathode, it is possible to obtain a small current without heating if you apply a very high voltage across the tube via something called field electron emission (https://en.wikipedia.org/wiki/Field_electron_emission). Essentially, applying a high voltage produces a strong electric field that allows some electrons to tunnel out of the cathode and accelerate into the anode. This is what I use for all of my X-ray experiments, because heating the cathode would produce so much current that it would likely load whatever power supply I was using and potentially decrease the x-ray output. In either case, it’s not needed here as the piezoelectric crystal isn’t able to supply enough current to make use of all of those extra electrons if I had heated the cathode.

      As for the anode, you are right – a positive anode is necessary for x-rays. I have set up this device so that the output of the microwave diode feeds into the plate/anode of the tube, so the AC output of the crystal is rectified and the anode does receive the positive voltage relative to the filament (the diode wouldn’t be necessary if the crystal didn’t self destruct via charging up the tube).

  8. Interesting you mention that. I found that it could be internal arc-through but this does not explain everything.
    Random: if you use a gas igniter with a PP3 instead it works well.
    Also be careful as very large X-ray fluxes can result.

  9. Also relevant: some small vacuum tubes used for display applications can be “Juiced” into the X-ray emitting region by zapping part of one grid with a near UV pulsed laser. It only works if the spot is very small though so you need to be precise with construction.
    This is similar to how 6VS-1 tubes emit X-rays but in this case the high voltage supply can be replaced with a safer lower voltage unit as the laser supplies most of the acceleration energy.
    Note that the hazards of laser light are well known and pulses are even more dangerous but careful construction may reduce risk.
    For added coolness, said 6VS-1 can then be used as a controllable X-ray pulse source at a well defined energy determined by the various voltages on each grid and the final acceleration voltage.

  10. I’m sure you can’t measure X-rays made with that because the EMF produced by the gas lighter will false readings on any Geiger counter. To read X-rays you should use a professional dosimeter shielded for EMF in use in nuclear industry.
    As far as I tried no X-rays can be produced with that. You’ll be more lucky with a wimshurst machine.

  11. Also useful: found that some early car bulbs especially if used have enough tungsten deposited on the inside to act as a secondary electrode increasing glass coupling if the right kind of foil is used.
    If the vacuum is moderately good then X-rays are possible.
    It won’t be as good as a regular tube or even something like a diode rectifier with a large gap NOT having leaded glass such as ones used in portable scopes of the time (cough 5642 /cough) but it will work.
    Onlyu sharing here because someone may accidentally injure themselves without intending to because the resulting X-rays are in the low energy region that isn’t detectable by most counters and normally shielded by dental setups etc.
    Disd I mention that this method is effective for temporarily fixing “unwipeable” microSD cards and chips on N4 smart phones with the infamous wear leveling issue.
    Bidding on some broken SSDs to see if I can repair them this way for storing write once data only.
    If you are messing with this stuff be warned: you are dealing with something that can potentially cause serious harm and should be shielded properly when in operation ideally with multiple layers of shielding, in another room by remote.
    Sourced (haha) some actual imaging sensors and going to test shortly.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.