Here’s a YouTube video that I’m sure will be called a fake. For one thing, the notes below say that the effect is visible only with the camera, and not with the naked eye. I’m not sure why that is, but I still think this is real. Maybe it’s just blind faith. . . .
The YouTube notes say this (and also give instructions for making the device):
Ever since I created the first version of this video a year ago I’ve been wanting to try it again with more water and better lighting / footage. This is a really fun project and when you first see the results, chances are your jaw will drop. The main thing to keep in mind for this project is that you need a camera that shoots 24 fps.
The effect that you are seeing can’t be seen with the naked eye. The effect only works through the camera. However, there is a version of the project you can do where the effect would be visible with the naked eye. For that project, you’d have to use a strobe light.
I’m sure some of the physicists who read this can say a. whether it’s real and b. if so, how it works.
sub
woofer
sub
I will take a stab at a possible reason.
You can see the hose end moving as it puts out the water. It moves in a spiral wave, so the water comes out in a spiral pattern.
The speaker sound is at 24 Hz and the camera is at 24 fps, so therefore it is equivalent to a strobe.
It looks real to me. It is just that one cannot see it with the naked eye because it needs a strobe to detail it.
I would think this rather inexpensive(?) trick would be in use all over trendy clubs in Vegas for example (add colored lights, dubstep, etc -wow..am I old – that was the first time I actually typed out “dubstep”). I do think it’s the hose movement, not the sound vibration acting directly upon the water, although I can’t explain the reverse effect.
I think the forward or reverse is due to 24 fps vs 23 or 25 hz
Yes, phase shifting.
The apparent reversal is only a matter of timing and it depends on the phenomenon being periodic; you can do the same with a strobe and regularly timed water droplets to make droplets appear to fall up rather than down. You get the same sort of effect with fan blades in a room lit with fluorescent lights – the blades may appear to spin the wrong direction.
Yep, Roy Rogers/Tom Mix/John Wayne/Toyata ad wagon wheel effect. As old as them thar phase shifted hills.
I vote with this explanation also. The reverse/forward effects is due to the strobe or FPS of the camera to freeze the action so that it appears forward and reverse. Anyone that has played with an oscilloscope knows how this works.
I’ve seen similar things with strobes. Not with water, but there is no reason the same effect could not be produced with water. So it may be fake, but it is at least plausible.
That or the video is in slow motion. At normal speed the water would look like it’s coming out of the hose chaotically, only in by slowing it down to we see a pattern.
You can tell it is not in slow motion when he moves the camera. Unless he has the camera on a rocket sled or something.
I saw the hose oscillating as well, which clearly seemed to be physically guiding the water. I can recall playing with a garden hose in this way as a kid, waving it back and forth to make loopy patterns with the stream of water.
The only baffling thing here is the claim that you can’t see this with the naked eye.
My first thought was that perhaps the sound waves in the air are distorting the light in some way that compensates for the vibrations in the water and syncs with the camera. But this doesn’t seem a promising direction.
The amplitude of the wave pattern in the water seems large enough that the eye would have to notice that much spatial displacement.
I would like to see this with my own eyes.
If it can in fact not be seen, it could be that the 1/10 of second refresh rate of the eye, and the brain’s ability to smooth out surprising inconsistencies in our visual interpretation of our environment, as we frequently see demonstrated by optical illusions, is somehow collaborating to cancel out the 24 Hz pattern in the water. There may be some interaction between the 24Hz pattern in the water and the acceleration of gravity that conspires in such a way that the naked eye only detects water at or very near the zero (central axis) amplitude, but the max and min amplitudes somehow “fall” out of the eyes view before it can form the image.
Or perhaps what happens is the naked eye sees a straight stream of water that is much thicker than the actual volume of water flowing through the tube. With the 24Hz wave pattern and the acceleration of gravity, the eye would see at least a full cycle of the falling water wave during each 1/10 of a second interval, so it sees water occupying every location on the full horizontal span of the wave amplitude from peak to trough.
I had the same reaction at first — because it is easy to make a spiral stream of water just by hand. I actually didn’t realize what was “special” about what I was seeing until the “reverse” motion at the end.
The naked eye is going to see the coils of the spiral fall quite fast — with the initial velocity of the stream, further accelerated by gravity. But what this video shows is
(1) the coils are stationary,
(2) the coils flow down very slowly,
(3) finally the coils seem to move upward.
Only case (2) above is physically possible — but would require the initial velocity of the stream to be very slow, and trick to be performed in weaker gravity, perhaps on the moon.
All 3 of the above cases are achieved with the stobe effect — I explain in detail in a standalone post.
While I’m not sure about the validity of the video, camera sensors can produce interesting effects:
Well, makes sense.. having 24 hz and a camera of 24 Frames per second will capture each amplitude of the propagating wave, and as the sound propagates in water, it gives this wonderful image!!
This technique of using sound waves to levitate tiny droplets is used in pharmacy widely! 🙂
I think #2 above has explained this well. No wonder you cannot see it with your eyes. What you would see is the water draining out of the spiral hose, sort of sweeping back and fourth or sweeping in a gentle spiral. In some clips you can see the hose slightly bobbing back and forth, which would set the pace of the lashing movement of the water stream. The film introduces a kind of strobe effect.
Reminds me of other camera tricks that can make a spinning fan blade appear to stand still, or even rotate backward — on film.
Since we do not see real film being used too often these days, these old tricks must seem sort of magical now.
It happens a lot with helicopters in movies.
Not a physicist, but I would say almost definitely fake. That phenomena is large and stable, and should easily be seen with the naked eye. The motion of his hand is normal when he’s changing the volume, so there doesn’t appear to be a need to use slow-motion to observe the effect. Also, 24 Hz is just within lower limit of human hearing, so why can we clearly hear the water but never the sound, even though the volume should be “up”? Inputting variable force (via sound waves) into a closed tube should create longitudinal (compression) waves and should lead to variable efflux rate or velocity at the end of the hose. This video has transverse waves in the water. There are a bunch of other oddities, but I think its pretty clear that this is FAKE.
His hand appears to moving normally because the camera video is being shown at the exact same frame rate as it was recorded. There is no slow-motion or speeded-up effect involved. Also, there is nothing being done to the density of the water – the effect is caused by the motion of the hose, not any compression of the water. Remember, sound travels faster in water than it does in air because water is not compressible.
If you have access to a strobe light you can see this effect with your very own bare-naked eyes.
Not AS compressible. If water was completely incompressible, sound would move at an infinite velocity.
Lot’s of luck getting compressed audio on YouTube to go down as far as 25Hz.
Not so sure. You should have to explain how you would fake this. You think purely with digital editing? That might be harder than it sounds.
When the waves move forward or in reverse, it bears a strong relationship to the effect of seeing a gradual reverse rotation in a rapidly rotating car wheel or plane propeller. It has to do with the phase of the oscillating image and the refresh rate of chemicals in the rods and cones of the retina.
Did you consider that it would actually be more effort to fake this than to create the effect for real? I don’t think you’d hear a 24Hz sound wave once it had been through his camcorder’s mic, whatever audio compression YouTube uses, and your computer’s speaker. Though you might feel it if you were there. Are you familiar with sampling theory and aliasing? (Other than the alias you use here, I mean.)
Here’s another clue, by the way: in addition to the 24 fps frame rate, he’s using a very short exposure time, probably 1/500 or 1/1000s. That’s an important aspect of why you couldn’t possibly see this effect with the naked eye, and why he’s recording it in broad daylight.
This absolutely is real and it has everything to do with the strobe effect.
The motion induced in the hose by the frequency and amplitude of the signal being fed to the speaker is being “frozen” by the frame rate of the camera. I’m sure we’ve all seen the “stop-action” effect of a strobe light on moving objects. You may have seen a strobe light being used to make a spinning image or spokes on a wheel appear to slow down, stop or reverse direction. The exact same effect occurs with cameras when the frequency of a cyclic or recurring action coincides with the frame rate of the camera. There was a Youtube video making the rounds a few months ago of a helicopter that appeared to be flying with it’s main rotor almost stationary.
If the frequency of the strobe is a simple integer multiple of the frequency of the wave or repeating motion then we see multiple occurrences of the object/image.
Someone mentioned oscilloscopes above. There is all kinds of fun to be had by feeding sine wave signals that are integer multiples of each other into the horizontal and vertical inputs of an oscilloscope ( just google Lissajous figures).
I vote real.
I think the effect is produced, not by sound, but by the in-and-out motion of the speaker cone wagging the hose (notice the duct tape). Then the frame-capture of the camera, in sync with the wagging, produces the strobe effect.
You can hear a kind of buzzing sound-that’s the 24Hz ‘tone’, which, as pointed out above, is just at the lower limit of human hearing.
That is very cool.
I’ve done this with a strobe light and a guitar–you can time it so the light only goes off when the string is deflected off to one side, making it look like a string bulging out to the side in an impossible curve. When it’s slightly out of phase, it slowly waves back and forth.
Sometimes something similar can be seem in videos of turboprop aircraft, where the footage appears to show a plane taking off with its propellers motionless.
Another weird sight caused by the camera.
Or the bizarre helicopter illusion.
Physicist.
This is “not fake” in the sense that it is a real stream of water, real loud-speaker, real video camera, with no gimmicks.
This is “fake” in the sense that the water is not actually behaving (stationary, slow forward motion, reverse motion) as it appears to. Rather the apparent motion, or lack thereof, is due to the relative frequency of the spiral wave in the water and the frame rate of the camera.
This is the same phenomena that makes car (or wagon) wheels appear to rotate backwards in film or video. Consider a wheel with 24 spokes and spinning at 1 rotation per second, filmed at 24 frames per second. Each frame the wheel turns just enough that spoke #1 is now in exactly the same position as spoke #2 in the previous frame, likewise for each pair of adjacent spokes. The spokes all look the same, so on film they appear stationary. Speed up the wheel slightly, and spoke #1 is slightly ahead of where spoke #2 was in the previous frame, so you have the illusion of rotating forward, but slowly — much more slowly than the wheel itself is rotating. Slow the wheel down a bit, and spoke #1 is slightly behind the position of spoke #2 in the previous frame, so the wheel appears to spin backwards slowly.
With the stream of water the stationary (or slow-motion, forward or reverse) the illusion is enhanced by the isolated and seemingly stationary drops. But we know physically that each stationary drop of water we see must actually be replaced by another drop that fell from above to take its place in the next frame of video. The fact that this happens is, perhaps, the most surprising aspect of this phenomena. But with a steady stream of water at constant pressure and a steady clean sin wave being the only significant force other than gravity (a constant) acting on the hose and stream, we actually should expect a repeated pattern of water droplets — a pattern that repeats with the same frequency as the oscillation of the input force.
Of course, slight variations due to air currents and so forth, mean each of the successive drops that make a seemingly stationary drop will be slightly different in shape, size, position. But this just makes it look like the drop is wobbling in response to the 24hz vibration, which, if anything, enhances the illusion further.
Non-physicist here who had an actual physicist confirm this exact explanation. It’s a nifty illusion, but because the film-maker explains exactly how you can set it up for yourself, it’s not fair to call it a “hoax” or any synonym of “lie”. In fact, any person with a hose, a large speaker, and knowledge of the FPS in their smartphone’s camcorder can reproduce this at very low cost. It just takes the effort of finding a recording of a tone in Hz that matches the phone’s FPS. (I have a samsung galaxy, which Google suggests records at 30 FPS, so I’d need a slightly higher tone for this.)
Thank you very much for taking the time to write this explanation. It was very helpful.
Looks real to me, and reminiscent of ink-jet printing and flow cytometry/cell sorting – vibrate the nozzle at 25 kilohertz and make 25,000 droplets per second. Add a strobe at the same frequency and the drops appear to stand still; change the strobe and they will appear to move up or move down. And if the nozzle is allowed to move, spirals can be made.
I vote fake. The water looks like CGI. It seems to “float” above everything else. If you stare at the point the water comes out of the nozzle, it seems to hover a bit. Like they aren’t really connected.
Plus…
Well, there you go. That’s as clear an admission of fakery you can get. Case closed.
I take it you don’t believe in anything that you need a telescope to see, either. 🙂
This is an awesome post because it has elicited quite a variety of reactions. Many people have it right, but what interests me is in how certain people have gotten this wrong. And, I don’t mean wrong from a technical standpoint, but rather from the cynical dismissal inherent in the cry “FAKE” along with pithy and poorly thought out reasoning to support their dismissal. The hallmark of the closed mind.
When someone says that an illusion that they are presenting can only be seen on camera, specifically one that records in 24fps, how does that admit fakery? The illusion is BECAUSE of the camera. It’s a demonstration that the person who made the video says you can do yourself, a repeatable experiment if you will. Wouldn’t the final arbiter of “Case closed” be in the failure to reproduce these results? Or at least an attempt to understand the claim being made?
Others have explained this far better than I could, and probably understand it better, but I understand what 24Hz means, and I understand what 24fps means, and I know how the “strobe effect” works, so here’s a try. The sound wave being generated oscillates 24 times per second, which means the cone of the speaker will move 24 times per second, thereby moving the hose 24 times per second. If the camera records 24 pictures (frames) per second, then it will only capture one position of the hose in all those frames, and therefore only one position of the water in those 24 frames. Adjust the motion of the speaker cone up 1 time per second (25hz) and the images capture the position of the hose and water out of phase so that the water appears to move forward, albeit slowly (I’d venture to say about 24 times slower, but I’m not sure) If you drop the motion of the speaker cone back by one time per second (23Hz) and you are capture the position of the water and hose out of phase in the other direction, this gives the illusion that the water is moving backwards. If you draw a sine wave on a piece of paper (24 peaks, 24 valleys), then draw another slightly shorter wave over it (25 peaks, 25 valleys) at the same length as the first, where the waves intersect is where the images would be out of phase. You can then imagine a line of images beneath and where in the motion of the water and hose the 24fps camera would capture the 25Hz water and the effect becomes clear.
It’s a neat illusion, but then the person making the video never claimed it was anything else. To posit an alternative like CGI is a lazy analysis because making realistic water with CGI isn’t easy, and this would be some amazing CGI water. CGI proposes a more complicated solution to faking this effect than the effect being generated really warrants, especially when the claimant says what you need to do to do this yourself.
Just as another example that I haven’t seen brought up, if you’ve ever tried recording your TV or CRT computer monitor with an older video camera (24fps was the standard VHS framerate) you’d see lines traveling up the computer monitor. That’s because your older CRT monitors refreshed at 60Hz, or 60 times per second. Meaning the entire screen redrew itself. Since the cameras were recording at about 24fps, you’d see a block of light moving along the screen.
Brusspup is an illusionist and his channel is full of illusions like this one. He usually shows videos of how he made these illusions along with other fun science tricks. So, calling this video a “FAKE” in such a flippant manner is calling a very good YouTuber, one whose videos are used to educate people about about illusions, a liar. Such a claim should be made with care and supported by a strong argument. Even a cursory glance at his channel should give anyone pause before dismissing his video as a fraud, and shouting “FAKE” with very little reason says far more about those making the claim than the person making the video.
I am liking this comment so hard. To evaluate something scientifically, we need to state a hypothesis, e.g., “If we build this device, will it make this result?” And the answer may be counterintuitive. Or if scientific questions always had intuitive answers, then why would we use the scientific method?
You also mentioned Brusspup’s YouTube channel for more good effects. Thank you, I didn’t know about it.
I, too, wish to compliment you on this comment. Polite, helpful to those of us who want to learn, and well put.
Google & Wikipedia, ever our friends…
http://www.youtube.com/watch?v=YpMtanpsVeE
http://en.wikipedia.org/wiki/Wagon-wheel_effect
http://en.wikipedia.org/wiki/Stroboscopic_effect
http://en.wikipedia.org/wiki/Temporal_aliasing
Oops… sry about the video embed… forgot about that. Please feel free to edit or delete the comment.
Eye has persistent of vision, the camera does not.
There CRT televisions and monitors rely on persistence of vision otherwise we would see the image on screen being painted as the beam traverses the screen.
The camera takes an image every 1/24 of a seconnd the same frequency of the oscilating stream.
To the naked eye it looks like it is spraying because the image of the falling water persists. but the camera picks breif and decrete moments. It is same thing when wagon wheel look like they are running backwards in an old western.
I am a physicist, a high school teacher, and I have done this myself. A student at my school also did it. It works wonderfully. We used a strobe light an no camera.
When you set the speaker to play at 24 cycles per second, the tube shakes around and the stream of water makes a spiral and breaks into drops as it falls. All of this happens too fast for your eyes to see, but the neat thing is that it happens exactly the same way, 24 times per second. So if if you take pictures exactly 24 times a second, or illuminate the water exactly 24 times a second, then you see the water in exactly the same shape every time. It’s actually not exactly the same, which is why the drops seem to have some jitter, especially the tiny spray drops and drops near the bottom.
With a strobe light you can adjust the both the speaker frequency and the strobe frequency. So I adjusted the speaker frequency to get the size drops I wanted, then adjusted the strobe to make them stop, move slowly down, or move slowly up. It takes a fair amount of messing around to get something nice.
Are there frequencies at which no such effect is seen, or can you make it work at any frequency?
What does the stream of water look like without the strobe?
It will work at any frequency (within reason) so long as it matches the camera shutter speed. The limits are set by the behaviour of the water stream – that is, it just needs to be a frequency at which you can get a decent spiral flow. Maybe anything between 10fps and 50fps? (I’m guessing there).
The other lower limit on fps is the eye’s persistence of vision when viewing the film in realtime – slower than about 16fps looks very jerky. And of course in this particular setup, the woofer probably doesn’t do much below 20Hz anyway…
Outside those frequencies you can still get similar ‘strobe effects’, just different in the details.
This isn’t fake and the claim is true: you cannot observe this with the naked eye unless you have a strobe light. The reason is that your persistence of vision spoils the effect when you observe continuously; if you observe intermittently as with a strobe or via the camera you will see what you see here. It’s easy for people to forget that motion pictures are not continuous – after all they were created precisely to fool people into seeing smooth and continuous motion and this is why we have a minimum 24fps these days and not the 15fps back in the earliest days of motion film (or 5-15fps with the nickelodeon).
sub
I’m really not sure why anyone has e problem with this. It is a straight forward visual illusion that has been seen numerous times before in different settings. This is just another example of that visual illusion called the stroboscopic effect.
Yes, this is real.
It’s an excellent example of “aliasing.”
The 25Hz vibration has the same phase increment per frame as a 1Hz vibration, so that’s why the water looks like it’s “falling” — in fact it’s falling much much faster than that, but it’s the pattern of droplets itself that is progressing down.
Similarly, the 23Hz vibration has the same phase increment per frame as a -1Hz vibration, so the pattern of droplets will move in the opposite direction even though the droplets themselves are still falling.
You’d get similar results if you vibrated the hose at 49Hz and 47Hz, or any multiple of 24 +/- 1. And the water would look “stationary” at any multiple of 24.
In real life, you’d just see diverging spray of water falling.
Cool experiment. Thanks to all for explaining the details.
It’s an absolutely genuine illusion produced by the camera (‘strobe effect’) (I find it quite bizarre that anyone should fail to realise that).
You would need a very stable flow out of the end of the hose in order that successive oscillations would match precisely in amplitude (which is why the water drops apppear to be stationary).
And there’s nothing magic about the 24fps, that just happens to be his camera speed. It would work equally well for any camera shutter speed within reason.
This effect can give filmmakers headaches if they’re shooting scenes with video monitors, computer screens, etc.. If the frame rate of the screen is different from the frame rate of the camera, then the camera sees a black or white roll bar going through the monitor. As in this case, the effect isn’t happening anywhere except in the interaction between the camera and what it’s shooting.
Which is why it is entirely normal to film such scenes with the monitor / screen/ whatever blank, and to add the imagery to it in post production. When to use which technique depends on the relative cost of film set time (including all the standing around while the technicians fiddle with the monitors etc) versus hiring moderately skilled post-processing technicians.
Same problem with fluorescent lighting, and you can get some really interesting effects with the current pulses travelling along the tube too. Really hard to cover up.
Running the speaker at 36Hz might be interesting as well — I think you’d see two separate shimmering streams.
Also, I’d like to see what happens with more complex oscillations, or with amplitude or frequency/phase modulation.
This video compares to Prof. Derin Sherman’s video.
In the comments, Prof. Sherman says, “An alternative to the pump [in Sherman’s video] is to vibrate the end of the hose using a speaker and oscillator — this is a much cheaper alternative.” So I bet it’s the speaker cabinet vibration making the droplets regular, not the acoustic waves in the air per se.
Sherman’s comments also credit Doc Edgerton with inventing the original strobe light version of this experiment.
Sorry about that link, I mean to credit Doc Edgerton.
With that method, you won’t have the reverse effect where it looks like the water is going back up.
Watch both videos:
• The video Jerry posted,
• The video I linked in my comment 25.
Both videos show the effect where it looks like the water is going up. The video I linked shows the reversal effect from 1:57 to 2:42.
The optic illusion is not fake. Yes: it is fans, wheel cars, including dancing under strobe lights (or any light that pulses like fluorescents)
Good explanations given all around in the comments. Just to add my 2 cents that this is a fantastic experiment to reproduce, talk about sine waves, measuring and how the eye perceives reality. Also Douglas Hofdstadter describes another trick, that works also the frames and the loopback. Here one example: http://youtu.be/Y-gqMTt3IUg
It’s just a form of moire.
Search youtube for these:
– Strobe light makes water drops fall in slow motion
– Strobe light water droplets – Steve Barbosa
– The use of strobe light to measure RPM – physics experiment
School physics: Unknown RPM of object; adjust strobe known frequency until stationary pattern appears = multiple of rotational frequency.
By the way, if you think water + sound is great, try fire + sound!
I tried doing that for my Physics ‘A’-level project! Well, strictly the idea was to try to measure the propagation speed of explosions in different gas-air mixtures.Well, OK, the real idea was to spend two weeks in the school labs over Easter, making lots and lots of explosions. Using a 5kV power supply.
I succeeded at the “real” aim. Didn’t demolish the labs either (OK, I broke a few of the ceiling tiles and singed several benches. And people.), though not for lack of trying. The teacher who was attempting to supervise us from underneath his car in the metalwork workshops at the other end of the school … developed a nervous tic. Poor guy.
In the unlikely event that any of the readers here meet me at work on one of my oil rigs, and then sees me start running for the horizon … it would probably be a really good idea to try really hard to beat me to the horizon ; you can always ask me why I was running later, but probably best to attend to the running first. (OTOH, if you realise that I’m running TO something … consider a 180 degree turn. Urgently.)
I couldn’t see the video last night – not a very good connection in the accommodation levels – but I suspect that this was the effect under discussion.
People have appropriately characterised it as a stroboscopic effect, including the classic “wagon wheel” effect of rotation changing direction as the “beat” frequency phase changes. (Did anyone mention beat frequencies? Phases? A nice demonstration, but no earth-shattering physics there.
I did look at it and think … if you set up a second speaker stack perpendicular to the first, and fed it from the same signal through a chunky air-cored transformer … you’d get a Lissajous figure when you viewed the water path along the z (gravity) axis. OK, you’d be limited to Lissajous figures where your two sinusoids were of the same frequency. But by changing the reluctance of the transformer (i.e. moving a lump of iron into and out of the windings), you’d change the phase relationships. And have hours of entertainment until the mixture of water and electricity went the way of all flesh.
Errr, “DO do this at home”, but preferably with components on an RCD circuit, or better using car stereo equipment and powering everything from a 12V battery. “Caveat emptor”. “Your mileage may vary” and “the cheque is in the post”. And for the last part of the traditional imprecation, I’ll resort to a link to “NSFW” offsite imagery : http://1.bp.blogspot.com/_7UfM3XH-UqI/TM_Xvz_REUI/AAAAAAAAAMg/XnB6J_TyRcg/s320/priest_with_boy_in_suggestive_stained_glass_image.jpg
I still am unclear on the explanation for the reverse effect.
You can think of it as you might a travelling wave — when you drop a stone into water, the ripple waves on the surface travel out, but the water molecules themselves do not. They’re just the elastic medium over which the wave travels. Same deal when you play a note on a piano — the hammer hits the string, and the wave travels down the string and reflects back up, and eventually creates a standing wave on the string.
In this video, the backwards effect is not water travelling up, but rather the pattern the droplets make changing over time. You can imagine it like this:
The speaker causes the hose to wave back and forth in a roughly sinusoidal fashion. When the speaker runs at 24Hz, it is by definition completing one cycle of oscillation in 1/24 of a second, so when the camera takes a picture of it the hose is always in the same place, but you see the effect of its movement in the pattern of the water droplets.
At 23Hz, the hose completes one cycle of oscillation in 1/23 seconds, which means that it has only completed 23/24 of one oscillation per video frame. The effect will be that the hose will appear to move at 1Hz, and you can see this in the video — and the pattern of droplets will look like it’s moving upward, because the pattern will have shifted up by 1/24 of a cycle each frame (or you can imagine it shifting down 23/24 of a cycle — these are equivalent).
Any frequency 12Hz<xHz<24Hz will create a backwards pattern like this — the lower in this range the faster the pattern will progress upward. It's unlikely that the woofer goes as low as 12Hz, though. And any frequency 24Hz<xHz<36Hz will have the "forward effect."
Incidentally this is also a great way to show that gravity causes acceleration — the wavelengths get longer near the ground because the droplets are falling at a faster velocity at that point.
Got it! nice. thanks.
Awesome. Takes a bit of mental gymnastics but I think I got it.
Short version:
In the stationary case, it looks to be frozen in air because the camera is taking a picture each cycle of the wave. If you think of a someone jumping on a pogo stick. Each time the person is at the top of the jump, you take a picture. If you play those pictures in sequence, it looks like the person is frozen in air.
Now imagine that rather than taking a picture at the peak, you take a picture of a slightly earlier part of the jump each time. So the first picture is at the peak. The next jump, you take a picture when the person is 90% of the way to the peak of the next jump. The third picture as at 80% of the way up the third jump, then 70%, then 60%, and so on.
If you play these pictures in order, it looks like the person is moving backwards because each picture in the sequence shows a slightly earlier part of the jump. The frames you are showing, in order, are 100%, 90%, 80%, 70%, …, in other words it is essentially showing the jump in reverse.
The same thing is happening here. In the reverse case, each frame of the camera is taking a picture of a slightly earlier part of the next wave. Since all the waves are basically the same, the effect is similar to playing a recording of a single wave backwards, so that is how our brain interprets it.
Imagine each wave from peak to peak is numbered 1, 2, 3, and so on. As you watch the stationary wave at 24Hz, focusing on one point in space, you would see these numbers incrementing by one each 1/24 of a second, or by 24 each second. The timing of the camera frame aligns exactly with the time it takes one wavelength from peak to peak to fall past the point you are focused on, so in each new video frame what you actually see is a new wave replacing the previous one in exactly the same spot.
Now if you use 23 frames per second, the waves don’t quite have time to fall the entire distance of a wavelength generated by the 24Hz speaker. You are shortchanging the process by just a smidgen. You would still see the numbered waves incrementing the same way, but each new fram would show the wave at a slightly higher location in space. This creates the illusion that the standing wave is rising.
Is it possible to recreate this experiment using 20fps of 30fps with a 20Hz or 30Hz tone?
If not, why does it have to be 24fps?
In principle similar effects can be achieved at other frequencies, and 24 isn’t a magic number.
But what probably makes 24 a good choice are the physical properties of the hose. The hose must be resonated by the speaker, which means there is some tuning between the dimensions and elasticity of the hose and the frequency of the speaker sound.
You could replace the hose and speaker with other vibrating or oscillating systems, and then tune the video frame rate accordingly, and get a variety of similarly interesting effects. Drum heads with sand on them, rotating wheels, airplane propellers, pendulums, and even CRT screens are possible candidates for observing this kind of apparent procession due to phase differences between the observed phenomena and the sampling rate of the observations.