I was curious about this. If i have a tube which has water flowing in it. and mounted vertically a small extension tube with a loop. The water will go horizantally and only a few drops will go vertically. Why is that? and how can u determine how much water will go vertically? I tried to search for a scientic explaination but didn't find anything. What if I want to design a tube with this loop but I don't want any water to go vertically?

Here is a diagram if you don't understand: enter image description here

Is there anyway to prevent the water from going verically if the tube is blocked?

  • $\begingroup$ Flip it upside down and check the results... This boils down to "why doesn't water flow uphill?" $\endgroup$
    – SF.
    May 26 '17 at 11:36
  • $\begingroup$ What would the loop be used for if it should remain empty of water? If it's a line tap or something for a gas hookup, you can install a check valve (or any valve really) that would prevent any upflow (water into the tube) $\endgroup$
    – J. Ari
    May 26 '17 at 17:37
  • $\begingroup$ @J.Ari: If so, the pressure after the valve would be different. the value is not the same before and after the valve. that's why I didn't add a valve. $\endgroup$
    – Manaf
    May 29 '17 at 6:05

Water likes to take the path of least resistance.

In this scenario, you have 2 paths for the water to take when it reaches your T-junction.

If we compare the two paths, we can see why it doesn't want to travel through the vertical tube. First, to go through the vertical tube it would need to overcome gravity. The tube seems thinner, so it would also have greater frictional losses as it travels through it. To add to that, it has a bend, which is another source of friction loss compared to a straight pipe.

If you're familiar with basic DC electric circuits, you may know that if you have current travelling through parallel resistors, the higher current will flow through the lower resistance branch. This is an analogous situation. The current is like the flow rate, the resistance is like the frictional and gravity losses, and the voltage is like the pressure.

You can use the equations of Bernoulli's Principle to get some idea on how these losses will effect the flowrate.

If you want to prevent water flow vertically when the tube is blocked, you need to make sure the pressure losses in the vertical tube due to gravity and friction are greater than the maximum pressure you can build up. This will depend on how you are getting your flow, if it's a pump, the pump may stall or throttle itself once it reaches max operating pressure. If it's flowing from a higher point, the safe thing to do would be to make sure your tube is higher than the high-point you feed in from. (you can probably get away with slightly below the high point due to friction losses on the way there, especially depending on how long the pipes are).

  • $\begingroup$ Thank you so much for replying. That explains everything! the flow source is a pump and by the time it throtles ( if blocked), the water will reach the vertical tube. Is there anything else which might work? $\endgroup$
    – Manaf
    May 26 '17 at 11:47
  • $\begingroup$ No worries. I usually try to go for the circuit analogy since most people studying fluids have already done an into physics course with simple electric circuits. Plus then you get to see that electrical and fluid systems can be very similar (and it gets really cool when you start learning mechanical systems and everything else the same way). For the pump thing, you'd probably have to play around with the pump setup or your piping. I don't know if you'd be able to adjust the stall pressure of the pump or not. $\endgroup$
    – JMac
    May 26 '17 at 11:49
  • $\begingroup$ @JMac How do you explain magnetism with a water analogy? Water pressure : voltage and water flow is current but capacitance? Inductance ... $\endgroup$
    – Solar Mike
    May 26 '17 at 11:51
  • $\begingroup$ @SolarMike That's quite a bit harder, but I mean if you get into the analysis of the controls system input vs. output you could come up with an analogous model. Capacitance, inductance etc are all modeled by the differential equations that represent them. It still uses currents and voltages (just the rates now matter as well). $\endgroup$
    – JMac
    May 26 '17 at 11:52
  • $\begingroup$ @JMac I find pressure and flow work fine and sets many up as a starting point - after that it does not help so much. $\endgroup$
    – Solar Mike
    May 26 '17 at 11:54

Depends on the internal and external pressures. Apply the Bernoulli equation to the situation. Also check out coefficient of discharge and the Coanda effect. To stop the flow going vertically block it same as the tube.

  • $\begingroup$ Thank you for your replay. I got the idea from your reply before btw and I tested the idea today. Using air it worked and i still have to test it with water! I'll check them today and hopefully I'll get them. Blocking the T-junction is not an option here since I want to measure the pressure. $\endgroup$
    – Manaf
    May 26 '17 at 11:45

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.