RF: beware of audible signals

GA from me.....

appreciate your good answer EMC C. without something to create a 'receiver' effect, even at or above 0 dBm (? threshhold of hearing) a radio wave could produce heat but not sound.

"rusty bolt".. I've heard this elaborated on as a 'resonant cavity' effect. It can happen in buildings with a particular geometry (big barns for example) under some circumstances.

re: monitoring and standards: Moose posted this link in a thread about DECT and TETRA not long ago http://www.tetrawatch.net/science/dect.php Pulsed signals with high peak amplitudes are often considered to be potentially more dangerous to health than other forms of RF.

Some scientists have also suggested that the unexplained problem in recent years of bees abandoning their hives and bee population decline generally could be related to mobile phone use.. we often don't know the effects we are having on our environment, unfortunately, until something is seriously wrong. There is a vast amount of wireless technology happening right now. Monitoring is not a bad thing IMO. If the RF we're generating is harmless, then monitoring will show that too.

The rusty bolt is different than a tuned cavity. The rusty bolt acts like a non-linear semiconductor band gap (diode) to rectify and detect rf and also can serve as a mixer: two different frequencies impinge on the rusty bolt, and a sum or difference of the two frequencies emerges. That can cause havoc on a ship if the sum or difference frequency is in band to a receiver tuned to it: the rusty bolt-generated frequency will interfere with and swamp the lower-level signal the receiver was originally trying to copy, since the rusty bolt-generated signal is at a high level because of the high power transmissions that caused it.

A resonant cavity has dimensions such that waves bounce around within it and you get local hot spots of constructive interference, as well as of course dead zones of destructive interference. If you took high school physics, think of a ripple tank. If not, think your microwave oven. A long time ago, they told you to stop cooking mid way through and move the plate. Now they all have a continuously rotating turntable. That makes sure you are never constantly in either a hot spot or dead zone, and over a complete turntable rotation you get illuminated by the average power out of the magnetron.

Finally, I'm not going to comment directly on health effects of low-level rf radiation. This is because I'm not qualified on the biological side. But I will say that in my opinion, most of the people opining on this subject are also unqualified, because of lack of expertise in biological and/or electromagnetics.

I once listened to an address at a prestigious gathering by someone well known in this field (medical effects of low-level rf radiation), with lots of alphabet soup after his name. His basic point was that human beings evolved in a (relatively) pristine environment, from an rf point of view, and that therefore any man-made rf radiation should be regarded with suspicion. But he hastened to add that there is background "galactic' radiation impinging on earth daily, so you also couldn't go hide within a Faraday cage all day and night, because our ancestors didn't evolve in a complete absence of rf, either.

The underlying argument is precisely the same as that made against street lights way back when: "If God intended there to be light at night, the sun wouldn't set." Or, "If God meant for men to fly, he would have given them wings." Or..., you get the point, this is an age old argument against progress.

All of these health arguments that are based on how our ancestors evolved, and in what sort of environment they lived, fail the acid test of being good predictors of how we ought to live, for one very important reason. Preceding the invention of agriculture, mankind in the hunter-gatherer stage had a life span of about thirty-five years. Absent an accident or fatal illness, and given enough of the basics: food, clothing and shelter, almost everyone lives much longer than that today. This is certainly the case in societies that have enough wealth and leisure time to worry about tertiary health issues such as the effects of low-level rf radiation.

There may indeed be health effects due to low-level rf radiation. Don't know and can't say. But when the basic premise is an appeal to how we lived as "noble savages," it loses all credibility in my eyes. The idea in the web site you cited, about the precautionary principle, is the most anti-technological philosophy I have run across in some time. Any time you do something new and different, there will be side-effects. A normal approach is to look at the effects and make a judgement as to the overall utility of the new device or procedure. For instance, prior to say 1900, how many people do you think were killed in horse and buggy crashes ever year? In the USA alone, we have close to 50,000 annual automotive fatalities. Clearly society has judged the risk and determined that overall automobiles are a net good in our lives. With the precautionary principle in place, we wouldn't have autos, or likely electricity. There are lots of potential problems with electricity - look at all the coal that is burned to produce it, or the risks with nuclear plants that generate it. And electricity can electrocute people. And before we had electricity and oil and natural gas, people burned coal for heat, and towns got stinky and sooty and people had breathing problems. Still, my guess would be that if you offered to liberate some family living in Liverpool, England in the late 1700s from their warm and sooty domicile to a cave somewhere, you wouldn't have had many takers.

Rusty bolt: wow. Is the mixer effect related to rustiness, ie causing more than one freq to be picked up because of discontinuities?

RE: health issues....when the basic premise is an appeal to how we lived as "noble savages," it loses all credibility in my eyes. I agree. Few of us would agree to live without electricity or transportation, we accept the risks. At the same time, people want to know what the realistic risks are, so they can make informed personal decisions about their health and so they can mitigate risks if they consider them serious. Without being anti-tech, and without depriving society of genuine benefits on the basis of groundless fears. Some people who are concerned about cell phone frequencies' effects on the brain are using shielding: better phones can be designed to mitigate risks. And that is ok, it's a good thing.

RE: pristine evolution? There has been far more radiation exposure at certain times over the span of evolution on earth: if I remember (not the evolution itself, just the course) there was very thin atmospheric protection and lots of gamma etc at certain times when the fossil record shows major changes: extinctions and divergence into new species. LIfe goes on.

Re rusty bolt. If you combine two different signals (frequencies) in a linear device, such as a resistor, all you get out is the same two frequencies. A non-linear device acts like a mixer: it creates sum and difference frequencies. The following url explains that the mixer is actually multiplying the two signals, and if you look back at your high school trigonometry identities for sines and cosine combinations you will see that multiplying two signals of different frequency yields sum and difference frequencies:

http://www.electronics-radio.com/articles/radio/receivers/rf-mixer/rf-mixing-basics.php

Wikipedia provides a little more math, including those trig identities:

http://en.wikipedia.org/wiki/Frequency_mixer

Regarding your response on the health effects. If you take what you said, and abstract it; that is, don't look at any other facet of the issue, then what you say makes good sense. But you can never do that. You have to look at the big picture. The big picture is that people don't always (or even often) act rationally in regards to risk taking. For instance, the same person who would sue because the cell phone company didn't adequately warn them about the possible dangers of low-level rf radiation will be driving down the road operating the cell phone and having an animated discussion all the while weaving in and out of traffic and making a menace of him/herself through inattention to his/her driving. How many people who would sue over cell phone radiation in an instant never once drove drunk or had unprotected sex or drove recklessly without wearing their seat belt? And the list could go on and on with behaviors demonstrably dangerous vs. maybe dangerous years down the road, if the behavior is continuous.

I think the real issue is similar to what you said, but different in a subtle but important manner. People want the freedom to take any risks they themselves deem worthwhile, but they don't want anyone else to put them at the slightest risk at all, even if that risk is orders of magnitude lower than what they accept themselves when they drive to work. Regardless of the benefit that might accrue from a product like a cell phone, they want that benefit to come with zero side-effects and the lowest possible cost. When you say that people simply want to be informed, that may be possible for some. I myself barely use a cell phone and try to use it with a wired headset if I will be on it for awhile, to keep the antenna away from my head and body. But there are an awful lot of people who expect a cell phone to be just like a land line, focusing on the phone functionality rather than the radio operation.

In the USA, we live in a society in which technology changes rapidly, while at the same time schools are getting worse and people's understanding of the world around them is much less than that of their forebears at the end of the last century. It is a slam dunk that the average person who graduated from the eighth grade in 1900 had a much better education than the average high school graduate in 2000. If you doubt this, check out the text of McGuffey readers, standard reading texts used on the 1800s and early 1900s. Here's one example:

http://www.howtotutor.com/samples1.htm

Don't know how old you are, or how old your kids are, but I can tell you this is superior fare to what my kids saw in the public school system right through high school graduation.

In closing, I would agree that the way you stated it is the way things ought to be, but not how they actually are.

I think it is fair to say that the government has done its homework and fulfilled the role of setting RF health safety guidelines on the basis of known science. Canada's Safety Code 6 sets limits for human exposure in various contexts, it is clear and easy to read:http://www.hc-sc.gc.ca/ewh-semt/radiation/cons/radiofreq/index-eng.phpAn overview of safety issues for amateur radio operators discusses the science, concerns, and precautions that apply. http://www.sss-mag.com/rfsafety_bkg.html Research on fields applied directly to the head (as in cell phones) is ongoing, since this has raised additional concerns. For recent answers about RF safety standards, this 2008 updated bulletin provides guidelines for assessing RF exposure and safety: http://www.fcc.gov/oet/info/documents/bulletins/#65 -----------

Looking at some real survey data as an example, collected in 2006 using Spectran 6060 frequency/field strength meter, this is what I found, which might be typical for small cities but perhaps lower than built up industrial areas:

Signal levels from commercial broadcast operations in my home area were well below any dangerous exposure limit as defined in Safety Code 6. 102 MHz was chosen as a benchmark for strong commercial signals since it was identified as a strong signal inside my home, measured around -36 dBm: the higher level was attributed to a 'mixing phenomenon' due to the density of commercial signals in that band. In other test locations the strongest level measured for 102 Mhz was -60 dBm. Using a conversion chart to V/m, -36 dBm amounts to 0.10706 V/m compared with the safety level for microwave exposed workers of 60 V/m. In other words, it would have to be about 600 times stronger before it could be defined as a measurable risk, even at the higher level found inside my home. That should give you an idea of the safety margin for typical commercial broadcast RF transmissions to the general public: a huge margin and a very safe level of signal. There was one reading: for 1 Mhz and below (another crowded band) which peaked the meter at 0 dBm and appeared to cause strong aliasing at 5 Mhz intervals. I never did find out what signal that was. Still, at 0 dBm or about 6.75 V/m that is still 100 times weaker than the limits for workers established in safety code 6. I concluded that there was no meaningful level of risk in these typical RF exposures of the general public.

To be honest though, this instrument and another (HF Profi II) made by the same company (www.electrosmog.com) only measure up to a maximum of 0 dBm. For a genuine risk assessment of resonant range frequencies, you would have to be able to measure stronger signals. Arguments by the manufacturer that safety levels are set too high are not so compelling, IMO, if the instrument can only measure safe levels of signal. Marketing it with scare tactics about levels the goverment and scientists consider safe doesn't seem entirely honest... (Got em there, EMC C! )

Something outside ordinary exposure circumstance, with respect to new technologies, is the issue of exposure levels for people with implants that interface with the nervous system (conductive stuff). EMC C, I would appreciate a professional take on this reasoning, as follows.

As an example, neurostimulator (aka microstimulator) devices such as the VNS (Vagus Nerve Stimulator) operate therapeutically at signal levels around 1 mA .http://www.vnstherapy.com/depression hcp/AboutVNSandTRDdosing_strat.aspxAdverse effects (such as choking) are induced by VNS at signal levels at or below the "maximum tolerable level" of 3.5 milliamps.(Vagus nerve stimulation therapy in depression and epilepsy: therapeutic parameter settings. Labiner DM Ahern GL Acta Neurol Scand. 2007 Jan;115(1):23-33.)

http://www.ncbi.nlm.nih.gov/pubmed/17156262?ordinalpos=1&itool=EntrezSystem2.PEntrez.Pubmed.Pubmed_ResultsPanel.Pubmed_DiscoveryPanel.Pubmed_Discovery_RA&linkpos=5&log$=relatedreviews&logdbfrom=pubmed

Suppose a person with a VNS (assume unshielded) came to my home, and encountered a -33 dBm signal, which by the conversion chart amounts to a 4 mA/m field, in excess of 'tolerable' levels for the device, would we expect to see adverse effects? the 0 dBm signal would amount to a 17.9 milliAmp field: well over the risk threshhold if this level of current is received by or induced in the device.

Questions unanswered: Is field strength 1:1 with expected levels of signal induction? would this risk exist only if the transmitted signal is at the specific operating frequency of the device? Or is there a risk of current induction associated with signals at relevant levels in general?

The issues you raise would take a book to answer.

At a very top level:

Established limits on human rf exposure are based on heating of tissue. These levels are well in excess of a radio broadcast, unless you are immediately adjacent to the transmitting antenna. There is not much controversy concerning heating. The controversy is due to claims that ill effects can occur at rf levels well below those that cause heating, hence my continual use of the terminology "low-level rf fields." The claim is disruption of cellular processes, cellular EMI, as it were. No immediate effects are noticed, but the claim is that long term exposure to low-level fields could cause cancer. This is the issue raised with cell phones.

When you start talking about implanted machines, the topic becomes much more tractable. It is easy to establish the susceptibility of the implanted device in the test facility, before implantation. You will know how many Volts per meter of electric field will cause a malfunction. Depending on the frequency(ies) of the susceptibility(ies), you could determine how much that field is attenuated by being placed inside a human body. You would need to know the attenuation per mm of human skin and underlying tissue would apply to the incident field, as well as how deeply implanted the device is.

You will often see signs around a microwave oven in a commercial facility that warn people with pacemakers of the existence of the oven, so they can keep away and not upset their pacemakers.

Your final questions: "Is field strength 1:1 with expected levels of signal induction? would this risk exist only if the transmitted signal is at the specific operating frequency of the device? Or is there a risk of current induction associated with signals at relevant levels in general?"

Here is the answer to part one, the relationship between field intensity and circuit induced potential:

Pr = Pd*G*(wavelength)^2/4 pi

where,

Pr is the power out of the antenna in Watts

Pd is the power density illuminating your antenna (Watts per square meter),

G is antenna gain (numeric), and

wavelength (meters) is 300/frequency (MHz).

You can see that the power delivered to the antenna terminal is dependent on the gain of the antenna. With precisely the same field intensity, you can get widely divergent amounts of power out depending on the antenna used. So there is no way to answer your question in a universal manner. What can be said is that whatever antenna is used with your field strength meter is highly unlikely to model either your body, your cells, or a small implanted device. The only use of your field strength monitor is what its name implies, checking the local field intensity relative to some target field intensity limit.

BTW, the calculation above is in terms of power density, but you can convert from power density to field intensity using the following equations:

Pd = E^2/120 pi

Pd = 120 pi * H^2

where,

Pd is as before

E is electric field intensity in Volts per meter, and

H is magnetic field intensity in Amps per meter

The above relationships are true in the far field of the transmitter, not when you can reach out and touch the transmit antenna.

Here is the answer to part two, "Would this risk exist only if the transmitted signal is at the specific operating frequency of the device? Or is there a risk of current induction associated with signals at relevant levels in general?"

The entire human body operates at sub-audio to maybe low audio frequencies. The transmitted rf spectrum in daily life starts at 530 kHz in the USA and 150 kHz in Europe. Clearly there is no direct relationship between the frequency of operation of victim and culprit field. Because the mechanism of cell EMI is unknown, (at least to me), I can't comment about the interaction of field and biological tissue, aside from heating.

However for the implantable device the interaction is well understood. Take a pacemaker. It monitors your pulse, at around 60 beats per minute (1 Hz) and if something is wrong, it applies a corrective signal, again around 1 Hz. Compare that to the broadcast spectrum. Clearly the interaction is not between the transmit frequency and the victim device directly. Instead, what happens is that the victim equipment acts as an rf detector. It can't react to the rf, which is too fast, but the modulation can be right in-band to victim operation. That is the mechanism by which rf fields cause susceptibility in electronics. The rf is the vehicle by which the modulation travels to and enters the victim, the modulation signal is interpreted as signal by the victim and that causes the susceptibility.

Therefore the precise radio frequency usually isn't a predictor of susceptibility, if a device is susceptible at one frequency, it will usually be susceptible in a range around that frequency. There are exceptions, but that is a good general rule.

ok, if I understand correctly, you are saying that the pacemaker-microwave problem is contingent on absolute proximity to the transmitting antenna (in terms of power density) and so the answer to that is to keep a distance from the mw oven.

this is similar to what is said about extra-low-frequency magnetic fields created by appliances, power lines etc: that the field strength drops off very quickly with any distance from the source, so to avoid exposure, don't place your bed etc on top of a field source.

Pacemakers like other commercial medical implant devices operate in the MICS band around 400-405 MHz. Where the modulation of a strong near-field signal is in band, it can affect the internal signalling of the device so that it delivers an impulse that is not required, or fails to deliver when required.

If I understand correctly, you are saying that a far field signal does not have the same capacity to interfere with internal device signalling. Random far field signals which appear to be strong (as in the survey) do not have the power to cause such interference.

The only exception, then, is where the far field signal is the one designed to communicate remotely with the device to adjust parameters, monitor etc. In this case, the only meaningful signal would be the one at the exact design frequency for remote communication with the device. Signal strength or field strength in this case may not be relevant, except for limiting the distance at which the parameters can be remotely adjusted.

So to the question: is monitoring necessary vis a vis incidental effects on medical devices, the positive answer would be that near field monitoring may be relevant but far field is not.

Re this: "ok, if I understand correctly, you are saying that the pacemaker-microwave problem is contingent on absolute proximity to the transmitting antenna (in terms of power density) and so the answer to that is to keep a distance from the mw oven."

Response: I would expect the pacemaker has a requirement to withstand some degree of microwave radiation. The combination of some basic level of immunity and a safe distance from the oven makes for safe operation. It may also be that a new microwave oven meeting its leakage requirements would not affect a qualified pacemaker. But a banged up oven may have degraded shielding and leak too much to guarantee safety. In that scenario, the warning to keep your distance is an unnecessary margin with a good oven, and a sort of fail-safe precaution if the oven goes bad.

Re this: "This is similar to what is said about extra-low-frequency magnetic fields created by appliances, power lines etc: that the field strength drops off very quickly with any distance from the source, so to avoid exposure, don't place your bed etc on top of a field source."

Response: Similar in result but not in the physics behind it. By its very nature, the magnetic field from an electrically small source falls off as the cube of the distance from it. Electrically small means the dimensions of the radiating element are small compared to a wavelength. So for instance if you have a small electric motor running at 60 Hz, the magnetic field immediately adjacent to it is very high, but by the time you back off just a few inches it has decreased dramatically. Now if you have a single high current 60 Hz feeder separated from its return by a distance close to the same as its separation from you (think overhead power line), and if the feeder length is very long compared to the separation from you, then the field will fall off between the first and second power of the separation. If the conductor and return were infinitely far apart, the field would fall off as the first power, and if they are right next to each other, and you are much farther away than their mutual separation, the field falls off as the square of the distance.

In the case of the leaking microwave oven, you would expect the field intensity to fall off in proportion to distance, that is distance to the first power.

The difference between the first two 60 Hz cases and the microwave case is in the nature of the field itself. In the first two cases, we are talking about the quasi-static field, where the electric and magnetic field lines start and stop on the radiating element, just like the static magnetic field lines of a bar magnet. In fact, if you imagine those field lines of the bar magnet, and then imagine rotating that magnet at 60 Hz, you get the field lines from the electric motor and the overhead power lines. That's why it's called quasi-static.

The microwave radiation is entirely different. It is just like light; it is traveling away from the source which created it and it is completely decoupled from the source. The power density decreases as the square of the distance, because a constant amount of power is spread over some portion of a sphere centered on the emitter. The surface area of a sphere is 4 pi r^2, thus the power density must decrease with the square of the distance. It's just conservation of energy. From the equations of the last post, the square of the electric and magnetic fields are proportional to the power density, therefore these fall off with the first power of distance.

Re this: "Pacemakers like other commercial medical implant devices operate in the MICS band around 400-405 MHz."

Response: I wasn't even aware of that, I had to look it up. More on that later.

Re this: "Where the modulation of a strong near-field signal is in band, it can affect the internal signalling of the device so that it delivers an impulse that is not required, or fails to deliver when required."

Response: Bingo.

Re this: "If I understand correctly, you are saying that a far field signal does not have the same capacity to interfere with internal device signalling. Random far field signals which appear to be strong (as in the survey) do not have the power to cause such interference."

Response: Strong is a relative term. Relative to the sensitivity of the victim receiver. Say you have a 100 microvolt per meter field impinging on a pacemaker which is looking at signals on the order of 1 mV. The pacemaker is an inefficient pickup, so the signal induced into the pacemaker will be much less than the field intensity, and there won't be a problem. Let's look at the effect of the same field impinging on a radio. Let's say we're in the middle of the FM BCB at 100 MHz, and our telescoping rod antenna is set to 0.75 meter in length, which is precisely one quarter wavelength. The antenna is efficient at that electrical length, and in fact the output of a quarter wave stub into a matched load is the wavelength /2 pi. So the radio sees

100 microvolts/meter * (3 m) / (2 pi) = 48 microvolts.

The noise floor of even a cheap FM radio is going to be -90 dBm or lower, or about 17 dB above one microvolt, or lower.

The difference between the incoming signal and the noise floor is over 30 dB, or a factor of 1000.

So the 100 microvolt/meter signal is a strong signal to an FM radio, but a weak one for our pacemaker. We wouldn't want a noise source adjacent to our FM radio to put out 100 uV/m, because it could interfere with the reception of an FM broadcast, but it isn't going to interfere with anything that isn't a radio - that isn't designed to look for it. You can't accidentally be susceptible to 100 uV/meter.

Re this: "The only exception, then, is where the far field signal is the one designed to communicate remotely with the device to adjust parameters, monitor etc. In this case, the only meaningful signal would be the one at the exact design frequency for remote communication with the device. Signal strength or field strength in this case may not be relevant, except for limiting the distance at which the parameters can be remotely adjusted."

Response: Bingo - this is precisely the example just given.

Re this: So to the question: is monitoring necessary vis a vis incidental effects on medical devices, the positive answer would be that near field monitoring may be relevant but far field is not.

Response: Bingo.

thanks. this is a very useful equation.
I see in my old notes that optimal antenna length is wavelength/2. and from a shielding perspective, the obstacle to an electromagnetic wave should be greater than wavelength/10.

would I be correct to generalize your formula for 1/4 wavelength example which is 1/2 optimal, so that the output induced is = (measured signal field strength v/m or A/m) * (wavelength m)/(fraction of optimal antenna length * pi)

(bearing in mind that in specific cases, attenuation may play a significant role in reducing the true value of induced current)

The example of stray signals received on a radio is a good one. Years ago I lived in a house where a specific radio station could always be heard in the background through the speakers of the stereo. I infer that the speaker wires were acting as an antenna and by virtue of the specific length of the wire were 'tuned' to that broadcast frequency. Back in the youthful day, the solution to this was: play your album at top volume. A better solution for 'audible RF' of this type might be to change the length of the speaker wire (tune out the strong signal) or take steps to attenuate the signal reception by shielding the wire?

At present, because of the strong signals inside my house, I have a problem with RFI picked up on cables in general. I have had to deal with bursts of unwanted signals interfering with everything involving listening: the stereo, the computer (with external speakers), and portable recording equipment as well. I found this interference to be generally reduced by using the shortest possible cables. I'm reasoning that a cable which is smaller than wavelength/10 does not pick up a significant level of signal by induction. (Ok I admit it: since short cables are less than optimal for stereo configurations, and I haven't found another solution, I still play my albums at top volume..)

Re this: "would I be correct to generalize your formula for 1/4 wavelength example which is 1/2 optimal, so that the output induced is = (measured signal field strength v/m or A/m) * (wavelength m)/(fraction of optimal antenna length * pi)"

Response: A quarter wave stub working against a ground plane gives exactly the same pattern as a tuned half-wave dipole, but it can only see the half of space above the ground plane, whereas the half-wave dipole can see all of space. The efficiency of each antenna is the same. I wouldn't worry about the difference between a quarter wave stub and a tuned dipole. When you start going smaller than a quarter wave stub, or the dipole becomes shorter than a half-wavelength, then the pattern doesn't change but the efficiency does. The source impedance of each antenna becomes capacitive, which means the shorter the antenna, the higher the impedance, until you can't get anything out of the antenna.

Re this: "The example of stray signals received on a radio is a good one. Years ago I lived in a house where a specific radio station could always be heard in the background through the speakers of the stereo. I infer that the speaker wires were acting as an antenna and by virtue of the specific length of the wire were 'tuned' to that broadcast frequency. Back in the youthful day, the solution to this was: play your album at top volume

. A better solution for 'audible RF' of this type might be to change the length of the speaker wire (tune out the strong signal) or take steps to attenuate the signal reception by shielding the wire?"

Response: Recall that you are listening to the modulation that is somehow being detected by your stereo system. Don't know where it is getting in, but if it is the speaker wires, you can filter it. Recall the rf is at either around 1 MHz (middle of AM BCB, or around 100 MHz (middle of FM BCB). A few tenths of a microfarad across the speaker outputs on your stereo should short out the rf while looking like an open-circuit at audio frequencies. You may also need to try a cap from speaker terminal to ground, can't say without trying it out.

actually, Ghovanloo reports on the same project page 38 of the 2006 report, but the optimal figures are revised: for forward 'data link' he says 10-100 MHz range.. I guess that wipes the 'body as antenna' speculation for this one.