Measure The Speed Of Light With Your Microwave

maddmike writes "There is a very interesting article on About.com that shows how to measure the speed of light using your microwave to melt chocolate. "

Comments

Half the experiment is missing

[panurge]

In a proper physics experiment all the quantities that affect the result have to be measured. In this one the frequency of the microwaves is taken for granted, so it is not a proper experiment. Just reading the value off the label doesn;t count. All the experiment actually allows you to work out is the wavelength of the microwaves, which is not hugely interesting. So how do you measure the frequency?

You cannot do it by measuring the dimensions of the magnetron cavities, because the calculation of the frequency based on dimensions assumes the thing you are trying to work out - the speed of light. Frequency counters that go up to 2.5GHz are a bit difficult to come by in most homes. One possibility might be to extract some energy from the cavity using a suitable antenna and mix it with the clock signal from a 2.4 or 2.53GHz motherboard, then try and pick up the resulting beat signal using a short wave or VHF radio. However, I'm not at all sure how to get the signal out of the P4.

Has anybody got a better and reasonably practical method of measuring the frequency?

Re:Half the experiment is missing

[zcat_NZ]

He's quite correct. If you want to measure the speed of light properly you really should establish for yourself the frequency of radiation you're using, or whatever other varables arise in the calculation. If you're prepared to accept 2.4GHz off the back of the oven, how is that any better than accepting
299,792,458 m/s directly from whatever source of reference you prefer?

Re:Half the experiment is missing

[aziraphale]

Well, then you'll have to define your own 'metre' or your own 'second' to your satisfaction if you don't like that definition of the speed of light, since that is the speed of light (in a vacuum) - by definition. A metre is 1/299792458 of the distance light travels in a vacuum during the time it takes the radiation produced from a particular transition of a cesium 133 to cycle 9192631770 times.

If you're making up your own units, you might as well say 'one second is the amount of time it takes the radiation produced by my microwave oven to cycle 2.4E9 times'

But of course, this experiment isn't trying to find the speed of light in a vacuum, it's trying to find the speed of light in a microwave oven (or possibly in chocolate). I think that means using the defined value for c is acceptable, if you can find a decent way of using it to derive the correct frequency of your oven. But refusing to accept c would be.... problematic, I think.

Re:Half the experiment is missing

[panurge]

I really don't see why this is modded insightful. It's actually possible to reference the clock of a computer, or any other clock, to the cesium standard because it is possible to make use of radio broadcasts referred to clocks based on that standard. So in theory you could measure the frequency of the microwave oven - the frequency, notice - in terms of the international standard.

The measurement you do in the microwave oven is a real-world measurement, i.e. of a displacement. If you have a real frequency and a real displacement, you can measure the speed of light in terms of your measured displacement and a time interval related tot he international standard.

Your comment that "refusing to accept c would be problematic" is invalid because, in fact, the experiment doesn't assume a value for the meter, or for c. You could compare your measured interval to anything convenient - using compasses, just like a Greek philosopher - and measure the speed of light in terms of a reference in your house. It would be a valid outcome of the experiment to measure the speed of light in teaspoons per cesium transition. It isn't a question of making up units: it is a question of deciding on the real world analog of the reference unit, whether it be a standard meter, standard kilogram, or standard sack of sugar. The international definition of the meter is handy if you have a suitable interferometer, but you can still only use it to produce a substandard by making marks on something, or counting interference fringes.

Anyway, why am I bothering? As I remind myself, this is /.

Re:Half the experiment is missing

[hughk]

Cutting open a magnetron to check the cavity size works only if you have a spare! And in any case, the resonance of a Magnetron cavity depends directly on wavelngth and is only related to frequency by the speed of light (which you are trying to measure).

The frequency will almost certainly be 2.45 GHz - a couple of other frequencies are allocated but aren't used much. The frequency isn't constant because the magnetron's operating frequency typically varies with temperature however for getting a 5% accuracy, it should be ok.

One complication of measuring by beat frequency is that a magnetron is pulsed. The duty cycle allowing power control. The problem is that this may mess up any indirect measurements.

Re:Half the experiment is missing

[Anonymous Coward]

the frequency of the microwaves is taken
for granted, so it is not a proper experiment. Just reading the value off
the label doesn;t count.

mix it with the clock signal from a 2.4 or 2.53GHz
motherboard, then try and pick up the resulting beat signal

I see. So reading the 2.4GHz label on the microwave doesn't count.
But reading the 2.4GHz label on your P4 is just fine.

Why bother measuring it?

[The Famous Druid]

The metre is defined in terms of the speed of light, so by definition c=299792458 m/s

Pretty pointless trying to measure it really.

Half the experiment is not missing

[ConfusedVorlon]

"In a proper physics experiment all the quantities that affect the result have to be measured. In this one the frequency of the microwaves is taken for granted"

This seems wrong to me. Experiments seek to measure the unknown using the known.

Why is it less valid to measure the frequency by looking at the back (another person has measured the frequency and marked it on the device) than it is valid to measure the distance by comparing to a ruler where another person has has measured a set of lengths and marked them on your stick of wood.

More generally - do you expect scientists to measure the speed of light and the charge of an electron for every experiment they perform? If c and e cannot be taken as known - how about Pi?

If science is about accumulating knowledge - it seems odd to throw it all away for each experiment...

This won't work for most microwaves

[ChrisCampbell47]

I'm surprised that nobody's posted this yet (or that that comment hasn't made it up above +2).

This only works if you can stop the mechanism by which the microwaves are scattered around to make for even heating. If you have a turntable in the bottom of your microwave, then removed it might do the trick, but most microwave ovens have a rotating metal "fan" that is enclosed in the upper surface over the cooking cavity, and that metal fan spins to scatter the waves around -- think of it like a flashlight and a mirrored pinwheel. Hence no turntable is required.

I'm not aware of any way of disabling that "fan", although I suppose you could drill a tiny hole in the shroud and poke in something to stop the spin, a la stopping a grinding PC fan. But I personally am not terribly interested in poking a drill into a microwave oven ...