1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Observation on a Radiometer

  1. Jan 11, 2014 #1

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    Off a g+ post:

    The device pictured (see link) is turning away from the dark side so thermal transpiration dominates.

    Data is a little hazy, ill-defined terms, and there is poor control of variables.
    But I'd put it down to the the laser spectra and the absorption spectra of the surfaces.

    Also reference:
    How does a light mill work by Philip Gibbs July 1996 (Usernet Physics FAQ)

    Note: Terrin is one of those happy individuals who loves to "play" with science equipment for entertainment but does not automatically assume some groundbreaking discovery whenever something happens against his expectations.
  2. jcsd
  3. Jan 12, 2014 #2


    User Avatar
    2017 Award

    Staff: Mentor

    The laser hits a single side, so radiation pressure and thermal expansion work in the same direction.

    The absorption spectrum of the surface is certainly a possible explanation, the absorption spectrum of the glass could be interesting, too.
    Alternatively: can we be sure the 10mW-laser is really emitting 10mW?

    That is another interesting observation.
  4. Jan 12, 2014 #3

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    ... nice observation: can suggest pointing the laser at the other side, see if radiation presure dominates - however:

    The green laser must be emitting an IR side-band.
    Looks like radiation pressure may be irrelevant.

    Even if that is what the manufacturer wrote on the tube, and it is still accurate, it may be the power consumption rather than the output... or it may be the output in the specified bands rather than the total output or anything.

    Good point - had not considered the glass. I bet lots of people have that blind-spot.
    I've heard that a lot of glass absorbs infra-red well ... on the face of it, that would seem a bad kind of glass to make a radiometer out of. OTOH: the glass would heat up and emit it's own infrared (etc) spectrum...

    ... that sounds like it could just work as an explanation: the radiation pressure makes little difference and the thermal pressure needs IR, which it gets from the glass not the laser ... the different sources heat the glass differently?

    ... what else am I missing?
  5. Jan 13, 2014 #4


    User Avatar
    2017 Award

    Staff: Mentor

    Direct heating is much more efficient and should work with visible light (as the surface is black), and apparently it absorbs at least some IR as well.
    Heating via the glass tube is possible, but the glass will emit in all directions, only a small fraction hits the radiometer.
  6. Nov 27, 2016 #5

    I have never encountered the following explanation for the light mill's rotation, which suggests to me that it must be flawed. Wouldn't the air on the heated side of each vane rise and cooler air fill in beneath it? Some of the cooler air will come from the cooler side, wrapping around the lower edges and displacing the vane.

  7. Nov 27, 2016 #6


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    probably not relevant since there is a pretty reasonable vacuum in the radiometer
  8. Nov 29, 2016 #7

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper

    You are describing a situation that convection in the rarified gas in the bulb has a horizontal component near the vanes.
    It is mentioned in the link in post #1.
  9. Dec 2, 2016 #8
    I always believed the long mean-free-path theory. I wonder how that got into my head:


    Since the effect is actually due to forces at the edges of the vanes, does it make a difference if the laser is shining on a spot in the middle of the vane?
  10. Dec 2, 2016 #9


    User Avatar
    2017 Award

    Staff: Mentor

    Where did you get that from? The force is everywhere at the vanes. Well, at all places where they are heated.
  11. Dec 2, 2016 #10
    That's how I interpret the explanation in How does a light mill work by Philip Gibbs.
    Is the force proportional to the area or to the length of the edges?
  12. Dec 2, 2016 #11


    User Avatar
    2017 Award

    Staff: Mentor

    It is more complicated, no simple proportionality.
  13. Dec 2, 2016 #12


    User Avatar
    Gold Member

    I've decided that I still don't understand how these things work.
    I just watched a video where someone made the vanes turn by cooling the globe.
    No light was necessary.

    By an odd coincidence, I just picked up a radiometer last Friday, and just finished performing some similar tests.

    The first thing I did, was set it up next to my stovetop heating element, about 15 cm away.
    When the element reached about 260°C, the radiometer was spinning faster than I've ever seen it spin in sunlight.
    My infrared thermometer indicted that the internal components were somewhere around 100°C when I finished.

    I then set up a contraption, such that I could warm the radiometer with just air.
    I placed my hands next to the tile base to funnel hot air onto the bulb.
    It turned slowly in the standard direction.

    Finally, I took a plastic bottle filled with ice and placed it on top of the radiometer.
    As in the video, the vanes turned backwards.

    The bottle has a concave base.

    Like a lot of people, I've heard various explanations of how these things work.
    But I'm now back to not understanding it at all.
  14. Dec 2, 2016 #13
    Thanks, Simon, for resurrecting the original post by Peter Terren. I had just decided yesterday to get a radiometer and begin experimenting with lasers when I came across your post. The OP had lots of good info about laser interactions. If I am not mistaken, he came to the conclusion that his problem was resolved when he tested the violet laser and realized that the measured lower output wattage was erroneous based upon the non linear absorption response of the cell at the shorter wavelengths.

    My other thought about the "cause" of the effect is that the molecular flow about the vanes reminds me of that which happens concerning "winglets" at the tips of aircraft wings. A vortex develops at the edges which is attenuated by the winlets. Maybe there is something similar going on here with the thermal flow.
  15. Dec 3, 2016 #14
    Internal components, or the glass envelope?
    I wonder from which part of the contraption the infrared thermometer reads a temperature.

    Question: How is the thing affixed within the bulb so as to minimize friction.
    Is it a wire with the vanes attached that rotates?
    Does the wire have a glass bearing top and bottom?
    Or is the wire fixed and the vanes rotate around the wire.

    I have never seen one AFAICR ( as far as I can recall ).
    Thinking of making my own some day ( like a lot of things never seems to happen ).
  16. Dec 3, 2016 #15
    Area force, Einstein Edge force, edge shear (opposing rotation), and the thermal transpiration ( that some seem to propose as doing it all ).

    It should be taken into consideration that a vacuum is required for motion with a heat or light source, but not a strong vacuum - in the viscinity of 1 Pa.
    Varying the vacuum will, I believe, result in one effect being more dominant than the others.
  17. Dec 3, 2016 #16
    How is the thing affixed ...
  18. Dec 3, 2016 #17


    User Avatar
    Gold Member

    I'm now guessing, I don't know for sure.
    But I'm leaning now to the opinion that I'm measuring the surface temperature of the globe.

    The problem yesterday was that I was using 3 4 different temperature probes:
    1. my fingers (conductive thermal sensor: this doesn't feel "hot")
    2. my eyes (that heating element is definitely hot)
    3. my infrared thermometer (something is hot)
    4. my skin (radiant thermal sensor: confirms that that heating element is really hot)​

    I think I previously assumed I was measuring the internal temperature, as the surface didn't feel that hot: 100°C.
    But it's possible, because the glass bulb is so incredibly thin, that even though it's very hot, it doesn't have enough thermal energy to burn me, and by the time I sense how hot the globe is, my fingers have absorbed all that energy.
    But I've just stuck the radiometer in the freezer, and the globe registers a consistent 16.2°C, while the freezer contents are at -11°C.

    I got mine at a "going out of business sale". :oldcry:

    I have no idea what those things are.

    So I wasn't the only person to analyze the data from the video. :oldwink:


    Science! :partytime:

    ps. I redid the experiment this morning, and my previous comments from yesterday appear to be somewhat wrong.
    Today's observations:
    The device started rotating when the heating element was around 260°F, and did not really start flying until the heating element was at full output.

    pps. My infrared thermometer has a maximum limit of 400°C, so I tried to collect some spectral data. I have no idea if it will be of any relevance to this problem, so I'll just keep the photos to myself.
  19. Dec 3, 2016 #18
    OMCheeto; the Video in your first post provided some excellent info...and the raw data chart was nice.
    Where did you get the data chart above?

    How long did you keep it in the freezer before measuring ? It has to reach equilibrium Temp.
    And of course, the temperature , even if accurately measured, must be measured with constant ambient light in order to have any meaningful significance.
    That is one thing I wish the fellow in your first video would have measured INSIDE the vacuum chamber (with some sort of guage) when he did the pressure measuements. -- TEMPERATURE.

    Furthermore; the thing that really makes things rough is we don't know the pressure (in the dark) of these little Crookes gizmos, so we haven't a clue where on the data chart they are fallen.

    My other problem is how the heck did the guy in the video determine the avg. molecular collision distance at optimum pressure was 2 cm if he didn't know internal temperature ??

    At first I thought : well, we could assume ideal gas law and use Pressure = nRT/V, and since V and n are constant we could assume Temp. (if he knew it) changes linearly with Pressure, (and figure kinetic speed from that); but not so since this is NOT a closed system; energy is pouring in from both the glass being exposed to ambient outside temperature and from the photon radiation itself .
    So how did he come up with a figure of avg collision distance? Anyone?
    Last edited: Dec 3, 2016
  20. Dec 4, 2016 #19


    User Avatar
    Gold Member

    It came from the graph in the video. I digitized it and changed everything to standard SI units.
    Not long at all. I was only trying to determine how it measures temperature.
    Does it see through glass? I think the answer in this case is no.
    I don't really know much about types of glass, nor how it blocks or allows transmission of infrared radiation.

    I'm curious about those things also.
    Perhaps, in the morning, I'll do more research.
  21. Dec 4, 2016 #20
    The transmission of IR depeds upon the type of glass; here's great little tutorial showing the transmission of various materials at various IR wavelengths.:

    I'll address some more later as time permits.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted

Similar Discussions: Observation on a Radiometer
  1. Crookes radiometer? (Replies: 1)

  2. Weird observation (Replies: 3)

  3. Stange observation (Replies: 18)