Should the Population be Concerned About an Electromagnetic Pulse (EMP)?

The reliance of society on electronics invokes fear in terms of what happens when an Electromagnetic Pulse, or EMP disables these electronics. This conversation came up recently. Along with the 75th anniversary edition of the WWII Jeep, this made for a combination that I thought would be a good debate for Planet Analog readers.

The person who initiated the conversation said that “it kept him up at night and kept the director of the CIA up at night as well”. Granted, this person did not seem to be an engineer so there were obvious limitations to his concerns. I don’t think he understood the inverse square law or how a Faraday shield works. This was reflected in his statement that discharging a pulse over Nebraska was centralized enough to take out the US. Regardless of his technical limitations, as engineers we rely on electronics for our livelihood and to improve our lives. So “what if” the pulse did in fact occur? How would it change life as we know it today?

Before we get into that discussion, the anniversary edition of the WWII Jeep offered me an opportunity to write an article in one of my other areas of interest, four wheel drive vehicles. It also offered me a chance to revisit vehicle technology during that time when ignition systems were point based and thus less vulnerable to EMP. I began to wonder about how to combine the threat of EMP on everyday life with how it would affect transportation as we know it today.

There is no doubt that automobile function and navigation rely heavily on electronics that are semiconductor based. Although the right amount of EMP could potentially pop most types of electronics, semiconductors are especially vulnerable. As I sat in on a seminar for converting my Scout to electronic ignition and fuel injection, I envied the thought of increasing my gas mileage above 15 MPG, a level virtually unknown in early four wheel drive vehicles. I also thought of EMP and what the current carburetor and point based ignition system meant in terms of surviving an event.

As my thoughts progressed I realized if I kept the original system then I would have one of the few vehicles that was still operating after the pulse hit. I could charge any Uber price I wanted. Then again, the telecomm infrastructure and most phones might not be operational so nobody could contact me to begin with. Another get rich quick opportunity lost. My Scout’s value still rests in its lack of rust and not in being the only vehicle still in operation. Back to square one.

From here this blog diverges from the value of my vehicle and the effects of EMP on transportation. As a result of searching the subject, the results had three different themes: ‘what will EMP “fry”?’, how to prepare for EMP, and the usual Wiki article that interestingly enough focused on facts and myths. Wow. What a diverse set of topics. Perhaps I’ve opened a can of worms that is best left to a series on EMP and its impact on society. Certainly this wide variety can’t be covered in a single blog. However, we can dedicate a paragraph to each result and further introduce the theme it projected.

The article mentioned above with the subject of ‘what will EMP “fry”?’ was a very good technical assessment that originated with a concern for a GM electronic ignition system. This article was the only one of the three that focused on transportation. It also did a good job of going beyond transportation to a general analogy of the effects of EMP on society at a high level.

With regards to the article on how to prepare for EMP focused on survival, I’m not big on conspiracies or preparedness so this subject will be left to the audience to decide whether it’s worthy of debate. If so, I’m happy to address it further.

Finally, the Wiki article on myths and facts, was of the most interest to me. I guess that’s because it seems to diverge from the typical Wiki explanation of a subject more towards the manner in which it is viewed by the public. Perhaps we could take the subject there.

This blog jumps around like a movie without a plot not by accident, rather it is presented this way on purpose. In typical form I have thrown a bunch of subjects out so as to see how the audience perceives this topic. I hope you have read enough of my blogs to realize that I not only try to present subjects I am knowledgeable in, I also like to start debates on subjects that are of interest to analog engineers.

This topic in particular could go in many different directions and could have many different titles. Nobody is currently at the helm. Feel free to steer this towards the subject of your choice. If I haven’t presented a subject that interests you, introduce your own. If it concerns you, it probably concerns other engineers. I’m just the catalyst that starts the reaction. I hope it doesn’t keep you up at night. Yikes! I hope Beavis and Butthead don’t respond to that closing statement. Mike Judge, spare me…hent, hent. Huh huh, you jutht thaid…..


  1. ‘what will EMP “fry”?’
  2. how to prepare for EMP
  3. Wiki article
  4. four wheel drive vehicles

8 comments on “Should the Population be Concerned About an Electromagnetic Pulse (EMP)?

  1. j.sinnett
    July 27, 2016

    If there's an electromagnetic field so strong that it can melt semiconductor junctions in equipment that's not connected to the power grid, what else will it fry?  Consider that typical “antennae” on off-grid electronics have dimensions measured in centimeters (radios, cellphones, etc. which have intentional antennae) or unintentional antennae of at most a meter or two (desktops, ignition systems).  So let's say we have a device with connected wires of 2 meters.  Now we're saying that the EMP dumps enough energy into the system to melt semiconductor junctions.  In a pulse that lasts probably less than a microsecond, the pulse has to deliver power at a rate of thousands or maybe millions of watts (1 microsecond at 1megawatt = 1 joule)

    Now think about this.  An adult human body is roughly 1.5 to 2 meters long.  If there's an EMP impinging on our body, isn't it going to do some damage to our nervous system?  Much smaller amounts of energy are sometimes used to partially “reset” the “programming” of people suffering certain mental illnesses.

    And what about metal in our dwellings – won't there be sparks from our Venetian blinds, gutters, etc. etc. that will spark fires?

    Basically, the EMP sounds like a simultaneous lightning strike on everything within reach – people, trees, buildings, livestock, vehicles… I can't imagine that it's relatively benign and ONLY knocks out semiconductor electronics and large-scale power grids while leaving everything else alone.  (But no, I have not read up thoroughly on the subject, so maybe this is an ignorant position.)

  2. Effective-Technical-Writing dot com
    July 27, 2016

    Thank you for your reply.  Semiconductors were referenced as they are the typical microelectronic component that is relied upon.  In the way we dope electrons and create junctions, we create opportunities for the pulse to affect them rather easily.  That being said, the tungsten filament in a headlight will “pop” from a rapid temperature change when being switched on which happened to me recently.  As for the pulse, it all depends on how it is coupled by the electronic component.  There are millions of possible scenarios.  Punch through of a tiny digital gate is one of the more likely ones.  In the quest to make items smaller, lighter, and more portable, many of these items are encased in plastic thus lacking a metal Faraday shield to distribute the charge that results.  Reference 1 goes into more of the likely scenarios.

  3. RadioGraybeard
    July 27, 2016

    This is a messy topic.  There is so much, to be blunt, crap distributed around on the subject that it's hard to know things for sure.  It's hard to find reasonable summaries.  Much of the rest of the real information is classified.

    To begin with: it's an EM field, it's not magic.  It can be quantified and analyzed like every other EMI/EMC problem, and EMP is just a special case of EMI/EMC.  Do an FFT of the waveform and you'll find that most of the energy is below 100 MHz.  Like any antenna problem, full-sized antennas capture more energy than small antennas and bigger current loops pick up more than smaller loops.  These are very high voltage pulses but they last nanoseconds – the standard model is 50kV in 5 nsec, decaying to 1% in 100nsec, but it's a very manageable problem.

    There have been experiments published where cars were exposed to the pulse of a nuclear EMP (NEMP) and were fine.  Some stalled, but worked when restarted.  These were modern cars, not '60s relics with mechanical spark systems.  Many cars ran through the event.  Likewise the American Radio Relay League (the national amateur radio organization) has published radio tests and typical VHF/UHF radios hams use were unafected.  If you run the numbers, you can see why: the EMP put -16 dBm into a 146 MHz radio.  -16 is a big signal to receiver designers, but it's not burning anything out.  It's 25 microwatts!  HF radios will see higher signals, but you want to know how to minimize damage?  Bandpass filters on the radio to limit the energy coming through.  Even if the filter is a resonant antenna.  Maybe we're talking milliwatts here, but only milliwatts.  Passive filter components won't even shrug.  If anything gets hurt, it's very broadband, frequency hopping systems like the military uses. 

    Here's another surprise: even with that 50 kV pulse for 5 nsec, the E fields are tiny: we're talking .002 V/meter.  In commercial avionics, we routinely tested to 200 Volts/meter over much greater swaths of spectrum than an EMP generates, and you gotta know military systems are tested to higher levels.  That junk you read about airplanes falling out of the sky?  Yeah, just ain't gonna happen.  And in the same “ain't gonna happen” category: broadcast radio and TV isn't going down.  Add a few milliwatts to the thousands of watts coming out of the transmitter?  How would the transmitter even know?  I don't care if you add a few whole watts – that's not even a VSWR to a broadcaster. 

    The real risk I see is to the power grid, and that may be because I spent life as an RF designer and not a power engineer.  But the way I do the math, the power grid is one huge antenna and it's where the most energy in an EMP is.  (Same for the POTS phone grid).  I can see the power grid going down – but I've also read they've been working on this on their own, quietly, in an effort to improve service.  So, unlike everyone else, I can see a world where virtually everything we own works, but the power grid goes down.  


  4. j.sinnett
    July 27, 2016

    Excuse me, but Wikipedia says nuclear EMP is 50,000 volts per meter, not 0.002 volts per meter.  An A-bomb test over the pacific did substantial disruption to systems in Hawaii, 1445km away.  Can you provide a reference saying that nuclear EMP is only 0.002 volts per meter?  If that's the case, I'm not worried at all.

  5. RadioGraybeard
    July 27, 2016

    I wasn't clear on that.  I left out too much in effort to not run too long.  Yes, you start with 50kV/m – that's the peak field in what seems to be a standard model.  I have an old app note from ANSOFT that I've long used as the only hard numbers I can get on that.  They derive (curve fit?) a closed form exponential equation to model the pulse.  (I've been trying to find the ApNote online so I could link to it.  I'll keep trying.)

    If you do an FFT on that pulse to put it in the frequency domain, you turn it into the field strength in V/m/Hz.  The numbers come down to the vicinity of 2mV/m/Hz at maximum (actually around 1.5 mV/m/Hz).  I left out the BW dependency, and you need to multiply that times the usable BW, and I left that out.  By the time you get up to where critical avionics systems are, just above 1 GHz, which is what I was thinking of, that field goes down.  At 1 GHz, that field goes down to under 1 microvolt/m/Hz.  In the bandwidth those systems use that's still way under the levels the boxes are qualified to.

    Perhaps of more interest is that 99% of the energy is within DC to 240 MHz. 

    As usual, it would be better to think through a response, edit the heck out of it to make it shorter, and post, rather than just post.

  6. Victor Lorenzo
    July 30, 2016

    Using the power spectrum we may determine the amount of energy to be received by the system receiving the impulse and, of course, it will be a fraction of the total energy generated during the formation of the several components of the EMP, but real systems don't tend to have such a narrow bandwidth as to consider only the surge components in the vecinity of a single frequency.

    On the other side EMP pulses have two closely related components, the electric field and the magnetic field. Both of them travel and attenuate in different manners. The electric field has more local effects and, depending on the charges distribution and movement after the ionization occurred during the first ps/ns, will exhibit an attenuation inversily proportional to r^3. In the case of the magnetic field it will have an attenuation inversely proportional to r (being r the distance).

    Will they fry things? Yes, definitively, but only some things. In my opinion (not a physics expert) the effect of the fast changing electric field will have severe efects on charge carriers at silicon die levels in the relative vecinity of the EMP, maybe up to 100km depending on the EMP magnitude and altitude, but the largest damages will be due to surges induced by the also fast changing magnetic field, withing a probably up to 10 times larger range.

    I have seen the effect of both, electric and magnetic, fields over electronic circuits in our high voltage lab. If I find the time I will post a blog with practical results and damage samples in the series about lightning.

  7. RadioGraybeard
    July 30, 2016

    I would love to see any real data you could post or point to somewhere else.  I've spent tons of time in EMI and lightning testing labs, but have never worked with really high voltage. 

    Regarding the different rates at which the E and H fields decay during propagation, from what I've seen in EMI testing, that only seemed to matter in the near field, within a wavelength or so (real precise, I know.  Sorry).  I think what happens at higher frequencies when the target is several wavelengths from the radiator is that normal EM propagation dominates (decaying electric fields produce magnetic fields and decaying magnetic fields produces electric fields – just as we all learned in school).  

    From 300 miles up (the number thrown around for the altitude to blow an EMP) to the surface is one wavelength at~620 Hz, but that's straight line so just a rough guideline to saying different attenuations for magnetic vs electric fields will be really important at powerline and phone line frequencies, but in the AM Broadcast band and up, straight inverse squared law will dominate.  That's my take, anyway.

    My disclaimer is that I'm an RF engineer, spent most of my career in receiver design from 500 kHz to 10 GHz.  As the RF guy in a building full of digital guys I'd spend lots of time in EMI/lightning testing.  I've never worked on EMP problems. 


  8. Victor Lorenzo
    August 1, 2016

    Hi RadioGraybeard, I should say just like you that “I've never worked on EMP problems”. I mostly deal with lightning monitoring and protection. I also collaborate with our high voltage lab in the instrumentation and also in designing and constructing special purpose voltage and current impulse generators. I can't say I'm the RF guy, in this case I'm closer to be the digital/analog guy which still jumps off his chair whenever the current impulse generator is fired ;). Of my recurrent questions after each impulse (current or voltage) over any of our designs are Did it survive? Still working on specs?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.