Barking Dog (slow motion) – Periodic Table of Videos

I’ve seen the barking dog’ probably a hundred times maybe more and I never noticed this before because it’s just too fast for the eye to see this reaction is one of a real favorites it’s been carried out for many years as a demonstration to our students and particularly by myself and some of my colleagues but also by BD Shaw Colonel Shaw who used to do the thunder and lightning lecture many many times and apply alike it’s called the barking dog it produces quite a kind of a backing dog sound a wolf sound it involves a test tube a really big test tube this is just a model with a cork and inside is a mixture of carbon disulfide cs2 and n2o nitrous oxide the flame shoots out the top of the test tube so I guess you have this image of this this flame coming up and consuming your hand if you leave it over the top of the test tube too long my first few goals always quite you can see me flinching and in some of the videos but I am as I’ve done it now probably tens if not hundreds of times I flinch less and less the combustion front races to the bottom of the tube giving up this amazing whoop and normally what the audience sees is a great rush the note just goes up like the squeak I made then it can be drawn out it could be roof or sometimes it goes much quicker but when you watch it in slow motion it looks quite different now you may not seen exactly what happened what I observed was it although the flame started slowly it built up the speed as it went off I’ve never seen it using this really really ultra fast camera you still see the flame going down instead of going down in one great whoosh it jumps up and down the tube so it goes down a little bit and then up and down and up and down and so on and as it gets near at the bottom it seems to go up and down more and more okay so we take the top off without a match and I thought that this this this flame front used to go very very smoothly from the top to the bottom and hence the whoop but actually what this video footage is shown us is that it actually goes in stages it’s sort of good sort of chugs its way to the bottom of the tube we weren’t quite sure what was happening I think what’s happening is the pressure wave is bouncing off the bottom of the tube now a number of very curious things are going on in this to imagine this hand here the one with the watch is the flame front coming down and as it comes down it gradually goes faster as it heats up and burns the fuel and eventually gets the stage where it almost explodes and sends a shock wave down to the bottom of the tube when it gets the bottle the tube it can’t go any further it’s not strong enough to break the glass the only time it ever went wrong for me was we broke the bottom of the glass so I guess that the pressure pushed that the tube downwards and we broke the top off the end and I only know of that happening once and it happened to me so it’s reflected and comes up again and then when it gets the flame it gives it a push upwards so the flame is now burning up here where there isn’t quite so much fuel and it gradually warms up again till it gets here where the fuel is rich and then another shockwave good boom up and actually if you look really carefully it looks like it’s picking up speed as it goes down the tube and I think that’s because the distance between the the flame and the bottom of the tube is getting smaller and smaller each time boom cup till in the end is going really quite fast well we still have another Joule prepared and so you have another opportunity of observing this effect I’ve seen two or three Vulcan dog experiments since then but the flame front going down is really too fast to see this bouncing and anyway I’m too busy blocking my ears to really pay attention to the flame front I think of all the slow motions that I’ve watched that one was the most surprising because you could have expected that cesium or rubidium would cause a big explosion but none of us really expected this motion of the flame front I suppose once you’ve seen it you can think of an explanation it’s no longer such a surprise but none of us were expecting it okay PhD is in some of these speakers every time you do a new experiment you see something subtly different and not just ask it makes you ask more questions red circle red said oh yeah ready on the trigger ready yeah

100 thoughts on “Barking Dog (slow motion) – Periodic Table of Videos

  • I suppose it makes sense now that the noise is so loud, because the vibrations must excite the air so it generates sound!

  • I have a different theory, I believe the gases in the bottom become so compressed under the energy and force that it thickens to the point where the fire can't ignite it all at once, this is also helped by the small diameter of the pipe, and the noise is actually created due to oxygen or other gases trying to escape the tube all at once. I noticed an almost vacuum effect towards the bottom, it'd be interested to see if this experiment still worked the same on a larger scale and would also like to see the barking dog in slow motion but from a bottom view looking up to see if the fire has a inward dip on it. 

  • Each time a pressure wave shoves down to the bottom, you can see drops of CS2 condense on the inner wall of the tube and evaporate again once the pressure drops.

  • It is a bit a matter of chemists trying to understand physical systems.

    If you didn't think this (or similar) pulsing took place, where would you expect the sound to come from? Something have to create oscillating pressure waves, else it would just have been a pop.

  • When I listen to my dogs bark (actual dog) audio recording and look at it on an oscilloscope, if you were to zoom in on the wavelength you can see the individual vibrato patterns.  I love this because this is like a visual version of just that!  The amount of vibrations in the sample and the shape in which is vibrates determines the way it sounds, for example…sine wave shapes determines the resonance (the acoustics of the sound) and the amount of vibrations/cycles/hertz in the given second of the sample determines the pitch.  This initial sound is probably about 140hz but I wouldn't be surprised if you would get different sounds out of different sizes and shapes of tubes. =)

  • I love that something done so often is able to provide such a great learning experience, not just to us non-scientists but even scientist, learning shouldn't stop and it doesn't stop, well done well done indeed!!

  • Why not do a secondary experiment? Use different sized test tubes (widths and lengths), an oscilloscope w/microphone and test for sound pressure level and frequency differences to see if the flame front vibrations correlate with sound frequencies produced. Might be a fun experiment.

  • it would be nice to test the theory
    you clould for example place a verye fast working presure measurement device (i dont know the right english word) at the bottom

  • if we have different lengths of tubes does the frequency of sound change? if so then if the frequency resonated with the glass you could break the glass.

  • could tonal rods be placed in the tube – maybe one in the middle not touching the sides going all the way up to the top / to produce different sounds? would the rod then vibrate a frequency that would break the tube?

  • I think it may be an effect of the mass of the gas flowing from the top of the tube. As the burning gas leaves the top of the tube its momentum is acting like a plunger and drawing the flame front upwards and lowering the pressure of the gas below it. Hense why you see the vapour form. This will only last a certain length of time until the pressure drop is enough to draw the flame front downwards. This would set up an oscillation, very similar to the one that is used in pulse jets. The frequency change could be related to a perhaps decreasing density and hence mass of the material in the tube, or that the reduction in unburnt fuel changes the pressure/volume relationship of the system in effect changing the 'spring constant'

  • If look at this video with english subtitles, you have a funny mistranslating at the 1:46 mark.
    instead of "races to the bottom" you have "racist obama"

  • Would be interesting to hook up a pressure sensor a the bottom of the tube and sample at the same or close to the same speed as the high speed camera as well as maybe hooking up a microphone and sampling that as well and see how all three pieces of data correlate as well as to verify the correctness of the professors hypothesis.

    Either way very cool, thanks 

  • My guess , just like plain fumes in a container are more highly flammable then a heavier gasses , when the actual gas in the tube starts to burn , it gives-off a lighter burn-off secondary gas , and when the flame gets pushed-up by the pressure , it resences the lighter burn-off fumes from the original burn off of the original gasses , inturn does a multiple reaction effect from each of these processes burns !…

  • Could you measure the 'beat frequency' of the reaction by correlating the sound of the reaction to the slow motion video?

  • I have knew about the reverb back when I was around 10 or 11, having done this with some butane or alcohol and a milk bottle or pop can, and though not as magnificent you get about the same result as the air fuel ratio gets right.

  • Hm, if you look closely at the bottom part of the test tube thing during the slo-mo you can see condensation forming during every circulation and then disappearing just as quickly.

    Another proof of it being a pressure wave from the flame front, I guess?

  • I used to do this all the time at my friends house as a 10 year old. He had a weight bench. The weight bar was 5 or 6 feet long and hollow. I'd spray a generous amount of aerosol deodorant on one end thrown light it on fire. Id then spray more deodorant thru the tube, but from the opposite end. The fire would spray out the other end violently. Then receed as a "fireball" back into the tube, all the way thru and shoot out the other end. It made the exact same sound generated by your barking dog demonstration.

  • It looks like the pressure waves are creating vapor below the flame if you look closely.  The vapor even travels downward to the same rhythm of the wave.

  • "If you want to understand how the universe works, one may think in terms of vibrations, frequency and energy." As you can see the flame vibrates. Think of that, the first phrase is from Nikola Tesla. Put your head to think. XD

  • If you feel it of value, perhaps several pressure gauges or indicators could be moulded at regular intervals down the length of the tube to show, over a few experiments, just what sort of pressures this reaction is making.

    I'm sure that even the final graph of time/mbar would even be fascinating.

  • Also notice how when the pressure wave reflects off the bottom of the tube and is coming back up, it pushes through the flame, again pushing it up, but look at the top of the tube: some of the flame gets pushed out of the tube each oscillation. Very cool indeed


  • I don't think the pressure wave is bouncing off the bottom of the test tube, but rather off of the gas beneath the reaction. You can see it being pressed down with each "stage".

  • I was thinking at first that's really cool… then I realized if there wasn't some kind of oscillating pressure there wouldn't be anything to create a sound.

  • Had one ever tried to measure the speed of the flame front and the shock wave, so we could check how close we get to the speed of sound?

  • Is this reaction linked to pulse jets in any way? What would happen if you put a valve/spark plug at the bottom of the large test tube, and kept the ignition going by adding 3N20+CS2?

  • Always wows me seeing the brilliant blue flames chugging up out of the tube. I love seeing the blue flame at the bottom pulsing as the fuel burns down the tube. 👍

  • I actually expected this from the earlier videos. For one, a rapidly, smoothly moving flame in a tube probably wouldn't make that sort of sound. It makes more sense for there to be oscillating shockwaves. For another, I could see a series of flashes as the flame traveled downwards, which suggests that the flame's motion isn't as smooth as it appears. Plus, at the end, you sort of see the top of the flame rapidly changing in length.

  • + Periodic Videos hi, could you ask one of your physicist friends what would the fromula for the acceleration of the flame front in such reactions after it reaches speed of sound be? that is, when the mixture is ignited from the open end and the other one is closed. I looked literally everywhere but i cannot find this info and im certain it was something impressive!
    It would be awesome if ou answered, i love your videos!

  • If I might ask, what chemicals in the reaction causes that bright blue flame? Just for curiosity's sake.

  • i kinda get the feeling that these oscillations are what causes the sound, you know what i mean? of course the sound gets higher as the flame goes down the tube, it's causing pressure waves at a greater frequency.

  • we kind of should have expected something like this, if the flame rushed smoothly down the tube it shouldn't make such a loud sound, if any. Sound comes from the vibration of air, so unless the vibration were coming from the tube itself for some reason, we should expect the flame to vibrate like that.

  • So, is this "oscillation" of pressure the ascending roar we hear? It could be actually fun to calculate it if we knew the frames per second of the footage, and then correlate it with the pitch of the sound we would hear at 100% frame speed… I guess.

  • i think if you could keep the reaction going and surround the tube with photovoltaic panels then you would a new way to generate electricity. 🤔

  • Strange you didn't mention the correlation with SOUND…I guess the increasing of the speed of the oscillations is the increasing in the peach of sound…Why not to COMPARE the flame oscillations with the sound wavelenght and frequency? =)

  • Is there any reason why one could not generate the barking dog "effect" (the descending flame and the whoosh sound) with any volatile flammable mixture in a long tube? Is there something unique about the nitrous oxide + carbon disulfide reaction that makes it work, or makes it work better?

  • I guess the sound it makes comes from pushing air out of the tube.
    It explains the sound going up in pitch.
    Couldnit the effect come from a shortage of oxygen.
    It belches the burned gasses and gets its next gasp of air into the tube.

  • Have you compared the change in frequency of the oscillating flame front with the audio waveform? I'm curious if the rising pitch of the barking sound is causally related.

  • I wonder how this experiment would work with a variable pressure release valve on the bottom of the tube (pretty bloomin well attached might I add)

    The barking sound could become completely different as I imagine it occurs due to those oscillations .

  • "It involves a test tube, a really big test tube.. this is just a model" No one commented on that great joke? Come on

  • I know it's an old video, but have you guys ever considered putting an electronic pressure sensor inside to record the pulses? You could have one stuck into a cork plugging a tube that has both sides open (like the extra long one).

Leave a Reply

Your email address will not be published. Required fields are marked *