Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Technology

The Computer as Microwave? 32

Posted by Cliff from the the-consequences-of-the-Ghz-race dept.
Clan Hanna asks: "With the newest processors that AMD and Intel have released, running at 1.0 gigahertz, chip designers may soon have a new problem on their hands more than can simply be solved by placing a bigger heatsync on the motherboard. Microwave frequencies run from 1 gigahertz to 1 terahertz. Currently processors heat up because they emit infrared radiation, but what is going to happen when they start to emit -microwave- radiation? I'm sure chip designers think about this in the back of their minds, but I'm just wondering if they have considered any real world solutions. If so, I'd love to find out about them."
This discussion has been archived. No new comments can be posted.

The Computer as Microwave? More

The Computer as Microwave?

Comments Filter:
  • Consider the processor die as a microstrip that keeps the microwaves where they belong in the circuit flipping the flops and toggling the gates. If a significant portion of its valuable switching electrons were radiated as RF, not to be eventually used up as heat, the circuitry would be very ineffecient. That's why there is a ground plane under the chip and sandwiched inside all motherboards. We wouldn't want all those overclocked electrons to be radiated from the bus before they reached its I/O port. The resulting reflections caused by leaky transmission lines on the bus would cause errors and crash the computer.

    If there just a few milliwatts of radiated power, it should not be enough to burn tissue. Compare to the kilowatt of a microwave oven antenna, which was designed to radiate its power.

    Anyhow, the microwaves or xrays from a computer and its monitor will not kill a person. Sitting in the chair across from the monitor for extended periods of time may cause fatigue, depression, alchoholism, and madness, which is another story.
  • Actually, Slashdot is a great place to get smart answers to "stupid" questions.

    Unfortunately, the quality of the answers varies over a wide range and the person asking the question (who wouldn't be asking if they were already experts in the area) can have difficulty separating the wheat from the chaff, especially since so many of the answers fall somewhere in between.

    Here's one helpful hint: the more a particular post flames previous posters as idiots, et cetera, the more likley it is that *their* answer is based on an incomplete understanding of the issue at hand.

  • Flatscreens (Score:3)

    by jerrol ( 7184 ) on Sunday April 23, 2000 @05:46PM (#1114878)
    They emit very little radiation...

    There is basically a VERY bright lamp that is shown through a number of colored LCD screens. The Plasma displays work similarly. Electroluminescent displays emit even less...

    Any of these is much better than the electron gun in the back of CRTs. (basically a small particle accelerator..)

    Of course, if you are truly worried about radiation poisoning: never fly (cosmic rays), don't get X-rays (X-rays), stay away from cinderblocks (alpha-emitter), ditch that glow-in-the-dark watch (radioactive), discard you're smoke detector (radioactive), stay away from the Northeast (radon, radioactive)...oh hell, just leave the planet beacause a couple of years ago, a satellite with a nuclear reactor onboard burnt up on re-entry and spread plutonium over the entire planet, of course, interplanetary travel opens up another can of worms...

    =)

  • While the wattage is important, you got one fact wrong: A gigahertz Athlon produces much more than a few milliwatts, they are actually in the range of 50-100 watts IIRC. Sure, it's a far cry from my 800 watt microwave oven, but still significant.

    -Smitty

  • Microwaves aren't that bad (Score:3)

    by mind21_98 ( 18647 ) on Sunday April 23, 2000 @12:30PM (#1114880) Homepage Journal
    I'm an amateur radio operator, and the thing we have to watch out for is RF radiation. At the microwave bands, the body starts absorbing more radiation and the internal tissue starts heating up *shudder*. If you're exposed for long periods of time to this, eventually you get stuff like cataracts, cancer and so forth.

    However, to my understanding all modern computers are supposed to comply with FCC Part B regulations. This means that they don't emit anything that could cause interference or other harm. In other words, you have nothing to worry about unless you run the system with the case open.
  • Duh! It's Ask Slashdot. That makes it a question - not a story. It sounds as though you have a good answer as well - perhaps you should just have given the answer instead of whining - would have contributed more to the conversation dont'cha think?

    "Give the anarchist a cigarette"
  • The heat from that laptop is stressing your skin more than any RF...
  • Hold on here. If there was a microwave RF problem, we'd know it.

    We've all made coasters in the microwave oven out of AOL CDs, right? If a computer was creating significant amounts of RF, we'd be seeing the same sparks crawling around various metal components.

    Obviously, computer designers would have to block any such leakage or all the electronics would be destroyed.

  • I really doubt there is. Why would there be? If systems are designed to be Class B compliant with the lid on, that's good enough. People shouldn't expect a device to function as specified if they don't use it as intended.

    BTW, how does running systems without a lid make magazine reviewers better? Or power users? A properly designed (take IBM's PS/2 line) PC case must have the lid in place in order for air to flow correctly through the system to keep everything cool. Without a lid, the flow is changed an the intake/exhaust fans can't operate properly.

  • A LOT of people do run their systems with the case open, especially (good) magazine reviewers, etc. Power users do as well. There must be other shielding other than just the case, right?

    Chris Hagar
  • First of all, I'd like to say that I'm an idiot. What I meant by the case open with magazine reviewers was that if someone is reviewing a system, they should look around inside. Some good magazines have the amount of expandability (PCI, ISA slots open) and how well it's all tied up inside. Of course, this wouldn't be done while the system is running...

    I was thinking that a lot of cases/entire system/fanning is/are not designed properly and so in many systems keeping the case off might keep it much cooler. Also, if someone is doing any sort of upgrading, I wouldn't want to pop the case back on before I boot the computer up to make sure it worked correctly.

    Chris Hagar

  • It's not the voltage that kills you, its the wattage!! A gig Athlon isn't going to put out more than a couple of milliwatts. The magnetron in your microwave is rated for 1500-3000 watts. The microwave is almost a million times more powerful.

  • It's even more mundane than that. (Score:3)

    by Tau Zero ( 75868 ) on Monday April 24, 2000 @02:05PM (#1114888) Journal
    ...call me an EE major (I am one :-) but I belive infrared is ABOVE microwave in frequency.
    Call me a double-E (I have a degree that says that I am). Infrared is indeed "above" (greater energy, shorter wavelength) microwave radiation. The clock signal does affect the radiation from a device; the faster the clock and the sharper the edges, the more RF radiation you get from the leads and such. This isn't going to cook you, but RFI and Tempest emissions ought to be kept in mind.

    Cliff wrote:

    Currently processors heat up because they emit infrared radiation, but what is going to happen when they start to emit -microwave- radiation?
    Well, Cliff ol' boy, I've got news for you: processors don't heat up because they emit infrared radiation, they emit infrared radiation period. So do you. A square meter of blackbody radiator at skin temperature emits about 450 watts[1], mostly in the infrared. However, some of it is actually at longer wavelengths, down into the microwaves. (YOU are a MICROWAVE EMITTER! And if you don't pay me $10,000, I'll TELL YOUR FRIENDS AND FAMILY!)

    This property is used all the time. One of the methods for sensing temperature remotely is with microwave radiometry; it's how the surface temperature of Venus was determined before probes were ever sent there, and the bulk temperature of Earth's atmosphere is measured by satellite sensors using that same technique to this very day.

    You don't have to worry about microwave emissions from a CPU chip, even if it's running at a couple of GHz. The chip carrier itself usually has a metal plate for heat dissipation, forming a ground plane; ground planes provide "equal and opposite" mirror currents and mostly cancel the emissions. The metal lid is another ground plane. Even at 3 GHz, a wavelength is 10 centimeters; most traces on the chip are but a fraction of a millimeter long, and don't have the physical size to radiate well. What radiation they do emit will be cancelled by image currents in the ground planes and confined inside the chip carrier until it is absorbed by resistance in the silicon itself. The biggest issue is emissions from leads which go out of the chip itself, and those are usually running at a much lower frequency than the CPU core. What gets out is microwatts at most; this can be troublesome, but danger to people is way down on the list of concerns.

    I think the summary is, no need to worry yet but you should probably do some reading so that you will know if and when you should worry in the future.

    [1] Blackbody radiation flux is determined by the absolute temperature and Boltzmann's constant; the formula for the radiation per unit area is flux = 5.67*10^(-8) W/m^2/K^4 * temperature^4, so the heat flux increases as the fourth power of temperature.
    --

  • actually, it's the amperage that kills you, but you are definately on the right track

  • apples and oranges (Score:3)

    by Bourbon Man ( 76846 ) on Sunday April 23, 2000 @06:19PM (#1114890) Homepage
    Here is why you need not worry. We are talking about two completely different types of equipment. A microwave oven has a special device in it called a magnetron that produces microwave radiation. It is usually powered by a step-up transformer that puts out around 3000 volts. It was expressly designed to transmit radiation in the form of microwaves (a radio waveform). A CPU has no emitter, i.e it has no parts in it designed to emit radiation. It was designed to push electrons through predetermined paths. There is a big diference between something operating at 1Ghz, and something emitting a waveform at 1Ghz. The infrared (also known as heat) produced by a CPU is not directly a product of it's speed, it is a byproduct of resistance (basically the friction of electrons moving through an imperfect conductor). Chip manufacturers are reducing heat and increasing speed *at the same time* by switching to copper-based chips because copper is a superior conductor. They are also lowering the voltage needed by the chips to the sub-2 volt range, a long ways from the 3000 volts a microwave uses.

  • Wait a minute... (Score:5)

    by dufke ( 82386 ) on Sunday April 23, 2000 @01:13PM (#1114891)
    ...call me an EE major (I am one :-) but I belive infrared is ABOVE microwave in frequency. Anyway, most of the heat produced by a CPU is transfered through heat conduction, not radiation. I don't think the actual clock signal would cause much radiation. Clocking a CPU higher may increase heat production, but the frequency of the radiation (if any) is not directly related to the clock frequency.

    Anyway, even if it did, the high-amplitude microwaves in an oven are stopped by rather thin sheets of steel. And the amplitude from the CPU must be lower. Remember, your oven draws multiple kW, while I damn hope your CPU keeps below 100W. (Otherwise, you have an Alpha, and I WANT IT! :-)
    -
  • You forget mobile telephones... at least I get most of my microwaves from a crappy Ericsson, I think.

    At least in Finland mob. phones meet the radiation regulations only because of the fact that GSM sends data in bursts - half the time there's no radiation, half the time there's more radiation than what would be allowed continuously.

    Luckily (at least for the phone industry) the technology develops in such leaps, that the long-term health effects haven't got the time to develop... yesterday NMT's, today GSM/PCS, tomorrow GPRS and stuff. And as development continues and the frequencies go up, the transmitting power goes down.

    I think I should be more worried about using laptop, anyway. Not for the TFT, but for the PIII 450 MHz ticking at my groin...

  • heh....you are assuming a properly designed case...
    my case == immediate 20+ degree(F) increase in cpu temp...
  • Even if they are a square wave, they still are emitting energy at high frequencies. Instead of just a sign wave at the clock frequency, it is that plus higher frequencies at smaller amplitudes.

    A lot of people in the posts above are suggesting that the low power means that this is nothing to worry about, just more un-informed hyteria. While I totally admitt to the un-informed part, you have to realize that the "always-on" nature of the computer means that damage can accumulate over a long period of time. I spend more than half my waking hours (at least during the week) in a room with more than 3 computers within a few feet. Many people (especially college students) sleep in a room with computers close to the bed.

    I don't understand the very much about the physics of how RF does biological damage; I've read about how "harder" radiation, such as gamma and particle radiation, works. Basically, you want to assure yourself that the radiation is not enough to produce a free radical next to a DNA strand, so that the radical can combine with the DNA and break it. Now, because of the way that your cells repair DNA, you can tolerate a certain amount of damage; kind of like an error-correcting code will keep all the information as long as only a certain portion of the bits are lost or flipped.

    Particle radiation can make chemical changes just by hitting the molecules, but RF level radiation does damage (or can do damage) by simply heating an area, which can damge DNA or produce free radicals or oxygen which will damage it.

    So while I think the power observations are very pertinent, I don't think they can be the basis of dismisal of the danger. I think you have to be able to say that the amount of heating your body might receive from several computers 24/7 is on the order of what you get anyway from the electrical system, or something like that. There must be some effect, but as long as it is below what you can measure it is probably below the level at which DNA repair will protect you from cancer.

    So one final question -- someone (the first post) mentions that ham radio hobbists get cataracts more often. What is the physical mechanism that causes that ? I know that exposure to too much UV light is supposed to increase the incidence of cataracts. I have glasses with an anti-UV coating that the eyeglass specialist recommended on those grounds.
  • Yeah, that makes total sense. It can't be just heat. You wouldn't expect that there to be a positive correlation between climate or exercising in sweats with cancer, would you ?

    So what is it about the EM radiation that causes the damage ? Why is the EM from a leaking microwave oven bad for you ?

    This reminds me of that whole debate about whether living in the proximity of powerlines and transformers increased the incidence of cancer. No one ever made a convencing study as I remember, and they never did offer a good physical description of how that energy could harm you.

    I know one professor who teaches a bio-electronics class (6.121J) and I can ask him.
  • Yeah, but a lot of people run their computers with the case open, or without a case at all (you can just screw everthing into a plastic milk crate, or worse; just look at some of the overclocked systems -- a pile of wires and boards on the kitchen table or in the freezer or in cooled mineral oil).

    Before I buy a chip running at that frequency, I'd like to see some specs on exactly what I can and cannot do with it. If I need to keep it in a case that meets certain specifications, I should know that; is it safe to just not screw in the panels on the case, or should I always have everything tightened up before powering on ? I would prefer to have the processor package do all the necessary shielding for me.

    What about those guys doing wearable computing ? I know Steve Mann and that Media Lab freak club were walking around MIT with computers butt packs for a while. For what processors they had there was probably no danger, but with a gigahertz plus processor hanging over the genital area or close to your body at all -- cancer at least only affects the person who decided ot wear the thing, a birth defect punishes a child for the parent's negligence.

    I would prefer to get the information from an independent source; if AMD or Intel said that it was dangerous if not run as shipped, the information would be suspect because they might just be trying to scare off the overclockers, and if they said it would always be safe, they might just be hiding the truth for marketing reasons.

    So I would like to see some overclockers and someone with access to the right kind of equipment test the newest processors at some really high clock rates (even if the processor was unstable at that rate) and publish the results. The "right kind of equipment" is the kicker here -- I've heard those electrical isolation chambers they use to test the meeting of FCC specs are very expensive.

    I am planning to buy a really fast alpha in a few months after I move to my new place. I've heard rumours of 1.4 and 1.6 GH by then. I'll just get some really long keyboard and monitor cables until I see some trustable information on this issue. Whatever the manufacturer says, I'll still trust 1/R^2 -- you can't beat basic physics.

    Incidentally -- does anyone know if these expensive flat screen monitors produce signifcantly less EM radiation than the good old cathode ray tube ? I think the fact that they use less power is a good sign.
  • Thanks for the advice bud. Logitech was much less helpful after many many emails. =)

    -rt-
  • ... when I can hear the humming of my computer over the AM radio frequencies. And the movements of my RF mouse over my speakers. Anyone have similar experiences or a reason why this might be??

    -rt-

  • I get a little upset... (Score:4)

    by jdwilso2 ( 90224 ) on Sunday April 23, 2000 @05:53PM (#1114899)
    When /. posts a bogus story like this one... uPs aren't gonna produce any harmful "radiation" such as this story suggests. Sure, every electrical device emmits a multitude of interferance, but to claim that when processors start getting above 1GHz I can cook my TV dinner in my case is rediculous. Power has a great deal to do with such things, and so does the appication. I mean, come on... Let's be serious, a 2.4GHz coordless phone meets FCC specs, and it's *transmitting* it's 2.4GHz signal... RF stuff can't hurt you, and it's moving through all the space around you. Hell, the military use spread spectrum encoded wireless transmissions whose carrier frequencies are 10GHz and UP! And if RF doesn't hurt you, ain't no clocked digital logic circuit gonna byte you. Especially at those exceedingly low frequencies of under 100GHz. It's all about power guys... Do some reaserch before you start spewing your crap all over me. It's all about the power...
  • Yeah, I don't think that we're actually broadcasting radio waves at 1000 Mhz, are we? It's deep in a chip beside a heatsink, inside a case. It there really that much to worry about? -Pete
  • My guess is that the worlds of junk food and computing will finally completely merge. Then, as soon as I build that 30ft catheter system, I will have no reason to leave my room.

  • So while I think the power observations are very pertinent, I don't think they can be the basis of dismisal of the danger. I think you have to be able to say that the amount of heating your body might receive from several computers 24/7 is on the order of what you get anyway from the electrical system, or something like that. There must be some effect, but as long as it is below what you can measure it is probably below the level at which DNA repair will protect you from cancer.

    Heating? The amount of heating you get by turning your thermostat up a few degrees has to be orders of magnitude greater than any heating caused by RF radiation.

  • As you wil recall from scopin' around your computer, the clock pulse is a square wave, (computers are digital, remember?) However, even if they are oscilating at speeds that would put them in the microwave spectrum, cmos chips don't output or draw enough current to cause anyone any harm..

    --Alex,
  • Most of the Radio Interference from your PC comes from your fans. As all the Fans in your computer (Power Supply Fan, CPU Fan, Case Fan) are all DC motors they are extremly noisy (in Radio Frequencies as well as sound frequences). You should move your case away from your RF mouse. (:
  • "Honey! Is dinner ready?"

    "After two more minutes of cracking RC5 keys, it should be ready!"

    I can wish, can't I?
  • Go look at your microwave. It has lots more radiation in there then any processor has ever produced. Nobody worries about standing infront of a microvave. You simply need to make the case for the processor out of the same stuff they use to block the microwaves from leaving your microwave.
  • Doubtful. Look at your processor. Take the heatsink off. There are just a few mm of ceramic between you and the actual silicon wafer. Its the same way with all the other chips. The only shielding would be the heatsink, and that doesn't cover the entire processor, unless you encase your processors to ward off condensation.......

Slashdot Top Deals

Beware of Programmers who carry screwdrivers. -- Leonard Brandwein

Close