h a l f b a k e r y
The halfway house for at-risk ideas
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
or get an account
A common misconception is that household microwave ovens operate at 2.45 GHz because water resonates at that frequency, leading to
the water getting heated. Actually, water isn't especially sensitive to that frequency, and can be heated just as well by other frequencies. I
don't know why that frequency
was used initially, but the main reasons, AFAIK, that that frequency is still used are that the magnetrons and
waveguides for it are already developed and available and that it's already an ISM band and therefore legal to use for this purpose.
Actually, the 2.45 GHz ISM band was created because ovens were already producing a lot of interference there, rendering it less useful for
But there are higher ISM bands, such as 5.8 GHz, 10.525 GHz, 24.125 GHz, 61.25 GHz, and so on. (Actually, I haven't been able to find
anything saying there's a 10.5 GHz ISM band, but the availability of unlicensed radio products (such as radar modules) using that band
seems to suggest it's one. Can anyone confirm/deny?) So, let's use those for food heating as well. The shorter wavelengths could have
several advantages. One would be smaller magnetrons. Another would be smaller waveguides, which could be routed to more outlets
around the cooking chamber for more even heating. Another might be that the door could be shielded using a (semi)conductive coating or
a fine mesh like a window screen rather than a coarse mesh, improving visibility.
A microwave oven could even be made with two magnetrons operating at different frequencies, which would enable some degree of
control over the distribution of heat between the surface and the deeper regions of the food.
Scroll down to the Penetration depth heading
[hippo, Oct 10 2019]
More on microwave penetration depths
[hippo, Oct 10 2019]
||//two magnetrons operating at different frequencies//
That bit I like.
Something I always thought was weird about microwaves:
when you have it on less than 100%, the magnetron simply
switches on-and-off to regulate the power. But (AFAIK) they
do it on the order of (quite a few) seconds; where-as I would
have thought switching at 10th's (or faster...) would be
better. Or can magnetrons not switch that fast (ie. takes
time to start up)?
||Shorter wavelengths wouldnt cook food all the way through. I think
you need the distance you want heat to penetrate into the food to
be about half a wavelength - the wavelength chosen for microwave
ovens is about the distance from the outside of a chicken to the
centre of a chicken. See links.
||5G has already started and the FR2 millimetre wave
bands have frequencies from 24GHz to 52GHz. There are
also lower frequency bands below 6GHz, likely to be used
||The higher frequencies are already being used for Fixed
Wireless Access. An example is Verizon in the US which
uses 28GHz. The Fixed Wireless Access idea is to have
something on the outside of your house which is your
connection to the Internet. This re-broadcasts to the
inside of your house. Cunning beam forming technology
allows each outside receiver to be pinpointed with a
beam. This means that rather than wastefully radiating in
all directions into space an in annoying Physics inverse
square law like way, the beam concentrates its energy, a
bit like a virtual wire, to the intended recipient. The end
result is more bandwidth. People in rural communities in
the US can, at last, have decent Internet connectivity.
||Why, we ask does it have to be on the outside of the
house? It's because the frequency band does not do so
well through solid objects. It turns out higher frequencies
in general also don't do so well through leaves (clutter in
the jargon of radio planning) or in fact the air in general.
There's something in the air, a pesky greenhouse gas,
water vapour. It limits the range of the signal and forces
the carriers to build more micro cells.
||All we have to do is to get a load of Fixed Wireless Access
external antennae, power them up, and place them in a
Faraday cage. They'll try to phone home and, with a bit of
luck, there's a chance they will act like a microwave
oven. Does it match the resonant frequency of water
vapour? Who cares, crank up the power until it does!
||Handsets, like the new Samsung S10, are 5G capable but
these currently use the lower frequency bands.
||The future may see higher frequencies in the hand and
therefore nearer the head. Luckily, my head is full of air
so there's no danger of blowing my mind.
||// Or can magnetrons not switch that fast (ie. takes time to start up)? //
||Magnetrons can be gated very quickly ; if they couldn't, pulsed radar wouldn't be possible. And they can operate in the megawatt range.
||No, the reason for the long pulse durations in domestic ovens is simply that the power is switched by a simple relay, because it's a cheap and well-proven technology, and relay contact life is directly calculable from voltage, current, and switching cycles. Since it makes no difference to the performance of the unit, high-PRF magnetron switching doesn't happen in home ovens.
||It's entirely and simply possible to modify such an oven, using a triac, to zero-crossing switch on half-cycles, so you can pulse the energy at 25 Hz or derivatives thereof with a little microcontroller. It doesn't make the thing cook any better, just makes it more complex and less reliable, but it does score a whole load of Geek Points.
||With an Arduino or a Raspberry Pi running it, you can control the energy over a 2% - 100% range, pretty smoothly, by selective cycke dropping.
||// But (AFAIK) they do it on the order of (quite a few) seconds; where-as I would
have thought switching at 10th's (or faster...) would be better. Or can magnetrons
not switch that fast (ie. takes time to start up)? //
||Panasonic Inverter microwave ovens, with their modern high-frequency magnetron
drivers instead of MOTs, were (maybe still are) advertised as having smooth power
adjustment rather than on/off over several seconds, which I took to mean they
used a somewhat more reasonably high (i.e. not absurdly low like everyone else's)
PWM frequency, but could also mean that they're doing analog regulation on the
magnetron power. I think they said it results in more even heating, or something.
||// Shorter wavelengths wouldnt cook food all the way through. //
||// I think you need the distance you want heat to penetrate into the food to be
about half a wavelength - the wavelength chosen for microwave ovens is about the
distance from the outside of a chicken to the centre of a chicken. //
||Interesting. So the higher-frequency magnetron, if it is to cook only the outside of
the food, might need to use a much higher frequency. Fortunately, there are 122.5
and 245 GHz ISM bands, though we might need an array of microscopic
||[8th of 7], [notexactly]; thanks for the info.