• EEVblog #716 – Raspberry Pi 2 Xenon Flash Problem Explained


    Dave investigates and explains the Raspberry Pi 2 Xenon flash problem. Where the Pi2 will reset and lockup when a photo is taken of it from a Xenon flash camera. How and why the photoelectric effect is responsible.
    UPDATE: I have tried a UV filter in front of the flash and the problem remains.

    Forum HERE

    Be Sociable, Share!

      About EEVblog

      Check Also

      EEVblog #1040 – Caesium Beam Frequency Standards

      A look at the Caesium beam time and frequency standards at the Keysight standards lab ...

      • Nice catch Dave. Don’t forget: Blutak is blue (duh) and will likely transmit lots of blue light. Xenon flash tubes emit a large proportion of their light in the blue end of the spectrum.

        • RJ

          It’s blue because it reflects blue light. It absorbs other wavelengths. As such, I’d expect it’s ‘transparency’ to blue light will be lower than that of other wavelengths.

      • hdavis

        Dave you made an oopsie. Wavelength is inversely proportional to the frequency. The near IR (>800 nm) has less energy than the visible light. The spectrum below 400 nm which is the ultraviolet has more energy and that’s probably what is causing the issue.

        • No oopsie, the graph is in nm and didn’t mean to imply IR had more energy, it’s just that IR spikes are there.
          Yes, the UV will also contribute and will have more energy, that is fixed in annotation.
          BTW, I tried a UV filter and it made no difference.

      • Just mirroring my comment from youtube:

        By an insane coincidence I was just checking out the work functions (http://en.wikipedia.org/wiki/Work_function ) of several elements yesterday , some people might be a bit confused as to why the UV light doesn’t appear to be responsible for triggering the effect…and why the peak in the IR range might matter…and I think I’ve got the explanation…the thing here is that the Photoelectric effect as is normally taught in physics 101 tends to cover only the situation where you’re trying to expel electrons from the surface of crystals of pure elements …

        So if we were to calculate the minimum frequency for a photon to remove electrons from , say…pure silicon (best case work function 4.60eV ) we’d reach something like 270nm (UV) .. so yeah … no peak of light, no matter how strong (amplitude) over 270nm (towards visible light) should be able to extract electrons from the surface of a pure silicon crystal…

        BUT! (skipping to microelectronics 101 (taking my trusty Sedra/Smith from the bookshelf!)

        We’re talking about a semiconductor! there’s no need to overcome the work function and extract electrons from the material , we just need to overcome the bandgap (This is equivalent to the energy required to free an outer shell electron from its orbit about the nucleus to become a mobile charge carrier -wikipedia) , and this takes FAR LESS energy (http://en.wikipedia.org/wiki/List_of_semiconductor_materials) !
        Bandgap for silicon would be 1.12eV (I’m using Si as an example… it might not be Si but the orders of magnitude are mostly the same for most semi (check table above) ) would put the needed wavelength at 1.1micrometers (IR)

        So if we fire enough photons of shorter or equal wavelength than say 1.1 micrometers (again,Si only an example) to overcome the bandgap we should make something that shouldn’t be conducting at that moment temporary into a conductor which is more than enough to cause all sorts of craziness in there (current should be proportional to amount of photons (amplitude))

        Also a Flash Lamp emits something in the order of magnitude of a MILLION lumens (http://www.wolframalpha.com/input/?i=1000000lumen) over a few micro-seconds and lumens are only measured in the visible light range! …so according to the spectrum graph you’ve shown the peak in IR would be even greater …that’s a LOT of IR photons which are probably over the bandgap energy and are probably making a semiconductor into a “short” (whose conductivity is proportional to the peak amplitude of the light)…

        What do you guys think?

        • Correct. Anything from IR onwards will contribute due to the bandgap. So that includes IR, visible light, and UV. You don’t know which one (or a combination) without doing proper filter tests.
          The issue here I think is that the Xenon flash produces insane levels of intensity for very short periods.

          • Exactly!
            My bet is that in the Xenon flash scenario the UV component is having a far lesser impact than is being attributed (because people forget about bandgap and only remember the work functions…)

            Say you don’t happen to have a IR-Pass filter lying around? or some developed negatives (https://www.youtube.com/watch?v=aL3FKil6yj0 I know…I know who has these nowadays =) ) to try it out?

            If I had a RPIv2 here I’d test this out right now…

            Update : Found a very suspiciously similar IC on the Rpi B+ V1.2 (grabbing my camera to test it 😉 )

      • Ivan Berton

        Yep, nerdy mindfapper here again…I can`t resist….

        To contribute to the discussion, here a bit of info about the nature of light, for those which are not familiar with it.

        So far, we have two models to describe light. The one that describes light as a wave and the other which describes light as particle (photon).

        There mus be a connection between the two. It can`t be that two models describes the same physical phenomenon and have nothing to do to each other.

        The connection happens with E=hf (h multiplied with f). Where, like we saw in the video, E corresponds to the Energy of one photon and f corresponds to the frequency of the light. h is the Planck constant which you can see as punch tool able to determinate the smallest possible amount of energy(photon) of the corresponding light. So, the higher the frequency(the lower the wavelength) the higher the energy of the photon.

        If we have a lot of photons means not that the we have more Energy in the light!

        It means that we are able to affect much more electrons in the material hit by the light due to the photoelectric effect, resulting in a much higher amount of free electrons which makes possible a higher current.

        There are also different types of interacting between light and material.

        From “nothing happens”, because of the too low energy of the photons. The photons are just distracted after hitting a electron.

        If we increase the energy of the photons, absorption occurs. Photons will be absorbed by the electrons causing the electron to move to another orbit of the atom and also gives a momentum to the atom and if this happens in the right direction we can cool down gas, see laser cooling. The electrons jumps back causing a random direction of the momentum which cancels out in the average. Only the momentum caused by the light source will act to the atom and slows down his motion–>cool down.

        If we go on increasing the energy of the photons we are able to shoot electrons to hell, away from the atoms orbit, usually known as ionization.

        Do we have intermediate states between these three examples? Check out by yourself. Fascinating microscopy.

        Sincerely your background nerd.

      • יעל

        Buy Raspberry Pi 2 with 30$ in ebay deal -> http://po.st/zocdS1

      • Ivan Berton

        What`s about the Indiana Jones scale model train set? Is Sagan not impatient?

      • Syd Jessop

        We need a CSP filler exactly as Dave suggests. I cannot find one available in the UK. Loctite 3536 looks ideal – anybody sell it? What is anybody using in the UK for CSP underfill and where do you buy it, please?

      • Michael Buckley

        Back in the mid 70’s we were stobing a printer to capture its movements. After a while the printer stopped working. We found that the EPROM was being corrupted by the strobe entering the quartz window used to erase the device. Good times!

        • Joshua E. Hrouda

          The EPROM should have had a cover over its window. I hate it when I see them open, and installed!!! Very bad practice!

      • tlhIngan

        Just FYI, most modern semiconductors do use high-impedance inputs – they’re typically based on CMOS technology, and the reason is that MOSFETs used in CMOS have a very high impedance (because of the gate insulator).

        So a tiny current is enough to disrupt the operations of the switching regulator temporarily until the excess charge has drained due to operations elsewhere in the circuit. This is desirable as it means the regulator itself consumes very little energy (given it’s use in portable devices).

        Of course, the end result is it doesn’t matter what is causing the regulator to fail temporarily – the excess charge generated causes the regulator to go out of spec because the charge is causing transistors to switch when they aren’t supposed to. And given those transistors are high impedance, well, it doesn’t take much.

        As for why the Pi doesn’t recover, well, it’s possible the POR )power on reset) circuit wasn’t triggered – POR is normally hanging on a different voltage rail, while the output of the POR circuit is an open-drain so able to pull nRST at any voltage down. (Devices that can assert nRST normally are open-drain and never push-pull as there can be multiple reasons to reset). Because only the 1.2V rail is affected, the POR circuit doesn’t necessarily trigger because the other voltage rail it’s attached to hasn’t dipped to cause a low-voltage dropout reset. I.e., the POR circuit is tied to say, 3.3V or 5V, so it’ll assert the output when those rails dip below a threshold. But since those rails didn’t drop, the POR never triggers to reset the board. The CPu locks up because the 200mV dip causes either internal transistors to latch up, or computations to fail that cause the circuits to do the wrong thing.

      • I do not have one to test, but what if you covered it with something akin to the liquid electrical tape available from most hobby stores.

      • Pingback: Raspberry Pi 2 Released - The RadioReference.com Forums()

      • It is probably the peak luminous intensity, rather than the spectral distribution, that allows a xenon flash to trigger this effect when continuous lights won’t. Even a small Xenon flash outputs a surprisingly high maximum power. A small flash may put out a total energy of 1 watt-second or less, but it dumps that energy in less than 0.1 millisecond, so the peak power is well into the kilowatts. You would need very big continuous lights to match the peak luminous intensity of even a tiny xenon flash like the one Dave is using.

      • Pingback: Raspberry Pi 2 Model B Update: Xenon Flash Related Failures – Open Source Hardware Reviews()

      The EEVblog Store generally ships twice a week, on Tuesdays & Fridays, Sydney time. Dismiss