Repost: Simulations December 29, 2010Posted by mareserinitatis in computers, electromagnetics, engineering, geophysics, research, science.
Tags: electromagnetics, simulations
After reading this post and participating in the discussion, I felt that perhaps reposting this from the old blog was in order.
After posting this morning about how I hate computers, I figured I should temper that.
One thing I hear an awful lot of is how people don’t trust simulations. (They also don’t trust math, but let’s take one thing at a time.)
An awful lot of science can be done through simulations. However, as soon as you tell someone that you got your science out of a computer program that feeds you data or makes pretty pictures, you may have just said you made science with your kid chemistry set and drew your data in crayons.
Skepticism about computer methods is a good thing as long as you know where to draw the line. A couple years ago, I went to a tutorial session on different computation methods used in electromagnetic compatibility (EMC). At the end of the tutorial, a spontaneous discussion about the reliability, drawbacks, and validation of simulations came up. I’ll summarize some of the main points and talk about how I have addressed them.
I guess the first thing to address is that there are many different methods to simulate things, and these methods have drawbacks. As an example from electromagnetics (EM), folks often use something called Finite Element Method (FEM). FEM is not unique as an EM tool…it was actually first developed to examine mechanical engineering problems (think stress and strain). It works very well for electromagnetics as well, with one caveat: whatever your modeling needs to be enclosed. If you don’t have an enclosed area (say a shielded box over a circuit), FEM can’t mesh space infinitely. There are methods that have been developed to deal with items when are radiating in open-space. One is called a Perfectly Matched Layer (PML) which matches the impedance your radiator sees at the edge of the space and then attenuates the field beyond that area.
I give this example because, as someone who has worked with antennas using FEM-based software, it’s important to understand these things. I didn’t, at first, and it took a lot of work to figure out if the software was even simulating correctly.
How did I do it? I used a method that everyone who is a good simulation researcher does: I validated my simulations. In antennas, I started out by modeling simple known devices to see if the results matched the theoretical value. Since the equations to compute these values are based on the same equations as the theoretical value, they should be pretty close. Next, as my devices increased in complexity, I used another computational EM code called Method of Moments (MoM). MoM is awesome because it works differently than FEM. FEM jumps straight into calculating fields while MoM calculates the currents on an antenna (for example) and then is able to compute field at any given point. Once I was able to get simulations that matched either an analytical result or the other code, I could be fairly certain that I’d gotten the kinks out.
Researchers in other areas (say, global climate change) validate as well. While I would assume their approach would have to accurately reflect any analytical results, they can validate more complex code by seeing if their code generates something fairly similar to actual events and known history.
The final step for validation, in my experience, is to take the code and run it using an example of something more complicated. Usually, this is the point where you start looking for interesting journal articles to reproduce.
Now, in all fairness, I know that people don’t always follow these procedures, which is where I believe people should start to be skeptical of results. In fact, the last step of validation can be the hardest even though it’s probably the most important. I know that in my short life time in computational electromagnetics, I’ve had the misfortune of coming across papers which predicted a result, but it’s totally different from my results. In a couple cases, I ended up writing authors to find out that they had misprinted some dimensions on something On the other hand, you don’t want to pursue that route until you’ve exhausted all your other options. In my case, moving part of a device by a just a few millimeters (at high frequencies, a significant chunk of a wavelength) changed the resonance frequency of the entire device. That’s why learning how best to utilize built-in placement functions rather than hand entering things is preferable.
However, those papers aren’t all that common (I hope…but I can say I haven’t hit too many). More often than not, good researchers have tried to test their code to make sure it is accurate and representative of that which they are trying to model. They have also reproduced previous known results to show that their method is sound.
The next time someone tries to tell you it’s just a model, you can reply by asking them how much they know about code validation. If you read this entire post, there’s a good chance you’ll know more about it than they do.
Getting hot under the sheets August 1, 2010Posted by mareserinitatis in electromagnetics, engineering, physics, science.
Tags: bed, cancer, electromagnetics
A couple days ago, I wrote about a post which was a knee-jerk reaction to an article I’d read on Sci-Am. The actual paper is “Sleep on the right side—Get cancer on the left?” by Örjan Hallberg and Olle Johansson, published June, 2010 in the journal Pathophysiology.
Now that I’ve had the time to sit down and read it, I’m no less boggled. There are several problems with the paper.
As far as the paper itself, it does absolutely nothing in the way of taking measurements or providing simulation data. The data upon which they are basing their argument is this: there have been studies showing that there is a higher incidence of certain cancers on the left side of the body than the right. Also, people have a preference for sleeping on the right side of the body. Therefore, the two are connected together by beds and they state a proposed relationship between beds and cancer with absolutely no supporting evidence aside from those studies. In other words, they have correlation between cancer and preference for sleeping on the right side. That is the entirety of the empirical data.
First, the authors state that preference for sleeping on the right side doesn’t necessarily imply that people spend more time sleeping on their right side. Second, there are also anti-correlative relationships between other types of cancers and disorders and side-sleeping. Right there I would be scratching my head and wondering about the validity of any correlation.
I mean, here we have another correlation:
Okay…maybe they’re anti-correlated.
However, I’d like to focus on two excerpts from the paper in particular. The introductory paragraph states:
There is no doubt that horizontally polarised radiation
emitted from western-type FM broadcasting towers may
interfere with a human body at night during sleep. This is
a logical consequence of wavelengths normally used for FM
and TV broadcasting. A metal-containing mattress may act
as a half-wave antenna, thus the central parts of the body may
become more exposed to resonant currents than the head and
In the results and discussion section, they state:
Basic physics tells us that a structure that is approximately
1.5 m long may become resonant (half-wave resonance) at
frequencies of approximately 100 MHz, since half a wave-
length is 1.5 m. Half-wave resonance reaches a maximum
in the middle of the structure. If this structure is resting on
(above) a metal bed and is exposed to horizontally polarised
waves, the incident wave and a reﬂected wave will produce
a standing wave pattern. The ﬁeld strength will be almost
zero close to the surface of the metal base, but reach a max-
imum a quarter wavelength (75 cm) above the metal base,
another zero at 1.5 m above the metal base and so on. The
ﬁeld strength at the maximum is twice the ﬁeld strength if
the bed was not there, due to summation of the incident and
reﬂected waves. Without such a metal base there will be no
standing waves driving currents along and inside the body.
I am bothered because there are absolutely no references for any of the above assertions. How does it interfere? What’s the mechanism? Where is the data confirming this?
When looking at the electromagnetics, I am seeing the author set up two different scenarios, and I don’t think they realize they are doing so. The first excerpt states that the metal bed frame is behaving like a quarter-wave antenna with a current distribution strongest in the middle of the bed (near the torso) and weakest near the head and feet.
In the second excerpt, they’re saying that you have a resonator (a human being) above a ground plane, which is reflecting. Let’s look at both scenarios, beginning with the second.
Let’s think about the bed like a hunk of metal floating in space and, for a moment, ignore the person lying on top of it. I made the mistake before of thinking about the mesh and springs separately. Really, they’re all so interconnected and the spaces between them is small relative to the wavelength, we can treat it like a large piece of metal. So really, it’s a scatterer/reflector or possibly a ground plane. Either one will reflect a field that impinges on it.
We’re going to assume plane wave reflection. The angle of incidence will be the same as the angle of reflection.
If you look at the graphic above (which I borrowed from the hyperphysics website…you should go visit there), you can see that there is an interference pattern between the incident and reflected waves. The places where the blue lines cross or you can see the background will be constructive interference (the crests or troughs will add, making a larger wave) versus the place where a blue line and tan cross, where we have destructive interference. The authors claim that constructive interference is a problem in this scenario.
We can actually use simple geometry to determine where the constructive interference will occur. It turns out that the formula is d = λ/(4 sin θi) where d is the distance above the plane, λ is the wavelength, and θi is the incident angle.
The authors of the paper state that this interference will be at a maximum (i.e. the wavefronts will be in phase or produce constructive interference) at 75 cm above the bed. However, that’s only true if the angle of incidence is 90°. That means the radio waves would have to be coming straight down onto the bed. Not likely.
Let’s assume that a 100 m tall radio tower is about 4 miles away (or about 6500 meters), the actual angle, neglecting curvature of the Earth (which we frequently do in Fargo), is closer to 0.88°. The maximum in that case will occur at 1200 m above the bed. Unless you live in a fairly tall building (like skyscraper tall), I think that’s well out of the range of your bedroom.
If we give the authors a bit and assume that maybe the transmitter is on a hill someplace, an angle of 10° says the maximum will occur about 5.75 m above the bed…probably in the apartment two stories above you. (I hope you feel guilty for affecting your neighbors upstairs this way.)
Keep in mind, however, that we do not have an infinite ground plane; it is only about a wavelength long. Constructive interference only happens above the bed at angles within 45° relative to the normal of the bed. A low angles, like I’ve examined here, the constructive interference will actually happen somewhere away from the bed because it has to occur past the point where the wave is reflected. Therefore, a higher incidence of cancer in the torso versus the head and feet would actually imply that constructive interference probably does not play a role.
Now that we’ve ruled out constructive interference from reflection, let’s look at the possibility that the bed is behaving like an antenna. It can’t be because it’s isolated from any sort of path that would remove energy. But let’s ignore that (significant) detail for a moment.
Now it may be true that an antenna may be receiving power and, in this process, re-radiate some of the power. Let’s make the assumption that the bed is, in fact, acting like an antenna that is absorbing everything that it receives and then re-radiating everything. (This is also slightly unrealistic because, if the body is absorbing the wave, the wave cannot also be passing through to the bed, which will re-radiate it. Let’s not let reality stop us from believing this, though.) We need to compute how much power a person would actually be receiving from an antenna.
All we need is the Friis transmission equation.
Assuming an antenna 4 miles away, transmitting at a frequency of 100 MHz (giving a wavelength of 3 meters, as the authors assumed), a max EIRP of 100 kW (this is the product of the transmitter gain and power), and a perfectly absorbing human with a gain of 1, a person would receive approximately 136 uW (microwatts) over your entire body. This is equivalent to 136 uJ per second. However, we are going to assume that it’s twice that because we have both the initial wave as well as the wave radiated by the bed springs.
The most plausible mechanism for cell damage by electromagnetic non-ionizing radiation is by heating, so let’s figure out how much this energy will heat you up. Of course, this assumes that all of the electromagnetic energy will be converted to thermal energy, which is highly unlikely. Given our bodies are mostly water, and the lowest resonant frequency of water is 2.5 GHz (which is the operating point for most household microwave ovens), I’m not exactly sure any of this would be converted to thermal energy. Remember that it takes 4.184 Joules to heat one gram of water by 1°C. This is the specific heat of water or c. The average mass (m) of a US citizen is about 80 kg. Using the specific heat equation,
this means that our body will heat up 8*10-10°C per second or 48*10-9°C per minute. By contrast, standing out in the sun gives you 1.96 calories or about 8 Joules per minute, changing your body temp by 2.4*10-5°C per minute. This is, at least, three orders of magnitude larger.
Sometimes it gets a little hot under the sheets, but I don’t think it’s due to radiation from TV and radio antennae.
In any case, the bed hypothesis doesn’t make sense because there is no way for it to work. If the bed is behaving like a reflector, the constructive interference is so far away for a realistic case as to make it absurd to propose a relationship. In both the case of a reflector or antenna, there is so little power that it couldn’t do anything. Therefore, I am certain that, if there is a relationship between side-sleeping and laterality of cancer, electromagnetics plays no role.
Your bed may be killing you… July 28, 2010Posted by mareserinitatis in electromagnetics, science.
Tags: antennas, beds, electromagnetics, physiology, scientific american
(Note: I posted a much more complete analysis of the paper here.)
Scientific American posted a brief review of some research stating there may be a correlation between metal bed springs and incidence of certain types of cancers, such as cancer of the breast and melanoma.
The person who wrote the SciAm article obviously didn’t read the original paper very well…or even the abstract.
In the Sci Am article, it is stated early on that there have been correlative studies showing that there is a statistical relationship between incidence of cancer and proximity to transmission antennas for radio and television.
My personal interpretation of this data is that people who live in cities are more likely to get cancer than those who don’t. This could very well be a lifestyle issue because, as we all know, correlation is not causation. And transmission towers are going to be considerably more dense in urban rather than rural areas.
The article makes the point that the amount of electromagnetic energy is so small as to be nearly irrelevant.
Anyway, the claim is that the springs in a person’s bed are capable of capturing and amplifying this electromagnetic energy, just like an antenna.
Thus, as we sleep on our coil-spring mattresses, we are in effect sleeping on an antenna that amplifies the intensity of the broadcast FM/TV radiation. Asleep on these antennas, our bodies are exposed to the amplified electromagnetic radiation for a third of our life spans. As we slumber on a metal coil-spring mattress, a wave of electromagnetic radiation envelops our bodies so that the maximum strength of the field develops 75 centimeters above the mattress in the middle of our bodies. When sleeping on the right side, the body’s left side will thereby be exposed to field strength about twice as strong as what the right side absorbs.
The solution, obviously, is that one should get rid of their metal bed springs.
Without even running numbers, I can tell you that this makes no sense. Antennas are passive objects. Amplifiers, however, are not passive and require the input of additional power. An antenna cannot amplify a signal. In fact, most antennas, including the one in your car, require an external amplifier so that you can listen to the signal.
Unless you’ve figured out a way to plug in your bed springs, I don’t think it’s going to do much to the signal other than capture it passively. I cannot generate a field any stronger than the one it is receiving…and generally, if it’s receiving, it’s not going to be transmitting anything.
The other problem I see, from an electromagnetics perspective, is that it just isn’t the right shape to work as a good antenna. If you look at the wire mesh in a box spring, that may work as a decent receiving antenna. However, your mesh is covered with springs and metal springs are basically going to act like inductors from an electromagnetics perspective. That is, if you have a time-varying field from the mesh, as it tries to move upwards through the springs, it will induce a current in the springs. This current, however, will generate a magnetic field which will oppose the current that induces it. If anything, I would think that it would reduce the fields around a person.
Of course, the easy way to check this would be to actually use an antenna and take measurements of the ambient field as you move toward the bed. Curious, I decided to check the original article to see if that was attempted.
It turns out that the SciAm article is completely wrong. The original article abstract states:
We found that people tend to sleep for longer periods on their right side, apparently to avoid disturbance by the heartbeat. This puts the left side farther away from the field-attenuating influence of the metal springs in the mattress; thus the left side will spend, on average, more time exposed to stronger combined fields from incident and reflected waves. This hypothesis may also explain why body parts farthest away from the mattress (trunk and upper arms for men; lower limbs and hips for women) have higher melanoma rates than the sun-exposed face area. The implications of this study should promote a critical consideration of population exposure to electromagnetic fields, especially during the night.
This is completely the opposite of what the review in SciAm stated. They are saying what I just said: the bed springs will actually reduce (attenuate) the field. They also posit that the real danger comes from the left side being away from the bed and thus exposed to EM radiation.
I still don’t know that I buy the bed-spring argument. One could argue that the safest place to sleep would be in a giant Faraday cage. Frankly, the fields transmitted are pretty weak (especially with the conversion to digital). Also, if you need to put an antenna on the top of (i.e. outside) your house to get a good signal in many cases (which needs to be amplified by the electronics in your TV), the fields are obviously attenuated by your house to some extent. Are you really going to get much electromagnetic energy hitting your body? I personally am inclined to think that there may be some other physiological factor that has to do with sleeping position but not necessarily emag.
And yes, the fact that Japanese use different beds may very well affect sleeping position which may, in turn, affect left-side prevalence in certain cancers. To assume that they must sleep the same way is a bit silly. (I say this as a side sleeper, but when I sleep on the ground, such as when I’m camping, I’m far more comfortable sleeping on my back.)
Regardless, the Sci Am summary of the journal seems to be, by all appearances, completely backwards in it’s understanding of the original research…or is doing a very creative interpretation.
Tags: circuits, electromagnetics, hygiene, lightning
We’re having a thunderstorm, so I thought it was a good idea to pull this post from the old blog.
We’re having a thunderstorm right now. I’m supposed to be working on my thesis (I have decided that I will not go to bed until I am either thoroughly exhausted or have completed the changes on a couple chapters).
Being the highly distractable person I am when I am not happy doing something, I have been caught up in pondering the quantitative values that surround a “human resistor” in the presence of a lightning strike. I’m not concerned about the case where the person is the direct path to ground. I’m thinking about a comment someone made to me a couple weeks ago: You should never shower during a thunderstorm in case the lightning hits the plumbing. Here I always thought it was because you’d fall out of the tub when you lose power.
I have embraced a small part of me that thinks I am an engineer because I want to set up a simple model to determine what sort of currents would pass through a person.
So if you’re curious, here’s my solution…
This is going to be pretty basic model, since I really should get back to work. The source is the lightning, obviously, but we’re going to assume it’s a current source. It’s passing through two resistors in parallel to ground (literal ground, in this case…not too often that happens). Based on this info, let’s say the current is 10 kAmps.
According to (one particular) ESD human body model (HBM), the human body has a resistance of 1500 Ohms and a capacitance of 100 pF. I’m going to do something completely bogus because this would take a long time otherwise. The impedance due to the capacitance is going to be zero for this solution. I realize that lightning is a transient and is almost totally high frequency content…knowledge of which would take me a while to dig up. I’m going to ignore that knowledge, though, because I really don’t want to try to do anything close to Fourier analysis to figure this out. I’m lazy and I only want to look at the resistance. Aside from that, we’d have to have THz frequencies to provide any appreciable impedance.
The other resistance will be the plumbing from your shower head to the drain. We’ll assume a 3 cm diameter copper pipe that’s 2 meters tall. We’ll use this info to compute our resistance. The pipe resistance will be 17 nOhm m *length/area = 17e-9*6/(π*.0015^2)=14 mOhms. Now, I also realize that it’s not a solid copper pipe because (obviously) it’s filled with water. On the other hand, copper is lower resistance, so not as much energy will flow in the water as in the pipe. So I’m going to pretend that it is, in fact, a solid copper pipe, realizing that the resistance, in reality, would probably be higher (which is bad for the person in the shower…more resistance in the pipe means more current is shunted to the other resistor, the person).
(Note: in re-reading this, the counter to this argument is that a water filled pipe creates more surface area, i.e. the inside surface of the pipe, and so could potentially reduce the resistance of the pipe. However, the water inside the pipe will, as I said, have a higher impedance than the pipe. I’ll assume both effects are negligible.)
So we have two parallel resistors, one of 1500 Ohms and the other at 14 mOhms, giving us approximately six orders of magnitude difference. The equivalent resistance (R) will be 21 Ohms. Given a current (I) of 10 kAmps, we can whip out Ohm’s Law (V=IR) to find that the voltage drop across our human/shower parallel resistance is 210 kV.
But since we all know that current is what really kills, we need to figure out if enough current will be shunted through the pipes to make this a serious concern. Initially, it looks good since the person is such a high resistance compared to the pipe. We should note, however, that 100 milliamps is enough to shut down the heart.
What we need here is a current divider:
Ihuman=itotal*(Rshower/Rtotal)=10 kAmps * (.014/1500.014)=93 mA
Crap. At the very least that will feel terribly unpleasant, but with that sort of voltage, I imagine it could do a lot more. The deciding issues would be those little assumptions I mentioned above (the accuracy of the HBM model and the oversimplified impedance of the pipe).
Looking at this article, I’m inclined to think it might be a low estimate. This lightning strike was 7 miles away!
So yeah…no more showers during storms.