Science Buddies: "Ask an Expert"

Hello,
My son (6 grade student) wants to do a variation of the experiment named

"Come One, Come All! Explore the Effect of Light on the Speed of the Amazing Rotating Radiometer!"

He will like to compare which type of light makes the radiometer rotates the fastest. The different lights he wants to use are: LED, Fluorescent, incandescent and sunlight. He has all the materials except the different lights he needs to conduct the experiment. He understands the science behind the experiment and how the Crookes Radiometer works. He does not wants to test the effect of the increase in light intensity but to find out which of these lights makes it faster. Therefore he does not know how many watts of lights or lumens he should go for. He found a chart that compares the first three lights [link removed; no longer valid and he is still afraid to make a mistake on which one to purchase when deciding on the dependent and independent variables. His independent variables for now are: (1) to conduct the experiment in the same area with an amount of light where the radiometer is at rest, (2) the distance and angle between the light source and the radiometer, (3) a constant temperature (maybe 75 F, he has not decided yet, please give your input on this too), (4) he is not sure if the amount of lumens generated from the different lights should be the same (this is his biggest question). The dependent variables are the different types of lights, but the question is which one to buy and where. How many watts and/or lumens? Does the radiometer reacts the same to different lights radiating the same amount of lumens? or keeping the same amount of lumens in the different lights will give different results because of the heat radiated from the different lights?
His second option is to conduct the experiment comparing different lights in the visible light spectrum, like UV, infrared and sunlight. But still he doesn't know where to purchase these lights, the measurements will be in nanometer and he is not sure about all these. Please help him. Thank you

Probably the most important thing is to use light of the same intensity. Looking at the lumen output rating of the bulbs is a great start. If you find an LED that's equivalent to a 100W incandescent, an CFL that's also equivalent to a 100W incandescent and a 100W incandescent, you should have what you need for an experiment. Light meters might not help you since so many of them are blind to specific colors - so I'd suggest you rely on the manufacturer's ratings. Keep the radiometer in the same position and distance relative to the light source when you make the measurements.

You had asked about watts vs lumens - you can't really think about these alternative lights that way since a 100W LED fixture would be dangerously bright. Your son has probably read that the radiometer works by heat or infrared, so that should help tremendously when he makes his hypothesis.

Going the other route of trying different kinds of light sources, an IR heat lamp is pretty easy to find (http://www.amazon.com/Shatter-Resistant ... B002IL4Q2W). UV light is classified as UV-A, UV-B and UV-C. UV-A is the standard sort of party blacklight you can get at any party store. UV-B is the kind that causes sunburn and UV-C is the kind that's used to sterilize things. You can find UV-A bulbs at most stores. UV-B bulbs usually have to come from specialty distributors and MUST be used in special fixtures that keep you from getting exposed to the light. UV-C bulbs can be found if you look - be careful with those too. The problem with this version of the experiment is figuring out the brightness and power of the lamps. The heat lamp might be a 250 watt bulb where the UV bulb might be 25 or 40 watts. Lumens aren't applicable here either.

I'd suggest you stick with the original plan, find some different technology bulbs that are all 100W incandescent equivalent and see what happens.

Howard

Thank you very much for all the guidance. You have help him a lot!!!

Hello,
Thank you once again for your help in the past. We are encountering a new challenge in this experiment. We started the project using the differents lights (CFL, LED, incandescent) equivalent to 100W and testing under sunlight too. The radiometer worked outside and we were able to get data but inside didn't work at all. I was thinking ahead of the game and fortunately I bought three, since we were planning to break one of them to show in class how it does not work without partially vacuum. We used the second radiometer indoors and spun perfectly well. But now we have to redo the sunlight experiment since it was not accurate the first time when using the defective radiometer. I just mentioning this because if any other student is planning to do the experiment they should consider buying more than one since they are not very expensive and they are defective sometimes (specially because they are consider toys).
The problem we are encountering is the following. We purchased the Globee LCD digital tachometer and we set it for 4 blades. It did work under sunlight conditions but when we tested indoors, the reading was 1800 all the time and for all three different lights, but the radiometer was spinning at different rate under different lights. It did not made any sense. We checked the instructions and we were doing everything correctly. We also followed the instructions of Science Buddies. Therefore, we went to the website where we purchased the tachometer and there is a comment that reads as follows:

"Because of its sensitivity, the unit does not know whether it is looking at a propeller or an incandescent or fluorescent light, which typically strobes at 60 cycles per second or 3600 cycles per minute. When set for 2-blades it will read 360 or 3600, 3-blades: 240 or 2400 and 4-blades: 180 or 1800. This is a quick way to test the calibration. In direct sunlight this will read zero when turned ON."

What do they mean by that? We think that maybe they are trying to say that the tachometer does not work indoors. If this is the case, do you know if there is an option of a tachometer use by scientist in a lab? or there is no such a thing. I will like to read your input before knocking the doors of an engineering department of a university since I don't see any other option and is kind of late to change experiment. Also we purchase all the materials and my son is excited with all the background information he read.

I hadn't done the experiment myself so I wasn't familiar with the procedure and equipment. What you have discovered is exactly as you said - the tachometer doesn't work in fluorescent light.

The way that tachometer operates is that it expects the propeller (or vanes in your case) to interrupt a light source. It then counts how many times the light is interrupted per second and uses that to display RPM. Sunlight is continuous and just shines all the time. Incandescent lamps also just shine all the time. LEDs running from batteries will usually shine all the time as well, but one running on AC (like a new LED bulb might flash on and off 60 times per second). Fluorescent lamps also flash on and off, usually at 60 times per second. What the instructions mean is that a light source that flashes at 60 times per second will confuse the meter into thinking that a 4 blade propeller is turning at 1800 RPM. No matter what your radiometer actually does, since the light is actually turning on and off, the meter can't tell how fast the propeller (or vanes) are turning.

So it may be the case that if you only light the radiometer with a fluorescent, you may not be able to use the optical tachometer. If you can find a way to have the vanes break a beam of steady light while the fluorescent lamp shines in the vanes you might be okay.

Here's something you can try. The radiometer is only sensitive to light that comes in from the side and falls on the white and black sides of the vanes. It's not going to care nearly as much if the light was shining on the top or up from underneath - in that case it would be hitting the thin edge of the radiometer and not the white and black parts. You could taking a weak incandescent flashlight of some sort and put it underneath the radiometer. Put the tachometer up above and see if it can still detect the movement of the vanes. The fluorescent light shining in from the side will still be providing the energy for the motion.

If that fails, maybe you can try recording it in video and counting frames. Many cheap video recorders have the ability to record in slow motion. A normal camera records 30 frames per second and some of these cheap recorders can go up to 240 frames per second. You could record the vanes spinning and then play the video back slowly and count how many times they rotate in a given amount of time.

Approximately how fast would you estimate the vanes are turning from the fluorescent lamp? Is it slow enough that you can count with a stopwatch or using the video technique if need be?

Howard

Hello Howard,
Thank you for your promt reply. Both of your suggestions are great!. We don't know yet how we are going to implement them but it is worth to try. Placing a light underneath the radiometer is a challenge but in science you need to be creative and my son will figure it out. Thanks a million!!!!

Hi fpena,

I am working on updating the materials list for this project and had a question about your experiment. I actually bought one of the radiometers and tested it out with a variety of light bulbs. I found that, even though all three types of bulb (incandescent, CFL, and LED) would make the radiometer spin, they cast a very soft, blurry shadow of the vanes, which made it impossible to get a measurement with the tachometer (independent from the issue Howard pointed out about 60Hz flicker from certain bulbs). I had to use a bright LED maglite to get a sharp, crisp shadow that I could actually measure.

My guess is that maybe I did the experiment in a room with too much ambient light during the day - are you doing the experiment in a dark room/at night so you can get a sharp shadow with regular bulbs?

Thanks for posting, your discussion here is really helping us improve this project!

- Ben

Hi Ben,
You are welcome Ben!. We are glad to help as we are also learning alone the way.
You are right. We experienced the same problems. We tried the incandescent light in a dark room and still we get the wrong reading (1800 rpm). During the day the shadows of the vanes are very soft and blurry as you mentioned. They are more defined in a dark room (not perfect like under sunlight conditions but pretty good for the incandescent, we have not tried the other two lights in the dark room) but still the tachometer reads the AC effect and not the shadow of the vanes.
We have not tried the suggestions from Howard yet, but as soon as we are able to figure out how to do it and perform the test, we let you know all the details.
I see that for now the only way that the experiment can be performed is with the bright LED maglite. Therefore, maybe only the intensity of the light is what it can be measure as the experiment was designed for and not a variation using different type of lights. Please let me know if you come up with any other ideas to test the three different lights. Thanks for your input.

fpena

Hi fpena,

I like Howard's idea of eliminating the tachometer altogether (thus getting rid of all the problems with shadows and flickering lights) and using a slow-motion camera instead. Most newer smartphones can record slow motion video - my year-old Android phone can record at 120 frames per second, and it looks like the iPhone 6 can do 240 fps. This might put some limits on what you can test - the radiometer spins really fast in direct sunlight. I just tried taking a video with my phone at 120fps, and the vanes were spinning too fast to count rotations when I played back the video. However, it works great with a 150W incandescent bulb. Just make sure you get the math right when measuring rotation rates if you do this (e.g. if you record a 1-second video at 120fps, it will take 4 seconds to play back at 30 fps. That 4-second long slow motion video represents 1 second of real-world time. So if you count 40 rotations in that video, the rotation rate is 40 rotations/second, not 10 rotations/second).

Let us know if it works!

-Ben

Hi Ben,
Thank you very much!!! I started playing with my iPhone 5S and figure out what you explained in your post. We will try it and let you know how it goes. Thanks again you guys are life savers.

fpena

Hello Howard and Ben
We tried taking some data using the cell phone. First I followed the instructions of setting the phone at 120 fps but when you watch the video is not posible to count the rotations. I just want to ask you if you ment to count the rotations using the tachometer. I used the tachometer to read from the cell and it looks like it is reading data. My question to you is: is it reading the rotations of the vanes or the frames from the recording? I was very happy until my son asked me this. But I like that he is analyzing and comparying the scenario of the cell phone frames in a video with the effect of the AC. The experiment is making him think out of the box. Thank you guys!
I conducted the test in a dark room, with regular video setting (not in slow motion), and my three readings made sense. Let me know if I can post it here for your info in case you need to analyze it. I am not sure if I'm allowed since other students can see the information and you probably don't want that. I just need to know if we are going in the right direction or if we are still reading erroneous data. Thank you in advance,

fpena

Hi fpena,

I'm sorry if I wasn't clear before. I did not mean that you should try to use the tachometer to take a reading off the cell phone screen. I meant that you should literally watch the video and try to count the rotations manually. If that is too difficult to do on the phone, you can try uploading the video to a computer. Some video players will let you step through a video frame-by-frame using the arrow keys. That will help you slow down the video so it's easier to count the rotations. There is also a free program designed to help track motion in videos for physics projects, called Tracker:

http://physlets.org/tracker/

that will let you step through the video one frame at a time.

Let us know if that works. If the vanes are still spinning too fast to count, even if you play the video back one frame at a time, then you can try slowing the radiometer down by moving it farther away from the light or using lower-wattage bulbs.

You're right that we try to avoid posting results in the forums, so let's hold off on that for now - but I don't think that will be necessary if you can get the previous suggestion working.

Finally - it's great that your son thought about the connection between flickering light bulbs and the cell phone screen. For the record, I'm actually not sure what effect this would have. Cell phone screens, TVs, and computer monitors all do have a "refresh rate" measured in Hertz - for older ones it is usually 60 Hz (just like with light bulbs, since that's the frequency of the electricity coming from wall outlets), and it's higher for some newer ones. Your son could easily test this though - what happens if you just take a video of a white (or black) piece of paper, then aim the tachometer at the screen while playing the video? Does the tachometer show a reading or not?

Hi Ben
Thank you for the new suggestions! We are going to try counting the rotations using the program Tracker and let you know how it goes.
I did the test with the blank paper and the reading of the tachometer when I played back the cell video was zero. But we got data when we used the cell with the different lights and it made a lot of sense the readings of the tachometer when placed on the surface of cell phone. Two of the lights made the radiometer spin slow enough for us to count the rotations and compared the results against the readings of the tachometer. These were within the range of the readings the tachometer was giving us (I said range because looks like the tachometer is too sensible and fluctuates back and forth within 30 to 40 rpm between one reading and the next while the video is playing or maybe is reading the vanes and the effect of the shadows). One of the lights was impossible to physically count the rotations and in this case the tachometer readings were much higher than in the previous two lights readings (where the readings were very similar). We are going to download the videos in the computer and read it again using the Tracker. This is great because finally we will get an accurate data and also find out how close the tachometer can read from the cell or if those readings were not good at all.

Hi fpena,

Interesting, then it sounds like the tachometer might actually be capable of taking readings from the cell phone screen. I do not have that type of tachometer, so I can't test it myself. I would be wary of using that method for very high-speed rotations though, since there might be some blurriness or skipping in the video frames that would mess up the measurement. I'm interested to hear how things go in Tracker.

Thanks for your continued responses, when all is said and done I plan to update this project accordingly with clarifications about light bulbs and measurement techniques.

-Ben

What I have found in my experiments with my radiometer and different light sources is this: The radiometer is only affected by light that contains infrared wavelengths. I have used a bright (400 lumens) white LED and also an Ultraviolet LED with no effect, and several laser pointers–red, green and UV, none of which moved the radiometer. However, a warm hand or something small and hotter than room temperature WILL cause the vanes to rotate. Since the warm items show that the invisible infrared is what moves the vanes, I can only surmise that it is the same IR part of certain visible lights such as sunlight is what is spinning the vanes.