Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Q: Scavenging energy from EM-fields, general ideas

  1. Oct 9, 2011 #1
    Hey everyone!

    I'm looking for general ideas on how to take advantages of high EM-fields created in for example , lets say, a transformer to energize a small sensor. I am not looking for specific ideas like placement, voltages etc.

    The reason for me asking is that this probably will be the topic of my master thesis, I'm heading for an interview on Wednesday and I want to would like to be as prepared as possible as I can because I know the guy will ask about how it can be done and which one would be the best fit for this project, or maybe try all the methods to see which one is the best suited for different situations.

    So help me out guys, your help is much appreciated!
  2. jcsd
  3. Oct 9, 2011 #2

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    i used to work in a power plant.

    the machinery makes surprisingly little external magnetic field.
    to put it simply: if the machine is letting its magnetism get out there's something bad wrong with it .
    Exceptions are air core inductors that are used to limit fault current in buswork, and the main three phase wires coming out of the generator. Both had substantial 60 hz fields because they were carrying kiloamps.. I could get several tenths of a volt induced in my "flux detector " whhich was a coil of ten turns encompassing 0.1 m^2. I suppose a a Hall sensor would have worked also. Nowadays you can buy solidstate gauss detectors.

    Another troubleshooting hint is to take a cheap AM transistor radio, tune it netween stations and simply hold it to your ear as you walk around the machinery. Any corona from leaky insulation will announce itself on your radio but it's difficult to hear when standing adjacent a noisy several thousand hp motor. Headphones help. Again premise is, if the machine is electrically noisy it needs some attention.
    That one you should try yourself - take a cheap transistor radio and walk the power lines in your neighborhood early in the morning when everything is wet with dew. You'll find the dirty insulators.

    Lastly there were some IEEE papers in 1970's about noise masurement in central station generators. They placed a donut style current transformer around neutral lead of the generator and measured its RF noise in vicinity of 1 mhz. An increase would indicate insulation is degrading someplace allowing corona. Now 1 mhz is right in middle of AM broadcast band so they had to use a narrowly tuned voltmeter. The transistor radio technique i described earlier is poor man's version of that instrument.

    in summary - i will be surprised if you can scavenge energy enough to use for powering instruments or charging batteries
    but there's plenty to convey information about the machine.

    i hope this helps with your interview.
    Good luck in your studies.. i wish i were more academic.

    old jim
    Last edited: Oct 9, 2011
  4. Oct 9, 2011 #3


    User Avatar

    Staff: Mentor

    You may already know this, but "Energy Harvesting" has gotten more attention the past few years, because of the potential uses for small RF sensors/devices:


    The more energy you can harvest from light, vibration, external fields (your example), the longer your device's batteries will last...
  5. Oct 9, 2011 #4


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The term 'harvesting' is very loaded. It conjours up visions of limitless supplies of free energy, there for the taking. I don't think that reality is as good as that.

    Low levels of em or thermal energy are very difficult to use. In the case of 'heat harvesting' you would be dealing with a device that would be, basically, a heat engine, working with source and sink temperatures which would be very little different.This implies extremely low thermodynamic efficiency - less than could be gained by just improving the system from which you are trying to harvest the waste energy. Low levels of em energy would need to involve large structures in the form of coils or long wires which would need to be matched to some form of detecting device in order to, perhaps, charge batteries. Very low initial voltages, followed by the unavoidable 'diode drops' would surely militate against worthwhile energy gathering per unit of money spent on the equipment. Again, more juice in the lemon of improving electrical efficiency in the first place.
    The actual numbers count in matters like this and, before investing effort into any long term study, it really would be worth while calculating just how much actual energy may be available in the first place.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook