There is a hard metallic pipe in your dark wine chamber. You hear the water flowing slowly in the pipe but you cannot understand in which direction that water flows since it is a closed system.

What could be the easiest way to find in which direction the water flows?

  • 4
    $\begingroup$ As a physicist: You have a valid solution? Because if you have the real problem and asking for a solution, Physics.SE is the right group. Sound in water travels with 1500 m/s and water flows with 1m/s, so it is unlikely to have a layman sound-based solution. Convection has the problem that metal is a much better heat conductor and we do not know how long the system is. $\endgroup$ – Thorsten S. Apr 1 '18 at 13:43
  • $\begingroup$ Does "dark wine chamber" mean a tank full of red wine or a cellar with poor lighting? $\endgroup$ – Peter Taylor Apr 3 '18 at 6:48

You could try

heating the middle of the pipe, say with a candle, then touching either side of it. The side which is warmer should be the direction water is flowing, since the metallic pipe would conduct heat through to the slow moving water.

  • 5
    $\begingroup$ I haven't done any calculations, but one would think that the amount of heat transported by the flowing water (water has a notoriously high heat capacity, it takes a lot of energy to heat water) would be completely masked by the huge amount of heat conducted by the metal pipe. (Most metals are very good at conducting heat.) $\endgroup$ – Bass Apr 1 '18 at 7:24
  • $\begingroup$ I hope the heat conducted along the metal pipe would tail off due to radiation and the side with incoming water would cool the pipe on that side (and that the heat supplied by a strong enough candle would cause the the pipe to become sufficiently hot at the point applied). Maybe testing by touching both fingers at the same time/distance would be sensitive enough to detect a temp difference. $\endgroup$ – Tom Apr 1 '18 at 7:39

The direction is

towards that end of the pipe where you can hear more sound than the other end.


The water is not stagnant, it must enter at a point of the pipe with full force(thereby it touches the entire cross section/ internal surface of pipe...thereby resulting lesser noise)..and leave at a point with a lesser force/pressure.

  • $\begingroup$ Great solution, I think this will work as (rot13) gur nzbhag bs gheohyrapr jvyy cebonoyl vapernfr tbvat qbja gur cvcr. $\endgroup$ – tom Mar 31 '18 at 9:58

I am not sure how easy it is,

but one way would be to place two microphones at each end of the pipe, hit the middle of it sharply and record the signal of the microphones.. the sound that travels through the water will come slightly earlier to the microphone in the direction of the flow, but I think this is probably too hard to be the correct solution

  • 1
    $\begingroup$ probably faster in metal than in water. $\endgroup$ – Jasen Apr 1 '18 at 6:04
  • $\begingroup$ @Jasen - yes, of course, ... rot13 gur fbhaq gung geniryf guebhtu gur jngre - gurer jvyy nyfb or fbhaq gung geniryf guebhtu gur zrgny naq guebhtu gur nve, naq gur fvtany qhr gb gur fbhaq gung geniryf guebhtu gur jngre jvyy arrq gb or qvfgvathvfurq - ernyyl vg vf n pbzcyvpngrq zrgubq, juvpu vf abg gur pbeerpg nafjre urer ---- I didn't want to go into all of those details in the answer as it just gets too complex - hope this makes sense $\endgroup$ – tom Apr 1 '18 at 9:26

Use an acoustic Doppler current profiler (ADCP), although this may not work well through a metal pipe.


As the water is flowing, there must be a pump of some sort. Find the pump and check the labels on the pipes. This is more of a home maintenance answer than a 'logical deduction' so.. maybe I'm missing something clever in the question.


I would not use Tom's answer because

the pipe is against the wall and I don't want to burn the place down.

I would instead try to

get my hands wet and drip some water straight down onto the outside of the pipe,
and then observe the direction in which the drop flows.
If the drop stayed put, I'd drip more directly down into it, to enlarge the drop.
Or, I'd try a plumb line and a level, if those were handy.

  • 3
    $\begingroup$ why would water outside the pipe be influenced in any way by water inside? $\endgroup$ – Orphevs Mar 31 '18 at 17:02
  • 2
    $\begingroup$ Are you suggesting that the pipe is necessarily on a slope? It isn't, necessarily. And water will happily flow uphill locally, as long as it's flowing down globally (that's how a syphon works). Or it could be being pumped uphill. $\endgroup$ – David Richerby Mar 31 '18 at 22:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.