Cosmic Rays
 

In an attempt to cope with insomnia, I'm reading the new PMW-F3 manual. On page 134, I found this:

"Although the CMOS image sensors are produced with high-precision technologies, fine white flecks may be generated on the screen in rare cases, caused by cosmic rays, etc."

COSMIC RAYS!!!

I already wear my tinfoil hat whenever I leave the house, but now I'm going to have to make one for the camera, too. Life is getting very complicated.

I remember an old myth that being in a airliner at altitude reduced the amount of protective atmosphere and the sensors could be damaged, at least I think it was a myth.

When I get a good take, I'm going to be sure to ask my AC to check the gate for Cosmic Rays.

It's no myth that you are exposed to higher doses of radiation than you are on the ground. About equivalent to an x-ray on a 5 hour flight.

Yes and bananas are fairly radioactive as well. Enough so to set off detectors set up for people trying to smuggle uranium. If you eat one a day for a year you have definitely increased your chances of getting cancer. You know what that means right? Keep them away from my camera! :)

It's no myth, it's very true and well known that cosmic rays are more damaging at high altitudes. The atmosphere acts as an absorber and reduces the amount of the damaging particles from reaching the ground. Look at any video camera on the ISS or space shuttle and it will be plastered with white flecks. If you fly long haul with CCD cameras a lot the chances are that you will encounter this problem from time to time. The same cosmic rays can also corrupt memory cells in flash media and computer RAM. It's not normally an issue with a computer as it will normally have some form of error checking, but potentially it could corrupt footage on a flash memory card such as an SxS card. I have not seen any reports of this, but it should be considered, just in case. It's even been speculated that Toyota's problems with cars randomly accelerating has been caused by cosmic rays flipping data bits in the controllers for the engine.

Hopefully, someone at Zacuto is reading this thread and working on some sort of lead shield that will mitigate this problem. Bright yellow. Banana yellow.

Lead does not stop cosmic rays. Hydrogen absorbs them (or something like that) which is why the atmosphere reduces thier effect. There's an extended thread on this here: http://www.dvinfo.net/forum/home-awa...revention.html

i had a replacement harddrive reading out jumbled data once. phoned tech-support and after ruling out a number of causes they suggested that relatively high sunspot activity could be the cause.
...turned out to be a bad cable.

Alister, I just read that this hypothesis was disconfirmed. A recent report by NASA and the National Highway Traffic Safety Administration claimed that the unintended acceleration in Toyota vehicles was rooted in mechanical flaws rather than electronic defects. The report found "no causes for the unintended acceleration incidents other than sticking accelerator pedals and floor mats that jammed the pedals down." Those were the causes Toyota had identified. NASA studied whether electromagnetic interference may have caused unintended acceleration, which may be linked to 89 deaths in 71 crashes since 2000, according to the safety agency. NASA investigators used Chrysler Group LLC's test facility in Michigan for its vehicle testing work, and bombarded vehicles with electromagnetic radiation to test the potential effects on the car's circuitry.

Toyota problems were mechanical, study says

Kind of a fun topic.

Primary cosmic rays -- the type we most often hear about are actually not "rays" they originate as particles: mostly protons (nearly 90%) helium nuclei (9%) and electrons (1%) and some other smaller parts. Because the particles are magnetically charged, much of the cosmic "rays" are deflected before they get to the atmosphere. They get concentrated at the poles because they follow the magnetic fields and we sometimes see the effect as aurora borealis or australis.

When cosmic ray particles do hit the atmosphere they collide mostly with nitrogen and oxygen and break up into lots of smaller unstable particles that quickly decay into "muons"; very tiny particles that don't interact with the atmosphere very much so can often reach the earth and because they are small and moving fast, they can penetrate a little way into the earth. Because muons are ionizing radiation they can hit and affect a computer chip (in a camera or in a computer). A computer might get hit once every several years and the induced error in the past was minor enough it was not really noticeable. However as the chips get smaller and more tightly packed with microprocessors and supercomputers are built to have many processors, the frequency of hits on the supercomputer could be 10-20 cosmic ray hits per week -- enough to be a problem. It is real enough that Intel is working on a cosmic ray detector (basically searcing for electrical spikes and eliminating the calculations since the cosmic strike and redoing the calculation).

I don't know if the chips in cameras or our typical computers are really very vulnerable (unless you happen to own a Cray), but it is within the realm of possible. Cosmic ray particles are detected in experiments (normally undertaken in space or high altitude from ballons so there is no interference from the atmosphere) by exposing plastic sheets, then bringing them back and "etching" them using sodium hydroxide which dissolves along the path of damage faster than undamaged plastic. So by extension it is also possible to damage plastic parts of a camera including sensors. The chances of significant visible damage are really small -- but theoretically present.

As I say, it is a fun topic.

Alan

Tossed and turned all night. Wife said I was mumbling in my sleep. Something about "rays". Ate a banana this morning and felt better.

...the CCDs on NASA's Solar Dynamics Observatory seem to be doing just fine...

Thanks for the reminder. Bought some bananas today and took a reading. Nothing above background. Didn't peel it so could be some alphas hiding in there, but I kind of doubt it.

K40 gives off beta, not alpha. The amount of radiation is only "high" in that it is the largest single source of naturally occurring radiation exposure for most people.

Jim, even though the tin foil hat is a definite fashion statement, you might want to hand it down to your least favorite sibling or "friend" because as Alan pointed out "cosmic rays" -- or more correctly galactic cosmic radiation -- are high Z (very energetic) particles and when they strike metals produce a lot of secondary radiation that can actually be more damaging to pink flesh than the original particle. GCR is very difficult to shield, but fortunately the earth's magnetic field and atmosphere together do a great job of it so very, very few (still not "none") make it to our filming locations.

My knowledge of 'alternative science', garnered during years of searching for cryptids in the forests and swamps of North America, gives me what some may call "a unique perspective". I'll keep the tinfoil hat on, thank you very much.

You guys are the reason why marijuanna should not be legalized :)

It is not just theory. It happens all the time. Do your research and you'll see that an SD CCD sensor can expect to have a pixel damaged every year or so and an HD sensor around 4 a year at ground level. That's why masking circuits are built into most cameras. When you fly the risk is greatly increased especially on Polar routes, which is why Sony won't air freight cameras on polar routes. CMOS sensors are affected as well although the masking circuits tend to hide them so you rarely see them.

Hi Allister,

You are the expert in this field for sure and I don't question that. My intention was to clarify what cosmic rays are and how likely it is that visibly significant damage can be done by them on normal computers or normal (SD or HD) camera sensors. The most likely to be affected are CCD sensors.

What you describe as observed and in-practice policies perfectly fits the theoretical models for the way cosmic ray particles behave in our situation here on earth. I would agree further that a camera above most of the atmosphere and at the poles is more likely to encounter a full cosmic ray particle with its complete complement of photons, helium nuclei, and electrons. These can cause damage to a sensor for sure.

Again just for clarification, we can detect cosmic ray particle strikes on the earth (which are primarily muons, not the full primary cosmic ray particles) on average about 2 times per minute per square centimeter. The particles are so tiny that the number of times they would actually interact with some physical part of a sensor is quite small -- I accept your estimate of 4 hits per year on a pixel-crowded HD sensor given your expertise.

I may be quite wrong here, and I acknowledge that, but I understood that a cosmic ray particle (muon type) hitting a computer chip or a camera sensor causes a discharge of electrons (because it is ionizing radiation). In a computer, the discharge is equivalent to a signal asking for a computation. On a sensor, I understood a cosmic ray particle strike normally causes a discharge of electrons that would be indistinguishable from a photon particle strike and thus give a false reading for the pixel that was affected. Thus the picture quality (or accuracy) would be reduced in that instance.

If I understand you correctly, your experience suggests that the damage from a muon can be more than just the discharge of electrons messing up that pixel for that picture, but rather that the damage can be a permanent loss of a pixel. I assume that damage would in future pictures render a black pixel on the sensor and in the final image. Again taking your numbers, that would suggest that a four-year old HD sensor could have about 16 dead pixels scattered randomly on the sensor as a result of cosmic radiation from muons. Here again, I bow to your knowledge and experience in making spectacular video images. Can the human eye detect a single dark pixel in an image of about 2 million pixels (even in a pale background) even in unusual viewing situations such as a large projection screen in a theater?

Finally, and again, I am a novice in making videos, so as a result of your comments, I am now made unsure of my knowledge, but I understood the dark mask process for CCD sensors was to correct a wide array of sensitivity errors caused in the manufacture of the sensor (read noise, thermal noise, electronic noise from the camera, uneven silicon distribution resulting in different pixel sensitivities). In this dark masking process a series of comparative images are compared to a flat master frame and the differences from the flat (or evenly dark) image are then subtracted digitally when the camera is taking pictures. But perhaps you are referring to a different masking circuit? I made the assumption that this was the only correction that the camera makes while it is taking pictures. I was not aware that a normal camera such as we use could correct for random cosmic ray particle strikes.

GIven the difficulty (essentially impossible) of shielding against cosmic rays particles that actually make it to earth (muons), and given that electronic discharges resulting from those cosmic ray particles are therefore essentially unavoidable, but rare ( a few times a year), I speculated that the likelihood of visibly significant effects from cosmic rays would be really small -- but theoretically possible.

However, if your experience and observation suggests that cosmic ray particle strikes at earth level result in permanent damage at a rate of 4 dead pixels per year in an HD sensor and that a few (4 to 20) randomly distributed dark pixels in an image of over 2 million pixels of varying intensities are visible then I stand corrected, and am happy to learn. In addition if the current cameras have circuitry that can correct for the cosmic ray strikes then I am also happy to stand corrected and to be impressed by modern technology.

Alan

I'm happy, too! ; )

What happens to both CMOS and CCD sensors as well as flash memory is that the energetic particle punches a small hole through the insulator of the pixel or memory cell. In practice what then happens is that charge can leak from the pixel to the substrate or from the substrate to the pixel. In the dark part of an image the amount of photons hitting the sensor is extremely small, each photon (in a perfect sensor) gets turn into an electron. It doesn't take much of a leak for enough additional electrons to seep through the hole in the insulation to the pixel and give a false, bright readout. With a very small leak, the pixel may still be useable simply be adding an offset to to the read out to account for the elevated black level. In a more severe cases the pixel will be flooded with leaked electrons and appear white, in this case the masking circuits should read out the adjacent pixel.

For a computer running with big voltage/charge swings between 1's and 0's this small leakage current is largely inconsequential, but it does not take much to upset the readout of a sensor when your only talking of a handful of electrons. CMOS sensors are easier to mask as each pixel is addressed individually and during the camera start up it is normal to scan the sensor looking for excessively "hot" pixels. In addition many CMOS sensors incorporate pixel level noise reduction that takes a snapshot of the pixels dark voltage and subtracts it from the exposed voltage to reduce noise. A side effect of this is it masks hot pixels quite effectively. Due to the way a CCD's output is pulled down through the entire sensor, masking is harder to do, so you often have to run a special masking routine to detect and mask hot pixels.

It may not sound much getting a single hot pixel, but if it's right in the middle of the frame, every time that part of your scene is not brightly illuminated you see it winking away at you and on dark scenes it will stick out like a sore thumb, thankfully masking circuits are very effective at either zeroing out the raised signal level or reading out an adjacent pixel.

Flash memory can also experience these same insulation holes. There are two common types of Flash Memory, SLC and MLC. Single Level Cells have two states, on or off. Any charge means on and no charge means off. A small amount of leakage, in the short term, would have minimal impact as it could take months or years for the cell to full discharge, even then there is a 50/50 chance that the empty cell will still be giving an accurate ouput as it may have been empty to start with. Even so, long term you could loose data and a big insulation leak could discharge a cell quite quickly. MLC or Multi Level Cells are much more problematic, as the name suggests these cells can have several states, each state defined by a specific charge range, so one cell can store several bits of data. A small leak in a MLC cell can quickly alter the state of the cell form one level to the next, corrupting the data by changing the voltage.

SxS Pro cards (blue ones) are SLC, SxS-1 (Orange cards) use MLC as MLC works out cheaper as fewer cells are required to store the same amount of data. Most consumer flash memory is MLC. So be warned, storing data long term on flash memory may not be as safe as you might think!

Hi Allister,

Many thanks! Much easier to understand now. I especially appreciated the explanation of the camera start-up masking system for the CMOS sensor.

Alan

Hi Alister,

As requested, I checked on the cosmic ray question some more. Here is what I found -- for purposes of clarification.

Primary galactic cosmic ray particles are composed primarly of protons, some helieum nuclei, a few electrons and some other bits. These rarely hit earth as they are deflected by earth's magnetic field or collide with the atmosphere. When they hit the atmosphere they create a shower of particles, most of which decay quickly into muons, but there are a few neutrons and other particles as well.

Muons hit earth at a frequency of about 1 muon per square cm per minute. When a muon hits a sensor it causes a discharge of electrons indistinguishable from a photon particle. A simple test with a DSLR will show this result. Turn on the camera and click the shutter in complete darkness. Then set the camera for a time exposure of about 10 minutes in complete darkness. In that time the sensor will have collected about 10 muon strikes. Some will be tiny dots others might be a streak if the muon hit obliquely. Compare the first and second exposures to be sure you are not looking at something else on the sensor. The next exposure will reset your sensor. For a video camera, the muon significance is negligble because the exposure is short (usually 1/30 sec or less).

Neutrons are a different story. They cause permanent damage as you suggested. They are much rarer, however. I found direct observation results that suggest neutron damage depends on the energy of the strike. "Minor" blemishes (not noticeable but detectable) occur about three times as frequently as "major" blemishes (which I presume would result in a winking pixel as you described). The rates are as follows: Major blemishes: 2/3 inch sensor about 1 every three years, 1/2 inch sensor about 1 in four years. 1/3 inch sensor about 1 in 7 years. Minor blemishes would be about three times as common as major blemishes. This same study showed rates in airplanes over the poles to be about 100 times as frequent -- so a very significant issue.

If your suggestion that HD sensors will receive strikes four times as frequently as SD sensors is your observation, I accept that, but I was unable to find similar observations. According to the literature I could find, neutron blemishes are dependent on area, not pixel density.

Hope that helps. Thanks again for your insight!
Alan

The more pixels there are the more chance there is of the particle hitting a critical part as the sensor is more densely packed. What ever the numbers, you can expect to see bright pixels crop up on a regular basis, especially if you fly. I'm back and forwards to the US on a regular basis and it plays havoc with my cameras. I had to get my old Digibeta re-programmed by Sony as it had so many damaged pixels that there wasn't enough ram to store the masking data. The re-programming sorted it out. I see hot pixels appear almost monthly on my PDW-700.

I had to retire my Sony HDCam due to too many bright pixels to hide. My HDX has travelled to 13 countries in the last two years and it is starting to show the same issues. I have no doubt that all that air travel has hastened their demise. Might be the cost of doing business, but still painful nonetheless.

Re: Cosmic Rays
 

I'd be inclined to dismiss this as folklore and nonsense, but I guess the real life data doesn't lie.
Someone needs to make a flight case to help this...what would it have to be made of?

Re: Cosmic Rays
 

I respectfully have to ask this, Why does it only effect cameras ? Most of the cockpits in todays aircraft are called glass cockpits. There are radar and stormscopes and displays and I imagine a few cameras in todays modern aircraft. It would seem that if it could strike and effect a sensor it would strike and effect other things as well and have a material that could shield it. Muons often penetrate mountains, is there actually any atmosphere stopping them? If the atmosphere protects us then the cameras on the space station would be unprotected ? Hubble ?

Re: Cosmic Rays
 

The lower the atomic number the better, so hydrogen is actually the best element; polyethylene is cheap, easy to work with, and contains lots of hydrogen. But from an every-day standpoint, Galactic Cosmic Radiation (GCR) can't really be shielded because it either takes impractical amounts of mass and/or significant magnetic fields to make a difference.

That's why these probes cost hundreds of millions of dollars. You can't go to your local Fry's and get the kind of mil spec or custom hardware they use to build satellites. For non-critical purposes, the ISS -- relatively protected in low earth orbit by the magnetosphere -- uses off-the-shelf cameras that gradually suffer lost photosites and laptops that occasionally suffer unexplained bit flips. Don, aircraft avionics also are built to a more robust standard than consumer goods.

Re: Cosmic Rays
 

An interesting read from NASA etc.


Muons

Re: Cosmic Rays
 

Don,
Interesting particle physics info, but for the sake of truth in adverstising it should be noted that the web site is a Principal Investigator's site, not a NASA web site.

Re: Cosmic Rays
 

You are correct, some of the references were from nasa.

Re: Cosmic Rays
 

I have been away for some time, but am now back and can add a bit more information about cosmic ray particles and camera sensors. The information I provided on muons remains valid, however the information on neutron defects can now be updated. Not many of the neutrons actually hit anything when they pass through the sensor but when they do they can cause permanent defects.

The most recent article I can find that has actual data (based on 23 DSLRs, 4 point and shoot cameras, and 11 cell phone cameras) as opposed to speculation is February 16, 2011 in volume 12 of “Sensors, Cameras, and Systems for Industrial, Scientific, and Consumer Applications”. Previous results showed that CMOS (APS) sensors were somewhat less vulnerable than CCD sensors. This study shows high ISO ranges cause the defects to be more noticeable thereby increasing the “rate” of observable defect. It also shows that for a given pixel size the number of defects scales with sensor size. The study also shows that defect rate increases rapidly (an empirical power law of the pixel size to be -2.5; a somewhat faster rate than just area as pixels sizes shrink). A large sensor can reach higher ISO numbers so is thereby more vulnerable than smaller sensors to cosmic ray neutron damage.

A different study looked at the relationship between altitude and latitude. Given Vancouver as a base index of “1”, Denver was “4”, Hong Kong about “.5”. The closer to the poles and the higher you are the more vulnerable is the sensor. Living in mid latitudes at moderate altitudes like Denver makes the camera 8 times as vulnerable as living at sea level in the tropics such as near Hong Kong. Other studies have shown that at altitudes reached by commercial aircraft the defect rate can be 100 times greater than at sea level.

As cameras evolve to higher resolution and faster speeds, cosmic ray particle defects will presumably be an increasingly significant design consideration for camera sensors, memory chips, and on-board computers (which also can be hit by muons and neutrons).

I still think this is a fun -- and serious -- problem. Here we are talking about quantum physics affecting our everyday lives. So far no one has actually tested the idea of a tin foil hat worn shiny side out.

Alan

Re: Cosmic Rays
 

As soon as I can get the HDMI output problem on my F3 fixed, I'll do those tinfoil hat tests.

Re: Cosmic Rays
 

Would man made radiation from reactors pose a problem to the cameras in Japan ?

Re: Cosmic Rays
 

Hi Don,

Japan's situation is very nasty right now, and the radioactivity scare is definitely real but with so little reliable information it is hard to know what is really happening in and around the plants. The radiation from the core will be alpha, beta, and gamma radiation. Alpha and beta won't harm cameras because the outer camera shell will stop them, but gamma radiation can if the camera is within range of the radiation (takes about 2-4 inches of lead to stop that stuff). Several explosions (probably mix of gases and superheated water) will have released particles that have ionizing radiation so if that settles on the cameras, it could do damage. Certainly people are in trouble over the long-term if the gas cloud is captured and dumped by snow or rain in any significant concentrations where people have to get food and water.

No nuclear explosion is possible with the materials in the Japanese cores -- they have not been sufficiently enriched to blow up, but they can get really hot (enough to melt steel and titanium). In Chernoble the cores were not in a heavy duty container so when the gases blew up it scattered all kinds of nasty material all over the Northern Hemisphere. A similar mess will only occur if the Japanese plants get so out of control that the melt-down destroys the containment building and then fires and gas explosions can send very bad radiation into the atmosphere. I am not too sure about the spent rods (they usually still have about 90-95% of the original energy in them) and are covered in water in what looks like swimming pools. If they are too close to the reactors, that could become a problem as well (proper storage should hold them at quite a distance, but these plants are 40 years old so who knows).

So unless the unthinkable happens, and assuming the commercial camera storage air is filtered and the warehouses have not been compromised by the quakes, warehoused cameras should be OK. Cameras in stores could be another story depending on how the store is built and if the building was damaged enough to let radioactive particles settle inside.

Alan

Re: Cosmic Rays
 

If you know which radionuclides are involved you can use the nuclear wallet cards at Nuclear Wallet Cards to look at decay products of the isotopes in question and determine which decay byproducts (alpha, beta, gamma and their percentages) are formed, which isotopes result from the decay, and their half-lives, decay products, and so on.

Re: Cosmic Rays
 

Hi Jim,

At Chernoble and in spent fuels, the most common problem radio-isotopes are Caesium 137 (gamma decay with half life of 30 years), Strontium 90 (Beta decay with half life of 28.8 years but seems to seek out bone) and Iodine 131 (gamma decay with half life of about 8 days goes to the thyroid which picks up iodine).

Alan