Xenon Flashes Can Make New Raspberry Pi 2 Freeze and Reboot 192
An anonymous reader writes Unfortunately for Raspberry Pi 2 owners who are trying to photograph their devices, ... the Raspberry Pi 2 has been found to be Xenon flash sensitive. Any camera with a Xenon flash aimed at the device is causing the device to freeze for a few seconds before rebooting. The forum thread about the bug is an interesting play-by-play of how the problem was narrowed down.
SMPS? (Score:3)
Cool. The linear reg on the previous models worked perfectly, but was rather less than ideally efficient - 5V power in, but almost all the power consumed goes via the 3.3V rail.
The new power supplies may be sensitve to EMP (Score:5, Interesting)
Years ago when I visited an aquarium I encountered a very strange situation
I was in front of a tank which has 3 electric eels, and in front of the tank there was a 'meter' measuring the power the electric eels were discharging
So I took out my camera (real camera, with powerful Xenon flash light module attached)
Before I pressed the button the Xenon flash was charging (as I said, powerful flash light) and all of a sudden the 'electric meter' in front of the tank indicated that there was an electric discharge from the electric eels
At first I thought it was a coincidence. Then I wanted to take another picture. Again, my Xenon flash light module was charging, and again, there was a jump in the 'electric meter' reading. This second time around I started to suspect that there was a connection in between my Xenon flash light module and the electric eels
The third time around I only use the Xenon flash module. Again I hold it close to the tank, and charge it, and again, the 'electric meter' got another 'shock'. I repeated the experiment the fourth time, fifth time, .... every single time while my Xenon flash module was charging up,. the electric eels inside the tank somehow 'felt' something and gave an electric discharge
I never know the exact reason. My suspicion is that there might be some EMP effect, some wave or some magnetic field, or something like that
What I described happened years ago. I never get the chance to test out my theory
Perhaps someone can test if Xenon flash emits EMP, or not
Re:The new power supplies may be sensitve to EMP (Score:4, Interesting)
Back in the earliyish days of cell phones (1994ish) I had a cell phone that would cause my computer speakers to power off about a half second before the phone would ring.
Re: (Score:1)
My cellphone would make any nearby speaker go "chik chika chik chika chik wrrrrr" just before the phone rang.
Re: (Score:2)
Yep, I remember mine (Motorola? Nokia?) did that, too.
Re:The new power supplies may be sensitve to EMP (Score:4, Informative)
They still do this if you're on a GSM network.
Re: (Score:2)
Yep, early motorola and nokia here too.
Re: The new power supplies may be sensitve to EMP (Score:2)
In the US, in medieval times, cell phones used TDMA (IS136) or CDMA (IS95/2000).
TDMA is time-division multiplex, which means the phone radio would turn on and off on a fixed cadence to allow for other phones to share the same frequency channel. At the start of a call (either to or from the cell phone), the cell network still doesn't know how attenuated the phone signal will arrive, so typically the first communication bursts from the phone would be sent with much more power than the subsequent ones, when th
Re: (Score:2)
Re: (Score:2)
Re:The new power supplies may be sensitve to EMP (Score:5, Informative)
That's because you're hearing the pulsed transmission of a TDMA radio technology.
D-AMPS (AT&T pre-Cingular), iDEN (Nextel), and any GSM 2G (up to EDGE) all use/used TDMA to share the frequency, so they're all potential causes of this.
These days you won't hear it much because D-AMPS and iDEN are both dead and most GSM phones will be attempting to connect on 3G UMTS (which uses CDMA) or 4G LTE (OFDMA).
DECT cordless phones are heavily derived from GSM so it's possible that they may be able to cause the same behavior, but due to their significantly reduced range requirements the power probably isn't there. I haven't heard it from my DECT phones.
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
My cell phone still does that if I put powered speakers near it. Pretty much only would be close enough if they are computer speakers on a desk, most other audio setups with powered speakers aren't close enough to the places in the room where cell phones get set down.
If the effect is at all reduced in modern equipment, it is probably just that shielded cables got cheaper, or computer speakers are using smaller wires that pick up less interference.
In the 90s it was common for CB radios from passing cars to "
Re: (Score:3)
It's the pulse rate of the radio. The radio turns on and off at audio frequencies, so the interference smears out to become the classic chirp sound.
Re: The new power supplies may be sensitve to EMP (Score:2)
The radio frequencies are high, but the turning off/on of the radio transmitter, not.
It is that keying of the power amps which caused the audible buzz.
Re: (Score:2)
Re: (Score:2)
And even if it was spewing a huge amount of low frequency components out, good look getting much of that to couple into a audio system a meter away without only a watt of power and a cell phone's antenna
The AC before me said everything I want to say, except for this: clearly that works, since my old TDMA phone used to be able to make my computer's speakers buzz when they were turned off at the power strip they were plugged in to. You can theorize incorrectly all you want, but the best test of real world physics is the real world.
Re: The new power supplies may be sensitve to EMP (Score:2)
Of course it does. Since you mentioned Fourier analysis, do you know what the Fourier transform of a delta function is? It's a constant. Yup, equal power at all frequencies. A knife edge also has power in all frequencies. If you're "turning something on and off" at 100 hz then you have a situation somewhere between a 100 hz square wave, which has power in all frequencies (less possibly a few nodes of destructive interference), and a 100 hz sine wave, with power at only 100 hz. Either way, 100 hz is present.
Re: (Score:2)
IIRC the GSM frame repetition rate was around 400-440 Hz.
Many electronics will, when exposed to RF like this, behave exactly like the legacy "crystal" radios did - these were nothing more than a basic envelope detector (diode + low pass filter) combined with a tuned resonator.
Hit a crystal radio with a lot of local RF (1/R^2 remember?) and it'll receive a "station" it's not tuned to.
Re: (Score:2)
Re: (Score:2)
It's related to the GSM network, if you had a CDMA phone (Verizon, Sprint) you didn't experience the issue.
The problem wasn't only over the radio, it also affected computer speakers and headphones. The expanded frequencies the networks use might help the problem, but I've been on CDMA for the past eight years, so I experienced it for a while.
Re: (Score:3)
I still have on my desk now a modern 3watt UHF digital 2-way which causes a lot of monitors to turn off when I key it up.
There's nothing really about this phenomenon which links it to age. Modern equipment and old equipment can be affected.
Re: (Score:2)
I'm sure that's the case, but I've never had a cell phone since that had that kind of an effect on other systems.
Re: (Score:2)
Because the "systems" have evolved to be tolerant to the common frequencies of interference.
The signal strength has mostly gone up, not down, so without the improvement of the equipment receiving the interference, the problem would be worse now.
Just take a cheap powered computer speaker, remove the speaker wires, replace with the speaker wires from a 90s computer speaker, and place a cell phone next to it. You'll pick up lots of interference just off the unshielded wires.
Home stereo equipment picks up less
Re: (Score:2)
To be fair, most cell phones emit less than a watt of power rather than three.
Re: (Score:2)
I do. Every cellphone I've had has caused interference to something at some point. Typically it has been audio electronics. The worst one was my car radio. I keep my phone (Galaxy S4 if it's of any interest) underneath the radio in a little compartment when I'm driving. I don't have a centre console in my car.
About 3-6 seconds before I get a text message I would hear a very loud clicking noise from the speakers and also lose AM radio reception.
Whenever you introduce something into the airwaves you will inte
Re: (Score:2)
Back in the earliyish days of cell phones (1994ish) I had a cell phone that would cause my computer speakers to power off about a half second before the phone would ring.
Yep, same here
Re: (Score:2)
Re: (Score:2)
That is a feature. Auto speaker mute when you have a call. :)
Re:The new power supplies may be sensitve to EMP (Score:5, Insightful)
Or maybe the glorious eels didn't give a shit about your puny human flash, but your device was interfering with the meter.
Re: (Score:2)
I wouldn't read too much into the eel's behavior. There are too many variables that we can't confirm. They could have faced towards you out of coincidence; movement near you; your movement; changes in the sign; etc, etc, etc...
I think it's interesting that your flash set off the sensor. If you don't mind me asking, where is the aquarium?
I've noticed other odd things in the right conditions. Like there is a state park near here that has an underwater viewing area. My prescription sunglasses are polari
Re: (Score:3)
I think it is more likely that your flash unit was disturbing the sensor than the eels.
Re:The new power supplies may be sensitve to EMP (Score:5, Interesting)
Re: (Score:2, Informative)
I've worked with Nd:YAG lasers that had large xenon flashlamps in them (in principle like a camera's flashlamp, but much bigger, 100s of kW for a sizable fraction of a second), and pretty much any voltage monitoring in the room would pick up spikes from the start and end of the flash, even with a metal cover over the laser. Any large changes in current and unshielded inductance, and you will get some mess of EMI coming off. I doubt it would be enough to damage anything, unless you had a giant antenna feed
Re: (Score:2)
Are you sure the meter was registering the fish reacting to the flash unit, as opposed to simply registering whatever crap the flash was emitting as it was charging?
Re: (Score:2)
Flash of genius? Blinding intellect?
Or, perhaps it was dark and the subject of the photo was not actually in the tank...
Re: (Score:2)
Um.... No...
Technically it was the quick discharge of the capacitors that required a really high current at substantial voltage that jumped though an arc gap in the bulb. The arc and the quickly changing currents where what generated the EMF spike....
However, in the Pi2, the problem was light related. U19 isn't apparently shielded correctly from light and when the bright strobe light hits is it disrupts the 3.3v power supply regulation of the Pi2. When they block the light from hitting just that chip,
Re: (Score:2)
Before anybody corrects me.. It's U16, not U19....
OMG! (Score:2, Funny)
I took a picture of my Raspberry Pi 2, you'll never believe what happened next!
Maybe not the power supply? (Score:3, Insightful)
Reports are saying the power supply is causing this fault.
That might not be the case. Bright UV light will create electron hole pairs in the gate of transistors turning them all *on*, which will cause the chip to use much much more power since push pull output stages of logic gates will now be shorting the power supply.
Hence, even though it looks like the power supply is failing, it could simply be the power supply is turning off due to overcurrent.
Re:Maybe not the power supply? (Score:5, Informative)
Hence, even though it looks like the power supply is failing, it could simply be the power supply is turning off due to overcurrent.
No. Covering the regulator chip solves the problem. That means that it is the culprit.
Re:Maybe not the power supply? (Score:5, Informative)
If that was the case, putting a blob of material on the power supply chip (and nothing else) wouldn't remedy the problem – but it does (see the last post on this page [raspberrypi.org].)
Re: (Score:3)
Interesting. That thread has reached a point where they're trying to confirm that it's just the edges of the chip that are light senstive.
A while back I took apart a scrap HD-DVD player and I noticed black epoxy around the edges of some chips. I thought it was just an attempt to prevent hacking the player, but I think those were the same type of flip-chip packaging, with nothing but mirror silicon on top.
Also I seem to recall that CPUs and other chips with a mirror silicon chip in the middle always have t
Re: (Score:3)
I heard that if you color the edges of the chip with a blue sharpie, then your RPi mp3 player will have cleaner sound. /s
I was recently working with some LED lighting modules with the bottom of the PCB exposed (and holding the LEDs) and the constant current controller had a big blob of epoxy on it. The other model where the LEDs were mounted on a different board, with the controller inside a metal housing, didn't have that.
Re: (Score:2)
I don't know about cleaner sound, but if you use a red sharpie the sound will be warmer.
I can't seem to find it now, but I remember seeing a product review I think on Amazon that was exclaiming how awesome some ridiculously priced optical cable was for a audio system because the fiber optic cable had a red tint to it imparting a warmer sound for the listener.
Re: (Score:2)
Usually, "epoxy" around the edges of a BGA chip is neither an anti-hacking attempt nor a light-proofing attempt. It's called underfill, and its chief purpose is to increase mechanical strength and make the bond more durable than tiny bare solder balls would be on their own.
Re:Maybe not the power supply? (Score:4, Informative)
This thread [ycombinator.com] shows a quick experiment which confirms it's directly the light which is the cause, not the EM pulse from the capacitor discharge in the flashgun. Chip U16 apparently, which is part of the power regulator.
Re: (Score:2)
Not all plastic or glass that's opaque to visible light [naturfotograf.com] is opaque to UV [medicalpho...phy.com.au].
Bring out the tinfoil (Score:5, Funny)
I am guessing that wrapping it in tinfoil would fix it? I know it works great for stopping the mind-control waves from getting into my head.
Re: (Score:1)
The trick is to filter out the government / nwo mind control waves while still allowing youself to recieve signals from the cosmos.
Re: (Score:1)
Aluminum foil is cheaper, more available, and has better conductivity fo EMI shielding. Only a complete asshat would use a foil made from Tin.
Re:Bring out the tinfoil (Score:5, Funny)
AC is a coverup minion for them. Every good consipiracy buff knows that tin indeed blocks the mind control rays, as opposed to aluminum. Which is why they did away with tin foil, and replaced it with aluminum. Go ahead, try to find some tin foil nowadays...
Re: (Score:2)
Tin foil is easier to solder onto the board.
Re: (Score:2)
You need a full head covering there, not just a hat to avoid the mind control fields..
You need to make sure to cover all the way down, well past the neck with foil, making sure it is air tight with no holes and tightly sealed at the bottom. It doesn't need to be tight fitting, just 100% sealed...
You will know you are doing it right if you start to feel light headed and out of breath. That's just the mind control waves wearing off and your brain returning to it's normal state.
When you do this, take a vi
Re: (Score:2)
"human"
Found the lizard person.
So I can't play some old Sierra adventure games. (Score:1)
That means I can't play Space Quest and Leisure Suit Larry at the same time.
Claim to fame is important no matter how trivial.. (Score:5, Insightful)
He only mentions that it crashes, everybody else answers the question yet he now goes by "Discoverer of the PI2 XENON DEATH FLASH !"
Re: (Score:2)
So put it in a case (Score:5, Informative)
TFA found out precisely which chip it is (U16), covering it solves the problem.
Similarly .... (Score:5, Interesting)
"Nothing like this will be built again"
I've just had a really amazing experience: a guided tour of the nuclear reactor complex at Torness on the Scottish coast. ... Cameras were verboten -- not because of security, but as an operational precaution. For starters, some embedded controllers in racks in the auxilliary deisel generator control rooms have EPROMs which have been known to be erased by camera flashes in the past, triggering a generator trip ...."
http://www.antipope.org/charli... [antipope.org]
Re:Similarly .... (Score:4, Informative)
For starters, some embedded controllers in racks in the auxilliary deisel generator control rooms have EPROMs which have been known to be erased by camera flashes in the past
That's why people have always put metal foil stickers on the EPROM window to protect them. Even exposure to sunlight can mess up uncovered EPROMs. And a little sticker seems easier and more reliable than making sure not a single camera makes it through security.
Re: (Score:3)
Yup.
One would hope they've got a belt-and-suspenders attitude there, as stickers sometimes do dry out and fall off; people sometimes put stickers on wrong; and having one's auxiliary diesel generators fail can be embarassing.
http://sfcitizen.com/blog/wp-c... [sfcitizen.com]
Re: (Score:3)
That's why people have always put metal foil stickers on the EPROM window to protect them. Even exposure to sunlight can mess up uncovered EPROMs. And a little sticker seems easier and more reliable than making sure not a single camera makes it through security.
Another problem in power generation is that arc fault detectors also has a tendency to be tripped by camera flashes. So keeping cameras out of power plants/transformer sites/etc, is standard operating procedure since time immemorial.
And these EPROMS probably had that black gunk on them that were popular when EPROMS were used in production. Problem is that the gunk had a tendency to dry up and fall off after a couple of decades.
U16 WLCSP package inherently photosensitive (Score:2, Insightful)
The device at U16 on Raspberry Pi 2 v1.1 appears to be an ON Semiconductor NCP6343 [onsemi.com] DC converter provided in a WLCSP-15 package.
Like all CSP packages, the bare die is photosensitive and needs to be protected from incident light if fault-free operation is expected. Usually such devices are embedded in closed cases like cellphones which prevent light ingress.
However, if the normal operating environment includes uncased bare boards or transparent cases (which are both common and normal for Raspberry Pi), then
Enough! (Score:5, Funny)
Stop using Flash, it's a persistent vulnerability, and Youtube has an HTML5 video player now.
Re: (Score:2)
When did Adobe sell Flash and who is this new Xenon company? Never heard of them.
Flash Player for Xbox 360 (Score:2)
Xenon was the development codename for the Xbox 360 video game console. I was more like: Since when did Adobe port Flash Player to the version of Internet Explorer on Xbox 360? Was it Wii-nus envy to catch up with Nintendo's "Internet Channel powered by Opera"?
Another good reason for OCD unboxing... (Score:2)
Re: (Score:2)
Joke aside, with some many people getting one any problems are bound to pop up quite quickly (like this one).
Re: (Score:2)
I've had trouble before with EMP - one of the few here who can honestly say it was a true EMP, not just RFI. The device producing the EMP was a four-kilovolt capacitor bank used for creatively imploding aluminium cans and launching metal objects at quite dangerous velocities. It can put out enough of a pulse to crash a camera - all the control circuits are especially hardened against it.
Nothing unusual really (Score:5, Informative)
There's plenty of cases of electronics misbehaving due to exposure to strong light. Glass enveloped diodes (such as signal diodes) can be notorious for it, as can the black plastic encased units if the light is strong enough.
Small bare CoG (Chip on Glass) LCD panels will crash / hang when you use the flash on the camera taking photos of them in operation ( same reason, the controller die is exposed ).
It's not EM-pulse or xrays causing the problem, just good ole silicon junctions being exposed to intense light :)
Re: (Score:2)
Re: (Score:2)
It's not EM-pulse or xrays causing the problem, just good ole silicon junctions being exposed to intense light :)
Yupp. Einstein [wikipedia.org] for the win! :-)
Military & Industrial Test Suites (Score:1)
Military hardware for specific applications is tested for light sensitivity, but it is not a common test.
I have never seen any industrial control equipment subjected to such a test.
Betcha there's gpnna be a whole lot of "unofficial" testing done starting Monday morning, I'll betcha.
LED acting as a light sensor (Score:2)
I've seen it before back in the day. They are not very efficient but they could cause a critical spike if they are not isolated from a bus.
I am filing this with ... (Score:2)
I am blushing and filing this with "gullible is not in the dictionary".
.....
razn1 ~ $ w
23:14:15 up 11 min, 1 user, load average: 0.00, 0.04, 0.05
USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT
bob pts/0 whitechrome.wr.t 23:14 7.00s 1.29s 0.04s w
razn1 ~ $
razn1 ~ $
razn1 ~ $ # flash flash flash... must be one honking flash but mine is big.
razn1 ~ $
@razn1 ~ $ w
23:16:14 up 13 min, 1 user, load average: 0.02, 0.04, 0.05
Had a similar problem many years ago (Score:2)
Re: (Score:2)
Poor Design! (Score:2)
This is a simply a case of poor hardware design. The design engineer should have known that exposed silicon is sensitive to light (Remember the glass window EPROMS) and used a packaged version of the regulator.
EMI and computers.... (Score:2)
Wondering how your neighbor got your banking info? Well, you transmitted it to him...
Re: (Score:2)
I've actually seen data from my laptop showing up on a TV. It was out of sync, so it was only part of the screen, repeated multiple times and scrolling slowly. This was back in the 90s, probably a 640x480 screen, and probably the radiation was mostly emitted from the external-video-monitor port (VGA or whatever we used back then.)
Re: (Score:2, Funny)
Re: (Score:2)
ymmd
Re:Not Photosensitive (Score:5, Informative)
A 100mW red laser pointer aimed at U16 also triggers it.
Unless you want to claim diode lasers now emit x-rays and low rise time EM pulses... it's light sensitive.
And inspecting U16 closely, it's no surprise. You're not looking at a plastic package but the laser marked underside of a bare die.
Re:Not Photosensitive (Score:4, Informative)
Bingo, that's about what I was going to say. U16 is flip-chip bonded to the circuit board, meaning the naked die is exposed on the bottom. Even if it had a plastic or ceramic cover, it might still be photo sensitive to light getting underneath it. If the underside of the die (flipped, so topside) is really exposed, it basically becomes a silicium solar cell.
Re: (Score:2)
"A 100mW red laser pointer aimed"
You play around with a HAND-HELD class 3b laser? Still have eyesight in both your eyes? (And your pets/household members)
Legally, those should be operated only by trained staff and I believe they must be equipped with a key lock and interlock connection. That stuff from WickedLasers should really be banned. The laser inside a dvd writer is of similar power, but interlocked and much less dangerous due to the focusing lens.
Re: (Score:1)
Define "better". So very few devices work well for things in the $35 category. You typically have to spend double that for similar gear- and IT isn't any better- they're all bare boards and each have gotchas gallore for their use.
Most people aren't going to shell out $500 or more for the board that accounts for all the possible concerns- which is what you get to pay for someone to have done most of the gotcha removals on the design. Well, unless they're building a system to commercially control an indust
Re: (Score:2, Redundant)
He is probably one of those idiots that will tout the Beaglebone Black as superior....
A Raspberry Pi 1 will blow away a Beaglebone black in video playback or GUI rendering.
Re:Crap hardware, not surprising (Score:4, Insightful)
I use both BeathBone Black's and Raspberry Pi's each has their tradeoffs. The BeathBone is better suited complex embedded applications. It has more GPIOs, two built in 200Mhz in-order microcontrollers for real time tasks, it is faster (than the pre Pi 2's), etc. Not every application needs to play video. In fact almost every project I have done didn't need video. Most didn't need a UI.
Each has their strengths and their weaknesses. Each has its niche. There is no such thing as better for all uses.
Re: (Score:3)
The Pi was never intended for video playback or a desktop replacement. The main goal was a low cost learning platform.
Re: (Score:2, Insightful)
My impression is that most people are just getting them for cheap ethernet-enabled controllers, not as learning platforms. It doesn't make that great of a learning platform, anyway. Better to get a cheap PC or laptop for that.
Re: (Score:2)
How many PCs and laptops have GPIO pins and will run from a microSD card?
Re: (Score:2)
You are missing the point. This is a platform for hardware and software hackers. Somewhere between small microcontrollers (PIC/Atmel) and a PC.
Re: (Score:2)
In fact, the Pi2 is better at web browsing than the original.
Re: (Score:2)
And sometimes learning includes watching instructional videos.
Re:Crap hardware, not surprising (Score:5, Insightful)
Really not surprising. The pi2 is pretty crappy hardware. So many better micro computers for projects, not sure why 'geeks' obsess over it.
Oh wow. Random uneducated Pi bashing. Especially considering the device causing the problem is the latest and greatest in small SMPS chip regulators and nothing at all to do with any of the parts that are typically quoted in specs and bitched about by ACs on slashdot.
Re: (Score:2)
Maybe he meant a lower model or RPi is more than sufficient. Some people are still using ATmega328P-powered Arduinos to control their 3D printers and those move a lot of a hell faster than most CNCs.
Re: EMI Noisy environments (Score:5, Insightful)
Inefficient hardware is sometimes justified by development time.
You can spend many days hand-coding an ideal program for a PIC in ASM. Or you can use an arduino, which takes more power, more space and more money, but can be programmed in a tenth the time by anyone who knows C without needing any esoteric knowledge of harvard architecture and tables of port numbers. If you're doing things a bit more complicated like image processing or networking, the same applies to arduino vs pi: The arduino may be able to do your task if you'll put in the days of programming, but with the pi you're dealing with a familiar linux environment and all the classic libraries are there.
Re: (Score:2, Informative)
It doesn't take much to actually run most CNC setups. That hasn't stopped professional equipment, costing more than thousands of dollars, from having their share of problems and crashing on us or crapping out. More often than not it comes down to user interface stuff or fancy features not used often (I'll do geometry calculations at my desk, not standing in front of the machine) that crap out, but sometimes brings the whole system down. I've seen it even happen when a company rep was demonstrating their