Help a kid science the speed of sound
January 12, 2017 8:27 AM   Subscribe

Asking for my kid: My teenager got a science assignment to determine the speed of sound. She is planning to measure the time difference between a visual event and the accompanying sound, observed from a set distance. Can you help her find practical ways to do the experiment?

First, she'll need to determine the exact distance between points A and B, which obviously also needs to be long enough. Is there a smart way to do that? Counting your steps seems so inaccurate (and IME, a smartphone GPS isn't very reliable).

And then she'll basically need something that causes a flash and a bang (or something similar) at point A, loud and visible enough to be observable at point B. I suggested some kind of firecracker, but I'm hoping there's a better idea out there (because as her main research assistant I'd probably be lighting it, and I hate those things).

She'd like to repeat the measurement at least 10 times for reliability.

We have no fancy equipment, other than a video camera and a stopwatch app, which she will rig at point B.

How would you do this?
posted by sively to Science & Nature (22 answers total) 3 users marked this as a favorite
Can't you measure a distance between two predetermined points on a map? Google Maps has a Measure Distance function if you right click.
posted by EndsOfInvention at 8:40 AM on January 12, 2017

Or, our school always had a trundle wheel. Is there one available to borrow?
posted by EndsOfInvention at 8:42 AM on January 12, 2017

To illustrate the point when I was in elementary school, we popped balloons. For distance, you can get a few different long-ish range measuring devices for not too much money. There might be a longer range laser measure, but most of the ones I've seen are limited to 100' or so anyway.
posted by LionIndex at 8:43 AM on January 12, 2017

Measuring distance like that is usually accomplished with a surveyor's wheel, the making of which could be a nice addition to the activity. If you have a handheld hiking GPS you can get lat/lon coordinates accurate enough to do this experiment at a high school level (and, again, converting those coordinates to distance measurements would be an interesting side activity).

The accuracy of your observations is going to be the hard part here, I think. If you're going to measure this by manually pressing on a smartphone to start and stop a timer, those errors are going to be high compared to the actual speed of sound. However, the video camera could provide a pretty good measurement device - you know the capture rate of the camera (24 Hz or whatever), and using a computer you can scrub the video back and forth until you find the exact point that your event occurs. Then just count the video frames between that event and when the microphone picks up the sound.

For an event, I would suggest hitting baseballs off a tee with an aluminum bat. Someone can hack at them from home plate at your local little league field, and the camera can be set up at the back fence. The aluminum bat will provide a nice sharp cracking sound, and with a decent zoom on the camera (and carefully choosing your viewing angle) you should be able to get a good picture of the initial contact.
posted by backseatpilot at 8:43 AM on January 12, 2017 [2 favorites]

Low accuracy, no cost, easy: I would find a running track, go 100m away, and get two sticks which I would smack together with big, easy-to-see movements. You should be able to get a few frames in between the visual and the auditory hit.

One step up: same thing, but figure out a way to set yourself up 100m away from a brick wall. Take the video at a close distance and listen for the echo. You get twice the distance and no error from the speed of light (if we are pretending we don't know any of these values to start with.) You may need something louder--canned air horn? But you're closer to the camera so the visual is easier.

One step up: take a brief video of two watches, clocks, etc. to confirm the time is synched. Find a long straight distance (ideally a multiple of 100m for ease of calculation; for the accuracy you're dealing with here GPS is fine, or measuring the points on a map. You take video from a distance filming only your clock, she, from the starting point, films her clock and the noise maker (I'd prefer an air horn over a firecracker; check local laws). Take another video of both clocks at the end to verify that they're still synched. The start time would be the time on her clock at the moment she started making a sound, the end time would be the time on your clock at the moment you recorded the sound.
posted by tchemgrrl at 8:45 AM on January 12, 2017 [1 favorite]

When Mr. Wizard did it, he used a walkie-talkie (which moved the sound at the speed of light). Failing that, you'll need to be close enough together to see the action, which probably means you'll want to film it and count frames of video rather than use a stopwatch (unless you can find 400+m of flat land).

As far as measuring distance, you could always use a bicycle with a piece of colored tape or somesuch, and ride / walk slowly so you can count revolutions.

For the sound, wood blocks struck together should work easily enough (or something hard struck with a bat).
posted by uncleozzy at 8:46 AM on January 12, 2017 [5 favorites]

Distance is fairly easy accomplished with a either a long tape or a length of string/rope of a predetermined distance. You can use say a 10m length of rope to repeatedly measure out your distance in 10m increments. For best accuracy you want something that doesn't stretch and you can buy a 25' steel tape for less than $10. Amazon has this lufkin as an addon item for $5 and this 100' Komelon for $10.
posted by Mitheral at 8:46 AM on January 12, 2017

A flash might be too much, but I remember independently discovering that sound was slower than light as a young kid watching someone pound a sledge hammer from half a block away. It was very obvious when the sledge hammer hit, even without a flash. Possibly to make it more so, pound the sledge hammer on to something hard, but scatter some flour in the area first?

For measureing the distance, steps are awkward. In the world of measureing a (running) race course, they'll calibrate a bike tire, and then ride the course in the most direct fashion multiple times and take the average of their measures.

for the purposes of this experiment, I'd take a bike, mark one part of the tire. Mark the ground with chalk, and walk it forward 5-10 revolutions (if you have a long enough tape measure, more is preferred). Then measure and divide by the revolutions to find circumference of the bike tire. Count of the number of revolutions of the bike tire from the place that you're creating the sound to the place that you're measuring - that will be the most convenient way to measure. A distance of 0.5k should probably be fine.
posted by nobeagle at 8:47 AM on January 12, 2017 [1 favorite]

Oh and your student's school athletic department probably already has a 100' or longer reel tape they may be willing to lend out. It's used for measuring many track and field events.
posted by Mitheral at 8:50 AM on January 12, 2017 [1 favorite]

Without getting spoilerrific about the actual speed of sound, you'll want to do this in an area with some serious distance, otherwise you're going to be trying to capture milliseconds, which might be tough with a stopwatch app under human control.
posted by soundguy99 at 8:57 AM on January 12, 2017 [2 favorites]

Also, if you are going to use video/audio to test the time, I wouldn't depend upon the audio/video of the recording camera being perfectly in sync. So in addition to testing from as far away as you practically can, you'll want to get a calibration from as close as practical. Ideally the calibration will be several videos with a start/stop inbetween each attempt so you can see if any offset of the phone is consistent.

Most smartphones record at 30 frames per second (some newer phones do slow motion in up to 240fps). Assume you want at least 10 frames of waiting for the sound, meaning you need 1/3 of second, so the minimum distance that you can test from is 120 meters to get within 10% of the speed of sound with the accuracy of a 30fps camera.
posted by nobeagle at 9:09 AM on January 12, 2017

Excellent ideas, thanks!

We have no measuring devices we could borrow (no real athletic department or school track field either, this is Holland...). But the suggestions to either use a rope or a wheel are good. There's plenty of empty, flat beach nearby we could use for this.

We've been looking at the Google Maps' distance function, too, only then we'd need to find suitable exact locations for points A and B. (Our location is in my profile, if you want to take a look.)

Without getting spoilerrific about the actual speed of sound,

Haha, it's already there in the question tags.

Her idea so far was to have a video camera pointed at Point A, with a stopwatch visible on screen as well. So that when the camera captures the Visual Thing and a moment later the Auditory Thing, the exact moments would be there on tape. (The assignment would have allowed using a manual stopwatch and substracting 0.3 seconds for the average reaction time, but she disliked this idea for being inaccurate.)
posted by sively at 9:14 AM on January 12, 2017 [1 favorite]

Using a wall or similar and then timing the echo might be a different approach.
posted by Heloise9 at 9:14 AM on January 12, 2017 [1 favorite]

The assignment would have allowed using a manual stopwatch and substracting 0.3 seconds for the average reaction time, but she disliked this idea for being inaccurate.

She could always do both and compare the results!
posted by grog at 9:55 AM on January 12, 2017

One other thing I thought of for your distance: survey monuments are at exactly known locations. You'd have look up the locations (no idea how that works in Holland) but if you can find a couple sufficiently far apart within line of sight of each other it would give you a very accurate distance.

Disposable cameras are sometimes available for free from places that do developing. They have a fairly powerful flash (especially visible at dusk) and could be hacked to fire from the action of a clapperboard by running leads from the camera switch to a switch (or heck just a chunk of aluminum foil taped to each half) that closes when the clapperboard closes. A clapperboard could be improvised with something as simple as a couple lengths of 2x4 lumber slammed together or even a car door.

Be careful if you go this route; the capacitors that fire the flash have the ability to store a very painful charge if you short them with your skin.
posted by Mitheral at 10:08 AM on January 12, 2017

Is there a local football pitch or some other sports field with a known pre-set length? That'll make getting a fixed distance over a level surface really simple. If you were in the US, I'd point you to a football field or a baseball diamond, but I'm guessing round-ball football is your local game.

If not, I'd go with the string method described above by Mitheral.

For fun, see if she can measure the difference between a cold day and a warm day (if you can manage one of each). Might also be worth noting the barometric pressure around the time of measurement. Speed of sound varies with air density, so a "complete" measurement would be a speed measure that's qualified by the air temperature or air pressure.
posted by Sunburnt at 11:12 AM on January 12, 2017

You need something that is simple and physical that can be induced to make its sharp noise with almost infallible time and with clear visual indication. This noisemaker is a basketball, being bounced as regularly as possible.

You observe that noisemaker from progressively further and further away. As you recede from the noisemaker, the noise will lag the noisemaker more and more, until you achieve a distance of P/2*1100 feet, where P is the period of the noisemaker.

As you recede further, the noise will appear to be coming more and more closely into sync with the appearance, until you are at P*1100 feet, at which point the noisemaker will appear to be in sync with the sound that it makes.

Shoot some video here. Measure the distance to the dribbler with the tape measure. Elaborate as needed.

Another approach is to use 2 FM radios, one very loud. Tune the loud one to a station that always has a good beat, and do the same coincident thing that you did with the basketball, only this time coordinating the beats with an FM radio tuned to the same station carried by the observer. For each song, compute the beat frequency, and measure the distance to the loud radio.
posted by the Real Dan at 11:13 AM on January 12, 2017 [1 favorite]

Computer sound cards are very good for this: normal sound sampling rates give enough resolution to time experiments really precisely. You could even use the sound recorder on your phone: stand with your phone a known distance away from a wall and clap: if you put the audio file into something like Audacity, you should see the initial impulse and then a fainter reflected impulse. The time between the initial and reflected sample is the time sound takes to travel double the distance between the phone and the wall.

(I just tried this with a phone less than a metre from the wall, and got a rough-and-ready estimate of ~380 ms¯¹, though the trace has some other reverberations that make it a little hard to see which is the primary reflection. I've also done this before with stereo mics at opposite ends of a ruler with similar results.)
posted by scruss at 12:33 PM on January 12, 2017 [2 favorites]

As an undergrad I did the Mr. Wizard walkie talkie trick. We stood on opposite ends of a football field so we knew what the exact distance was. We both had walkie talkies, and I had a tape recorder. Loud noise was generated, picked up by the walkie talkie, and then later by the sound travelling through the air.

Digitized the tape (Windows 95!) and used something like Audacity to measure the time between the two sounds. After that it's simple math.
posted by spikeleemajortomdickandharryconnickjrmints at 1:34 PM on January 12, 2017 [1 favorite]

Get someone on a cellphone call at night, on a nearby hill, then get them to light a firework. Then video the whole shebang. Use windows movie maker or w/e to time the difference.
posted by Sebmojo at 6:08 PM on January 12, 2017

Have access to a tuning fork?

Here's an alternate approach using resonance

In physics, the sooner you can start thinking about things in terms of waves, the better!
posted by rocketbadger at 3:48 AM on January 13, 2017

Thanks again to everyone for an abundance of suggestions and ideas. Here's how it went down:

My daughter ended up using the Google distance function to measure the distance between some landmarks that were visible on the satellite map: park benches along a long, straight stretch of road in the dune area.

Two metal lids were banged together at point A and recorded on video at point B. (She then repeated the experiment using different distances, and also to calibrate on a short distance.)

Afterwards, she analyzed the time difference between contact and sound on I think Movie Maker or some other very basic free video editing program that also has audio graphs.

The result she got was 352m/s, which was off by a few percent. Her middle school science teacher was chuffed to bits. Apparently, all the other kids had results that were at least 20-30% off the mark, or worse. (I think my daughter was the only one who hadn't used a manual stopwatch, so it turns out everybody learned something about reaction time, too.)

I've marked a few best answers, mainly those she used or was inspired by. But you all get a gold star.
posted by sively at 2:16 PM on January 18, 2017 [2 favorites]

« Older 1999 Chevy Malibu - locating coolant leak   |   How can I repeat or sustain a sample using... Newer »
This thread is closed to new comments.