Sucking down on this cone makes it rise, and Feynman sprinkler

Trying to reproduce the Feynman sprinkler experiment, I tried a few more things, and then it got interesting. I love experimenting with fluid dynamics.
I figured out why the straight tube would also lift up:
• Levitating pipe myster...

[@eVITORIOe]

Maybe, the low pressure caused by the air bending around the upper edges of the straight pipe is pulling it up ?

[@matthiasrandomstuff2221]

Yes, that turns out to be it! There is enouch vacuum that the wall thickness of the pipe is enough to lift it by.

[@eVITORIOe]

@@matthiasrandomstuff2221 does a thicker pipe increase or decrease lift ? By the way, since I've got your attention, I would like to let you know that I love your videos and so does my grandpa. Even though he doesn't know English, he watches the video and we talk about it afterwards. You're awesome!

[@matthiasrandomstuff2221]

Making a pipe from thinner material lifts much less, while widening the top rim increases lift by a lot, so its gotta be coming from there. I made a short video on that, will publish it tomorrow.

[@123xqp]

​@@matthiasrandomstuff2221this is fascinating. Could/would you repeat this experiment after adding the O ring from the other experiment?

[@weldabar]

good job man!

[@ramosel]

In the late 70s as I was preparing to leave home for 4 years at Canoe U. and then on to a life in fighter jets. Some High School and Hockey buddies decided we should all go camping. Knowing the trouble this could be I drove myself. Just before Priest Grade outside Yosemite I stopped at a Gas Station/Mini-Mart. I noticed the old brown van ,that was the only other vehicle in the gravel lot, had a bunch of white squares all the way around the van under the window line. Each square had a Feynman diagram. A lady came out carrying bags of groceries and fumbling with her keys. I offered to take at least one of her bags and she handed me one and unlocked her door. As I handed her the bag I asked why she had Feynman diagrams all around her van? She responded, “well, I am Mrs. Feynman”. About that time, her husband came out the door carrying 2 more bags, which he too handed one to me. Then after adjusting his gear took the bag, turned around and said “Tanks”. (Not a typo). She then presented her husband with my question, to which he responded, “Hi, I’m Mrs. Feynman’s husband. Call me Richard”. I responded “Call me Richard, too”. We talked about 10 minutes, he asked Me lots of questions. I attended one of his lectures a few years later. He remembered me, told the Navy Brass I was one of the best helpers he’d ever had. I got an A in Physics that year.

[@matthiasrandomstuff2221]

Unfortunately, the comments are so full of random guesses, its hard to figure out which is right -- but I did figure it out in the end.

[@JasonJonesoriginal]

better than nothing but "this video sucks" comments

[@yoshimario3466]

Search "quasi 1D flow" for pipes. Basically, flow in straight pipes behaves like flow in nozzles due to friction.

[@FriggOff361]

Tell everyone how much you hate them matt.

[@leightonandrade8755]

Matthias, I’ve studied this effect for the past hour and I have come to the conclusion that this effect is caused by the air around the tubing being moved by the suction moving through the tube. I am contemplating making a video here on KZbin to explain, I’ll let you know if I do.

[@-O-_-0-]

Smoke could be helpful to see what is happening

[@markjacobsen605]

I recently read, "Surely You're Joking, Mr. Feynman!" A wonderful book I recommend to all. Great video, Matthias!

[@Drokkstar_]

Matthias, you're an absolute gem.

[@eric13hill]

I love the curiosity you have combined with your determination to measure and explore truth. Wonderful!

[@bigpushing7167]

This is so interesting. I'm a mathematics graduate but I've always been fascinated by physics. I especially love seeing physical experiments with surprising results. Thank you for the video! I really enjoyed it.

[@bigpushing7167]

I do not know what exactly is going on in your experiment. However, I do know of a video where a mathematician working in mathematical physics, Tadashi Tokieda, discussed something that's just as surprising to me when we talk about fluid dynamics and lift. I don't know if it will help you but here's a link if you're interested. kzbin.info/www/bejne/bHmZdoNjdtKAfK8si=_AvRwYJVaswK90Bk

[@WJCTechyman]

You can't do physics without the math. I know that fluid dynamics is a tricky one for sure.

[@69dblcab]

I always enjoy the rabbit holes!! Thank you Matthias.

[@MCsCreations]

Fascinating! And Feynman was a fantastic guy indeed... More than brilliant! Anyway, thanks, Matthias! 😊 Stay safe there with your family! 🖖😊 And happy holidays!

[@barkingsquirrel2082]

I really appreciate your curiosity and thinking. I actually built a dish for my wifi repeater to get signal to an outbuilding. It's been a few years now and was a great solution. Much thanks for your content!

[@JPGuay]

Just loving the direction this channel takes. Popularisation of science. Thanks !

[@michaelladue5655]

Thank you for making me think this morning and engaging my brain.

[@ronboe6325]

Try a clear tube and fine smoke source to "see" what is going on in the tube.

[@MotoRideswJohn]

I never imagined getting a lesson on fluid dynamics on this channel. Super fascinating!

[@deanperkins2091]

really cool video, your explanations are very intuitive.

[@MiscMitz]

Dude! I actually understand this now! Thank you so much. Great experiment.

[@shaunhayward]

Everything you do is fascinating!

[@lesterdelacruz5088]

Watched your videos as a teenager. Loved your approach to building bandsaws. Now I’m 30 and just stumbling upon your videos again. I’m glad to see you still making the awesome videos you make. Inspirational.

[@HeyImBubProductions]

I've also been watching since my teenage years and am 31. Matthias has always been an insta click for me. his videos are so consistently good

[@ScottFreeVideos]

Man, I love engineers!

[@craigcolavito5606]

Awesome, what cool effects. Fluids behave in mysterious ways.

[@danielmackey6594]

Fascinating. Fluid dynamics is (are?) amazing.

[@kevinetheridge6153]

This was an amazing video. I really enjoyed it, though I understand very little of the math / physics / fluid dynamics behind it. For some reason, the way you presented this one or the problem itself perhaps, I found incredibly enjoyable. Thanks!

[@sompka1]

Incredible. I just learned a few things about fluid dynamics.

[@MandoFettOG]

Now you've got me thinking about what the possible applications are for this.

[@joels7605]

That sir is the most interesting thing I've seen all week.

[@hootinouts]

Fun experiment. I love playing around with things like this in the shop. There is also a phenomenon called the Coanda effect with respect to flow and curved surfaces.

[@JoeGator23]

This explains how to repel black holes with a simple vacuum cleaner. Thanks.

[@HWPcville]

I'm neither a scientist, engineer or mathematician but I have a fondness for experiments of this type and greatly enjoyed the video.

[@ronwoodward716]

All very curious. Interesting stuff.

[@tedapke6519]

Really enjoyed this!

[@Watchyn_Yarwood]

So very interesting!

[@bugdozer314]

Consider if you pack the short tube with straws, would this tend to reduce or eliminate any possible recirculation effect? Rather than trying to visualize it with small threads, maybe try to remove it entirely and test? Great video as always. Thank you.

[@gorak9000]

Hmm, packed full of straws = laminar flow! I think that's the right answer - when the air is rushing around the edge of the pipe on the elbow directly, you get recirculation like he showed, so little effect of the air pushing on the back of the elbow, maybe due to forces canceling, or due to the flow at the elbow still being very turbulent, but when you put a longer straight pipe in front of it, that section of pipe is dealing with the recirculation forces, and by the time it reaches the elbow, you have more laminar flow, so it pushes the elbow backwards - but that separates the elbow from the pipe, so it oscillates between the 2 conditions, and it bounces back and forth! Pretty nifty. I wonder what an elbow with a longer straight section of pipe directly attached to it would do?

[@therealpanse]

I'd assume that would increase the friction between the walls and the air, producing more force in the opposite direction of the flow. Would need to test that though.

[@marcoschwanenberger3127]

Thats so cool! And it explains what I experienced just yesterday! I have 120mm dust collection tubing (ridged steel tubing) in my workshop and am currently upgrading my "main line" to 160mm. I have a little reducer piece to go from 160 down to 120mm and thats obviously a cone as well (About 100mm long). I was trying to - for fun - push the reducer piece the wrong way into the 160mm tube (So the 120mm side facing inside of the 160mm tube). And I just couldn't push it in! It kept coming out! I thought for sure it would get sucked in very hard and would be very loud with the air going through the now relatively small opening. Very fascinating indeed! thanks for sharing!

[@matthiasrandomstuff2221]

fascinating. I would think if you are sucking on the whole 160 mm diameter, that it would suck in hard enough to overcome whatever effect I was seeing with the cone moving away from the suction

[@D3moknight]

When the straight tube is closed to the end of the vacuum hose, the entire column of air is below ambient pressure because you are pulling a vacuum. This means that there is a force pulling the entire column forward towards the atmospheric pressure. Since the only articulating part of the machine is the end of the tube, it is pulled forward by the low pressure.

[@WJCTechyman]

Interesting as always.

[@Ekanselter]

You make physics fun! The TV series Mr Wizard was one of my favorites when I was a kid.

[@HexenzirkelZuluhed]

Truly random stuff. Yet so fascinating.

[@krenwregget7667]

always interesting stuff.

[@paulmeynell8866]

That’s brilliant so interesting.

[@blanchae]

This is quite interesting from a horn loud speaker. We use horns to better couple the acoustic energy to air and to make the speaker more efficient. Basically the reverse of what the bell opening does.

[@dannydodge7191]

I think the reason the vibe repels the vacuum is that the vacuum is sucking in air from the outside of the cone and the tube. Same thing with the cone rising, is sucking the air from around the cone and the shape of the cone is what causes it to rise.

[@htbyos]

I think that if you increase the wall thickness of the short piece of piping (and make a nice half-circle of the rim cross-section) it may become apparent: the suction on the top rim will pull it up - and as soon as it lifts and exposes the bottom rim, the same will happen there and it will be pulled down again. That would be my guess.

[@cest7343]

the Wandel effect 😂. Keep it up dear sir 🙏i really enjoy the wits and creativity.

[@pierrec1590]

Fascinating!

[@peterfordham3562]

The tube doesnt swing forward much because of the reaction force of the air going round the corner is cancelling out the -ve thrust. When air is flowing out the reaction force adds to the +ve thrust enhancing the effect. No idea about the straight side case at the end though.

[@justavian]

I thought the conclusion was going to be that the fast moving air started trying to drag air in from the outside, and somehow that was going to cause the motion. I guess that could be ruled out by making a diaphragm around the moving tube, such that all of the air would be forced to go through the tube, but the tube would still be able to move up and down. But with the tiny forces involved, the diaphragm would itself be adding a variable that would probably still be significant.

[@BLenz-114]

Yeah, maybe I'm thinking the same thing, that the air trying to get into the stream from outside the tube/cone was pushing it up.

[@manifold1476]

The motion induced in the intake tube by placing an object into the (converging) air flow reminds me of the motion induced in the magnet in the (diverging) magnetic field of the Stern-Gerlach experiment. (but I can't identify a similarity in the physics)

[@ashleyshearin3175]

I have not read all the comments so this may have been covered. I believe the lift is being generated by eddy currents as air gets sucked in from outside the tube. This air initially has less resistance and the tube is pushed away as it rushes in, then gravity overcomes the tube’s momentum and it falls so the process repeats. Adjusting the variables should act to control what can be viewed as a machine.

[@mdug7224]

The suction causes a vacuum around the entire hose and nozzle since air is compressible/decompressible. It is the same fluid within the system as outside the system - not air pressure on non- compressible liquid. This prevents the forward motion as the overall space around the piple loses pressure. The boucing is when the pipe is in an air pocket so free-falls back to position. From the original position the rebalanced pressure starts the process again. Layman speaking with limited experience of an open-faced helmet at 140mph.

[@thetheo2002]

Best. Video. Ever. The Feynman sprinkler. Yay.

[@MacroAggressor]

Fascinating.

[@woutmoerman711]

Thanks! I never thought of centrifugal force as a way of creating lift.

[@EeekiE]

These are my favourite kinds of videos. Hope it gains some traction and someone digs deeper.

[@andrewgalbreath2101]

Good ole Bernoulli. Explains so much

[@matthiasrandomstuff2221]

Actually, you haven't explained anything.

[@personacharles]

Matthias the momentum change from restriction with tube or cone into vacuum is the greatest. As the vac tube moves away from the straight or cone the vena cava of the air flow reduces the momentum change and pendulum effect of weight moves vac tube back in proximity to the restriction and the system oscillates back an forth.

[@DFPercush]

My first thought was entrainment. The high pressure atmosphere wants to push into that gap where the low pressure stream is, and if it's not able to form an airtight seal, there will always be a crack where it can start expanding and blow open. When the gap opens enough for the atmosphere to rush in, the velocity into the gap increases, and the pressure equalizes with the cone or tube. then the air current is able to pull the two pieces together again. I'm curious what would happen if you use a straight tube that's smaller than the inlet? (Mounted to something so it doesn't get sucked through.) Maybe there's an optimum size where the inlet would exert the most force on the tube, while still allowing atmospheric entrainment.

[@GoogleAreDumb]

I'm a fluid dynamicist by trade and training, and my best guess is fluctuating inflow from the gap between the "free" cylinder/cone and the fixed one. Air starts flowing in through the gap, and an interaction between that inflow and the main inflow sets up some kind of eddy or turbulence that exerts a force on the tubes. Would be interesting to see some smoke and a laser sheet used to visualise the flow in a few 2D planes to confirm what's happening. Fun little experiment!

[@GoogleAreDumb]

Further thought, perhaps the radial inflow somewhat constricts the axial inflow, creating a sort of "nozzle" effect that accelerates the axial flow through a smaller effective cross-section. The equal and opposite force to that acceleration would possibly be passed out through viscosity in the fluid and act to push the free cylinder back in the opposite direction? The force would be small, but the toilet tube is very light...

[@joshwand]

I agree-- the key is the gap-- my thought was that the air sucking through the gap is creating entrained flow-- more air approaching radially at the gap pushes the two pipes apart.

[@joshwand]

This is also perhaps testable with more viscous/visible fluid--a couple of aquarium pumps with repositionable silicone tubes. Or honestly, just use a sump pump to suck water through a submerged sprinkler! EDIT: there is a long history of this experiment being tested and re-tested! gizmodo.com/the-long-running-mystery-of-the-feynman-sprinkler-1497717528 sciencedemonstrations.fas.harvard.edu/files/science-demonstrations/files/reverse_sprinkler.pdf#page=2 en.wikipedia.org/wiki/Feynman_sprinkler

[@NotaRobot_gif]

Mech Eng here. I agree. There's less resistance to flow through the gap than thru the tube. I think adding a small collar to cover the gap may change the results.

[@CarAudioInc]

I love listening to people who know way more on a subject than I do. I have no idea what's going on, but I'm along for the ride.

[@migalito1955]

Cool concept to investigate. Now I really want to know.....

[@wallyhall]

This is the sort of curiosity I feel society is losing in the younger generation. Superb content, and inspiring!

[@silverXnoise]

I’ve got great news for you! You’re wrong.

[@brw3079]

Neil Degrasse Tyson would be proud of you 😅

[@k13ehr]

Interesting one Matt.

[@laureven]

Fluid dynamic master class :)

[@teenflon]

What an interesting phenomenon, physics is often very counter intuitive!

[@TurpInTexas]

That was unexpected. I should not have watched this so late in the evening. Now I'll be laying in bed, wide awake thinking about it all night long.

[@gshingles]

For the vertical case I think it could be a reaction force; the air at the side of the pipe detaches from the walls (coanda) and there's an opposite force as it does that. That would be why it is such a small effect.

[@zqzj]

Thought provoking!!

[@mikesmswoodworking]

Welcome to the new "Mr. Wizard's World". I love these videos. I would love to hear what The Physics Girl would have to say about this phenomenon...hopefully she can recover much sooner than later.

[@joethompson11]

Feels like one for Steve Mould. Would be interesting to see you guys collaborate on this.

[@billshiff2060]

18:30 Off hand I'd say because the straight tube will form a vena contracta at the inlet which behaves exactly like a bellmouth would even if there is no physical material shape of the bellmouth. The flow coefficient for a plain pipe like that could be ~.5 so it will behave like a bell mouth of half that diameter. Since there is no actual surface to suck against it will suck against the air in the vena contracta which will in turn pull upwards on the tube walls as it recirculates in the vena contracta.

[@matthiasrandomstuff2221]

yes, that was my initial theory too, but I could not find the air getting sucked back up when probing with a thread. But it's something along those lines.

[@billshiff2060]

@@matthiasrandomstuff2221 Probably any thread will be pulled into the main stream immediately. Can you try tiny short threads on the wall or some liquid in the inside wall and see if it climbs up or not. It MUST climb up because the torridal flow in the vena contracta certainly has to be there. Failing that a thread or smoke in the top will reveal the shape of the vena contracta and you could calculate the size of the throat. This is an interesting phenomenon and I'd like to know more. Something very similar has been used in hi perf engine intake systems.

[@dennisrech9149]

i believe that the high velocity air being channeled though the toilet paper tube contain an amount of velocity pressure. When it hits the elbow, or any restriction, it causes a reverse pressure wave. I believe that you have rediscovered "water hammer".

[@trunkmunk3y]

I think the air flowing through the pipe extension can only move so fast so the unrestricted air from outside can force it's way in which pushes the main pipe back but only so far before gravity takes over

[@frollard]

3:00 is neat. it makes sense that the air in the tube provides resistance...but in unintuitive ways: when pressed up close, it seals well and the straight horizontal tube air begins to speed up. This lowers the pressure in the tube, and creates the pull force...but then as it reaches steady state the pressure at the elbow and the pressure in the tube are equal-ish. ...but then the air has a high inertia as it hits the elbow and provides a large leftward force. Once the seal breaks there is a big glut of air available to the low pressure elbow, so the elbow is drawn forward once again starting the oscillation.

[@nascheme]

I think the behavior you see makes more sense if you change your perspective. The fan is lowering the pressure inside and at the end of the tube. The surrounding air in the room is at a higher pressure and so that surrounding air is pushing the air near the end into the tube. It explains why the "reverse sprinkler" doesn't turn backwards and also why you see that separation. The air from the room is trying to squeeze through the crack into the low pressure area inside the tube/pipe. The air in the high pressure zone is trying to find the easiest path to the low pressure zone. Thanks for making these kinds of videos. I'm trained in mechanical engineering but this kind of stuff is non-intuitive, even after studying. The simple "grade-school" or common sense explanations are often wrong.

[@mariecosmos4383]

my guess paused at 4:30 is that the L shaped tube pulls air through the cone, which then creates a force on the bottom of it, which pushes it back (we saw how much it was affected by small breezes just from blowing on it a minute before), and it's able to be pushed back since it's not attached.

[@rustyshackleford851]

The air rushing in is pushing left as it hits the 90 elbow so it's canceling out all the pulling force, when you lengthen the tube after the bend the air is accelerating faster and why it pushes the whole arm left away from the longer tube

[@danjeln]

This is so amazingly interesting. I’m watching this and try to get my spouse to come and watch and she goes. That’s nice dear. Clearly we have different interests :). Thanks Mattias

[@matthiasrandomstuff2221]

My audience is overwhelmingly male for a reason.

[@boelwerkr]

The movement of the suction pipe with the knee in front of the fixed straight pipe and the straight pipe above the suction use different effects for the movement. 0. The Bend pipe don't move much because it will suck in air from all directions around the opening of the pipe. Only a small section in the middle will make the pipe move forward. 1. The "loose bend pipe" experiment oscillated because of the weight of moving air that was changing direction and with that impulse. The air gets accelerated in the straight section of pipe and the enters the bend section, it has to change direction. It has to loose the horizontal momentum. This pushes the pipe away from the straight section. The movement in the fixed pipe stops and the bend pipe falls back into the default position. It starts to oscillate. 2. You're sucking air through the short loose end of a pipe that is not exactly the same diameter or orientation than the pipe sucking the air, these misalignment can result in wired forces pulling or pushing on the loose pipe. My guess here is that some of air is hitting the suction pipe not exactly straight causing sight changes of air movement and with that turbulence in the suction pipe that results in an repelling effect as long as the loose pipe section is close enough. As the pipe moves away the air flow changes and the loose pipe gets sucked towards the fixed pipe. EDIT: changed some wired wordings and typos. It's hard to write in a non native language after midnight.

[@Lyndalewinder]

As a retired electrical engineer I would never have thought I would be interested in a video about obsure effects of fluid dynamics......but I was!

[@matthiasrandomstuff2221]

And now you know something you didn't need to know :)

[@bradley3549]

Between this and Veritasium's new video, my mind is blown today by two things that I would have thought "this is obvious" before being proven wrong.

[@oisincarroll1171]

I'll preface this with: I'm not an expert, but I'm at least an engineer. I wonder if there could be some resonance going on in the long extractor pipe - maybe adding and removing the tube causes pressure waves that cause the air to accelerate and decelerate in the tube, and so cause the 'bouncing' we see. The fast-moving air in the short tube would suddenly come up against the slow-moving air in the extractor pipe and cause it to be pushed away. It'd probably be possible to disprove this by shortening the extractor fan pipe or adding a weight to the short tube. Fascinating video as always! Looking forward to the update🤞

[@Im_Ninooo]

I would love a collab with Steve Mould!

[@cardosera]

I was about to say that 😂

[@matthiasrandomstuff2221]

yes, it is indeed a very mould-y video.

[@yurishnirman3847]

Flow in the pipe, the speed increases with the length. A long enough straw and you blow into it, the exit speed at the other end would be speed of sound. Basically makes a invisible cone inside the straight pipe.

[@TheMrTape]

8:18 From the center of the holes perspective, the tube is cone shaped, and so it still makes low pressure on the outside. Air get sucked towards the center of the hole, at an angle towards the outer tube walls, rather than parallel along the entire flow-path. An analogy would be a 3D drawing or render, with perspective vs. no perspective; since the hole sees perspective, it effectively sees the tube as a cone.

[@GuitarDocAndMore]

the most astonishing thing for me is that you build an acurate 90 degree elbow out of paper towel rolls

[@matthiasrandomstuff2221]

I cut the towel roll on the table saw to get a nice miter cut!

[@CadeTrigon]

I wouldn't be surprised if I saw some of the big science channels taking their turn to explain this phenomena in a couple of weeks.

[@IvyMike.]

The tube is affected by turbulence, spinning air at lower pressure, that draws it towards it, but the air coming from behind filling the hole is keeping the tube down, it's all about equilibrium.

[@maitajack]

Very interesting. Aerodynamic is always complicated.

[@TKC_]

Possible it’s still the air flowing around the inlet edge having a similar but lesser effect to the low pressure entering a bell mouth. Would be interesting to modify the inlet geometry by cutting triangles or castellations to see the effect. I’m thinking the castellations could lower the velocity around edges that would tend to pull the pipe. The edges along the axis of the pipe would just cancel so the overall effect would be reduced if that is what it really is.

[@Gunbudder]

when in doubt, the answer is low pressure probably! lol i never fully understood fluid dynamics to that degree and luckily i don't have to for my work. i've learned that localized low pressure in a fluid can cause all kinds of very strange behaviors though. its similar to the "add a chunk of rubber" rule to solve vibration problems. i don't really know exactly how or why harmonics work, but i know gluing blocks of rubber to something will change its properties and make harmonic problems go away haha

[@bborkzilla]

When someone starts talking about eddy currents, I can't help remembering the great Douglas Adams. "Eddies," said Ford, "in the space-time continuum." "Ah," nodded Arthur, "is he. Is he." ... "What?" said Ford. "Er, who," said Arthur, "is Eddy, then, exactly, then?" Ford looked angrily at him.

[@tonyp9179]

The air flow around the top of the cardboard tube near the hinge is a factor

[@srmofoable]

That's a great demonstration for college level course. Unfortunately my fluid professor was so bad at teaching that they did away with it and changed the requirement. I had him for thermodynamics. I can attest to how bad he was. Incredibly gifted researcher though, just was not organized enough to put together a coherent lecture

[@ardenthebibliophile]

This would make a really fun demo in a physics class in high school

[@brandonhicks7549]

The air surrounding the pipe has no initial velocity and hence no momentum. The vacuum in the pipe starts accelerating the air just outside the pipe, and you would expect the bent tube to deflect in the direction of the outlet, but then the air hits the bend in the pipe which redirects the momentum towards the vaccum, but absorbs the imparted momentum, and practically stops deflecting.

[@thedevilinthecircuit1414]

Fascinating! Try attaching little threads to the *outside* of the short straight tube; I'll bet you'll see airflow moving upward around the outside in a pulsing oscillation. This upward airflow occurs so long as the tube's bottom is resting on the vacuum inlet beneath. But the internal downward flow creates low pressure that overcomes the weight of the tube and support arm, creating a lifting force--upward flow--on the outside of the short tube (Coanda Effect). This breaks the upward flow (Coanda Effect) around the outside of the tube when it lifts off the vacuum inlet, and the tube settles back down. The motion develops into an oscillation.

[@MacroAggressor]

To actually observe this you'd need to attach the strings to something stationary outside the tube. The tube's oscillation would spoil the results.

[@mckenziekeith7434]

I thought that when you sucked with just the elbow that it would move the same way as when you were blowing. I was thinking only of the momentum of the air molecules making the right angle turn. Not thinking about Bernoulli. Very interesting!

[@wtfrankian]

weight of tube could be damping the effect, possibly also some conflating airflow in the gap between the cardboard tube and the wooden plate which you can see at 0:39.

[@gavinmcmillan6222]

The eddies in the straight pipe you drew cause the same stream lines as the cone and Bernoulli velocity/pressure. You’d expect the same reactions but only have fluid resistance on the wall

[@ChristopherRoseP]

I have 8.5 years of experience designing turbo machinery, is just Bernoulli. The gap in between pulls air inward which separates both ends. Try inverting the suction cone you will see is not the slope of the first section is the gap that you have between both sections. The inlet also is pulled upward, you are right! Bernoulli again.

[@matthiasrandomstuff2221]

do watch the follow-up video for the answer.

[@csbarbourv]

I remember making circular paper airplanes that were amazingly efficient. No idea why they worked so well.

[@ohnonomorenames]

As the air is pulled through the gap between the hose and the tube by the venturi effect it is accelerated and constricted by the small gap. To constrict flow you need to confine it, this requires a force. With no force holding the gap constant the opening is forced open until the air rushing through the gap is greater than the air coming through the tube at which point the tube falls back to the hose. It may be possible to set up a stable arraignment where the pressure through the gap matches the weight of the tube, but this seems unstable. I think because once the gap between the tube and the hose gets too large the flow through the tube slows the venturi effect drops and the gap closes. As the gap closes it exceeds the stable point due to momentum and so the tube rises again.