Radiometer

The way I learned it is that the air molecules near the black side gain energy and velocity and bounce against the vein pushing it. The molecules around the white side are cooler, less energetic and exert less pressure than the ones near the black side. It only works with less than normal pressure because normal air density is too thick to push the veins through.

It's sort of like the boats we made as a kid. Take a Popsicle stick and put a notch on one end and jam a small piece of soap into it. When you set it in water the soap particles spew off the soap into the water and you get propulsion.

You might try looking at one using polarized light. I've seen some images where the polarized light captured the heat convections in air. But, it seems like the effect has to be very near the black veins to work, because the thing is spinning so fast.

Thanks for your response. I looked at his initially and found myself scratching my head.

If the excited atoms bounce electrons off the vein or the molecules vibrate, I would think they would bounce off the black surface as well, canceling out the effect, unless the electrons developed enough energy to escape - i.e. the black surface emitted electrons, in which case it would develop a positive charge.

It does not develop a charge nor are electrons emitted. Here's the kicker. It does NOT work in an absolute vacuum. It needs partial atmosphere. If it were simply molecular vibration, it should work better in an absolute vacuum.

The only thing I can think of is that the heated black surface attempts to give off heat by convection to the gas in contact with the surface causing a rapid expansion of the gas. Maybe the gas even cools and contracts when in contact with the white surface, creating a high pressure area on the black side and low pressure on the white side. But I don't see that causing 3000 RPM's. And then as you mentioned, there is the resistive force of the gas that is in the bulb.

This thing is apparently so sensitive that simply touching the glass with your hands or holding your hands near the glass will emit enough IR to cause it to begin rotating.

Does it not work at all in a good vacuum or does it work backwards?

Tom

The claim is, that it does not work at all in a vacuum.

There is a device called a Nicols radiometer that is used to determine the force or pressure of light:

http://en.wikipedia.org/wiki/Nichols_radiometer

It does not have enough force to spin the device in question.

All of this leads to an interesting train of thought though. We know that increasing the energy in a molecule causes it to vibrate because the atoms are excited. Which would tend to mean that an electron orbiting a nucleus which generates a centripetal force on the nucleus causes an atom to be in a constant state of imbalance, causing the vibration.

If this is the case, changing the gas in the bulb to another gas with a gas molecule that's composed of atoms of different atomic weight, different number of valence electrons, etc., might yield different results. Causing it to spin slower or faster.

Also, based on the vibration of an atom at a given energy level, it might be possible to begin to be able to determine the level of "clustering" of electrons in a specific type of atom......

Guest writes: "Which would tend to mean that an electron orbiting a nucleus which generates a centripetal force on the nucleus causes an atom to be in a constant state of imbalance, causing the vibration."

-----

An electron 'orbiting' a nucleus does not experience centripetal force because the electron is, in fact, not accelerating! (Recall that velocity is a vector having both a magnitude and a direction. Acceleration can be viewed as a change in either quantity - or both). Were the electron accelerating, it would radiate its energy¹ in a very short time and fall into the nucleus. This doesn't happen, obviously, and it doesn't because the electron does not take the form of a particle spinning about the nucleus, but as a de Broglie standing wave. Prior to de Broglie's hypothesis (for which he won the Nobel Prize in 1929) physicists, including Einstein, did not understand why the electronic orbits in the Bohr atom were restricted to integral values of the angular momentum in units of h. But if the electron were in some sense a wave, it would be very natural to restrict the orbits to those of standing waves, for otherwise the electron wave on going around the orbit would interfere with itself destructively.

Suppose now the electron, having momentum p, is moving in a circular orbit of radius r. Then for a standing wave, a whole number of wavelengths must fit around the circle, so for some integer n, nl = 2p r. Putting this together with p = h/l , we find:

2p r = nl = nh/p

so

L = pr = nh/2p.

The "standing wave" condition immediately gives Bohr's quantization of angular momentum!

Since the wave has an integral number of wavelengths, the electron can occupy only discrete energy levels that correspond to integral wave numbers. The electron does not actually "spin" around the nucleus, causing the atom to wobble. Is this clear as mud?

Note 1: An electron carries an electric charge. A stationary electron creates no magnetic field (like a wire with no current). An electron moving at constant velocity generates a steady magnetic field, but (like a stationary magnet in a coil of wire) a constant magnetic field won't result in another electric field. An electron moving with a changing velocity (ie. accelerating by changing speed and/or direction), however, generates a changing magnetic field, which will produce a changing electric field, which produces a changing magnetic field, etc. In other words, it generates an electromagnetic wave.

Sorry for getting a bit off-topic here, but I thought a bit of clarification on this point might be useful.

Excellent response europium....

Clear as mud? Considering that mud is generally not clear or even remotely transparent,,,, sure. You lost me.

I fear there is a communication gap with me being on the mechanical side of the aisle.

I understand particles and the perception of wavelike properties of a particle. I do not understand how something can exist "as a wave".

After reading up a little on the "de Broglie standing wave", he actually won the Nobel prize for suggesting that light is a particle that exhibits wavelike properties and the same for electrons. This implies that an electron is a particle.

From the time I was taught "light was a wave" back in elementary school, I have never been able to grasp something traveling as a wave not a particle.

Could you please give me your definition of a wave so I can understand your perspective?

What causes a molecule to vibrate?

There have been a coupla papers written, maybe 20 or 30 years ago, on that very experiment. It is thought that there are two competing mechanisms: (1) heating of the gas from the incident radiation, leading to higher pressure along the black edge than along the silver edge; and (2) photon momentum exchange, leading to higher force on the silver face than on the black face. At low pressures (~.1 mm Hg) (1) predominates; at very low pressures, (2) predominates. Somewhere in between, there should be a dead zone. (1) is sometimes called thermal transpiration.

However, the real question is whether I can now have your radiometer so I can break it under water to test Mr. Truman Brain's vacuum question.

Sorry, after re-reading your post,,, I missed the "air" molecules and thought you were talking about the molecular structure of the black surface. It makes more sense now.

Thanks!

It's Brownian motion in action! I think maybe the explanation could have been phrased better. I see little reason why the impact of the warmer molecules should be any less on the mid-face of a black vane than at the edge, but of course it results in less torque about it's axis.

I only saw this mentioned once here; but the theory that I remember is that photon energy exchange from the light spectrum. I am unclear as to the electron comments and why it should apply.

The originator asks for something about "colored gas", but hasn't apparently been addressed.

ot a physicist, but would rather believe that a radiometer works best in a total vacuum. No gases.

And yes, it is NOT a "free energy" machine.

A radiometer will not function if it is completely evacuated, nor will it function if the gas pressure is too high (see second link in Guest's original post). The Wiki article referred to correctly states that the effect is seen if the radiometer is only partially evacuated to pressures between 10^-2 and 10^-6 Torr (1 Torr = 1 mm of Hg). At these pressures the mean free path of the air molecules in the radiometer is comparable to the dimensions of the device itself. Gas flow dynamics are generally classified into three regimes:

• viscous flow
• mean free path << size of the system (D)
• gas - gas collisions dominate
• molecules "drag" one another along in the flow
• when D(cm) P (Torr) > 0.5
• for air at room temperature
• intermediate (transition) flow
• mean free path comparable to size of system (D)
• complicated flow
• molecular flow
• mean free path >> size of system
• gas - wall collisions dominate
• molecules move independently of one another
• when D(cm) P (Torr) < 0.005
• for air at room temperature
• for air inside a typical consumer radiometer
  

Viscous flow no longer occurs at the low pressures inside a working radiometer, and so convection currents will not be observed and the presence of a 'colored' gas (bromine, for example) will not reveal any useful information to the naked eye. Instrumentation can be used to infer the motion of the gas by other than visual means, but such instrumentation does not not require that the gas be 'colored' in some way to aid observation.

If you have worked with or seen vacuum pumps designed for operation at extremely low pressures - for instance, those used to evacuate an electron microscope - you'll notice that these pumps generally have very large ports, or openings, into the pump body. This is necessary because at very low pressures, gas does not actually flow. Rather, the individual molecules in the gas wander about in random directions like drunken sailors and (hopefully) find their way into the pump, which then removes them from the system. Cryopumps, for example, work by chilling gas molecules when they come into contact with a cryogenically-cooled surface (often called an 'array') inside the pump. A cryopump's array is usually operated at temperatures around 15 K, and contact with the array literally freezes the gas molecules to the array surface (via cryoadsorption for H₂, Ne, and He, and by means of cryocondensation for other gas species).

But the main point here is that at the pressures inside a radiometer, the mean free path of molecules present in the gas is on the order of the size of the device itself. This means that the gas molecules inside the radiometer stand a fair chance of striking the walls of the device (and cooling thereby) before colliding with another gas molecule. There are no convection currents to be seen and a colored gas would be seen (if it could be seen; unlikely at these low pressures) as being evenly dispersed throughout the volume of the device.

Really interesting, thanks. I'm delighted, in the words of the old school adage 'physics is phun!'

I was at a continuing education seminar early this sumer where Steve Hansen who runs the website "Belljar.net" Gave a good description of the workings of a radiometer. He is also the brains behind MKS instruments. I have not looked but there may be a good explaination on his website.

Thermal convection of enclosed gas revolves the winglettes?

See my reply (post #12). The gas in a radiometer is too rarefied to form convection currents, as gas at that pressure doesn't actually flow.

-e

So, all there is left there is photon kinetics..

Is this why the winglettes are painted black and white?

Wiki says that Photophoresis is not essentially Radiation Pressure, although I personally cannot see how.

I thought that particle kinetics is particle kinetics is particle kinetics.

Well?

Hi Yuval,

No, there's gas in the envelope, but at a reduced pressure. If you completely evacuate a radiometer, it doesn't work. On the other hand, if the pressure is too high, it doesn't work at that pressure either. A radiometer typically works at pressures between 10^-2 and 10^-6 Torr (1 Torr = 1 mm of Hg). The fact that the radiometer doesn't work when completely evacuated forestalls consideration of photon kinetics alone. Now, if we were to eliminate enough friction at the bearing, it might actually work based on light pressure alone, but this won't explain why it does work when the thing isn't completely evacuated, and does so without our eliminating (as much as possible) bearing friction.

Also, the vanes rotate with the lighter side leading. Seems to me that if a radiometer depended on photon kinetics only, the photons reflected from the lighter side would impart momentum in the direction opposite of what is actually seen, ie, causing the black side to lead. Which raises the question: does a photon impart more kinetic energy when it is reflected or when it is absorbed by a surface? Assume perfect reflectors and absorbers in each case.

When I was a kid (when was I ever not a kid?!?), I shined my 1 mW HeNe laser on each side of the vanes. Nothing happened. A more powerful laser would tend to heat the vanes - especially the darker side - complicating the experiment. It would be interesting to perform a similar experiment using a completely evacuated radiometer and a more powerful laser (Vermin probably has one), provided we didn't cause the vanes to out-gas or something similar that would cause material to be ejected from the vanes.

Straight forward post. Me likes.

"...cause material to be ejected from the vanes..." - Do you mean like evaporated molecules from the vanes' surface?

"...does a photon impart more kinetic energy when it is reflected or when it is absorbed by a surface..." - This seems to be the most intriguing issue of the lot, so far...

Here we are, from the Wiki page:

"Radiation pressure is the pressure exerted upon any surface exposed to electromagnetic radiation. If absorbed, the pressure is the energy flux density divided by the speed of light. If the radiation is totally reflected, the radiation pressure is doubled."

So, according the Wike page, if the motion of the vanes were due entirely to radiation pressure, the vanes would rotate with the black side leading.

Right. So the coloured vanes are an actual part of the mechanism.

I suspected so, but had no ground on which to rationalise it.

Thanks! I first saw the sucker in the Haifa Science Museum in 1993 and freaked out, and all I had to go with, was closed-circuit convection, since I assumed that the tube would take up most of the radiation pressure...

The caption on the display object there was something like "Solar-Wind Indicator", or something like that...

"Solar-Wind Indicator", or something like that...

-----

Are you sure the caption wasn't Solar Flux Indicator? You'd have to be outside Earth's magnetosphere (no small feat) to measure the solar wind directly. Moreover, a Crookes Radiometer wouldn't be up to the job. The solar wind is just too tenuous.

I saw my first radiometer in the window of a pub near the TV repair shop where I used to beg for parts (we didn't have much money). I was riveted to the window, trying to figure out how the damn thing worked. The folks at the table on the other side of the window stared at me. I don't blame them, really. They probably thought I was just another ten-year-old pining for a drink.

Now, if we were to eliminate enough friction at the bearing, it might actually work based on light pressure alone

True, but the vanes would then rotate in the wrong direction -- with the black faces leading (more photons would recoil off of white faces than the black faces, and such recoils imparts more momentum to the vanes). Hooke radiometers normally rotate with the white face leading (I am looking at one as I type these words). This can only be explained by the involvement of gas molecules.

Oh, okay, Europium already pointed this out: photon pressure alone would cause the vanes to rotate in the wrong direction. Many people think that Hooke radiometers are powered directly by the collisions of photons with the vanes. This is clearly incorrect.