Question about hydrostatic pressure

1. May 20, 2017

Tan

Hi,

I have a 12(W)x18(L)x1m concrete tank installed underground and it will be subjected to pressure of 0.5bar.

To test the tank for leaks, instead of pressurizing the tank using pumps or air blower, my supplier suggests we use a 1 inch hose with 1 end connected to the top of the tank (completely filled with water), and connect the other end to an elevated IBC 10m above the top of the tank.

He claims that this way the tank will be subjected to pressure of >0.5bar and <1 bar.

What do you guys think? I mean it sounds logical because he is trying to create the pressure using hydrostatic pressure. But is this 10m water column in the small 1 inch hose able to provide the sufficient pressure to test the huge 12x18x1m tank? If the tank is completely filled with water, there will be no flow, and Bernoulli equation would dictate headloss=0 and the static pressure in the tank would be 10m?

Back then in school the examples given were otherwise. The tank providing the hydrostatic pressure would be huge, while the receiving end at the bottom are pipes. I can fully understand this part taught in school. But to put it the other way round with huge tank at the bottom and pipes providing the hydrostatic pressure, I cannot reconcile. Advice please? Thank you so much.

2. May 20, 2017

scottdave

You would definitely need that much pressure to pump a column of water up that high.

3. May 20, 2017

CWatters

Yes. Provided the tank is sealed to the hose. Fit a pressure gauge to the hose where it connects to the tank to be sure.

You don't really need an IBC either, you just need some method of filling the hose that ensures it's full of water with no air.

4. May 20, 2017

Tom.G

Good idea. 1BAR is 10.5 meters of water, and you can use an ordinary household garden hose, it doesn't have to be 1 inch diameter. Biggest problem I see is getting the end 10M up in the air!

For filling the hose, use a Y connection at/near the bottom end and leave the top of the hose open.

5. May 21, 2017

scottdave

While it may sound difficult that this would work, just think about how hydraulics work: A small force, applied to a small area can create a large pressure, which can then be applied to a large area to lift a large load. Note that if you do have a significant leak, your water column will start dropping quickly, reducing the pressure (and the leak might then stop as the pressure reduces. Just some thoughts.

6. May 21, 2017

NTL2009

Yes, this seems to be the hard way of doing it. Since the tank is being filled with water anyhow, why not simply use an air compressor or even a bicycle pump to get to 1 bar?

7. May 22, 2017

Asymptotic

Air is absorbed by water. I don't know if the rate is high enough to affect a quick test, but it probably would be observable over the course of several hours. Before bladder tanks became the norm, home well water systems used air-over-water captive tanks (and still do, where their advantages outweigh the cons). Every couple of weeks or months these would become completely waterlogged due to absorption, and it would be necessary to partially drain them down to re-establish the air headspace.

I like the 10 meter hose idea - neat and simple - except for the part about getting one end of it 10 meters into the air. If clear tubing is used, elapsed test time and water level can be checked and recorded periodically to perform a leak rate test.

If a pneumatic or hydraulic piston cylinder is handy (2" bore x 6" stroke should do), one way to approach it would be to use a shorter hose (say, 1 meter) with the piston side of the cylinder connected to the upper end of the hose. Plumb a ball valve to the rod side cylinder port, and a pressure gauge to the piston side. Open the ball valve, and fully retract the piston rod while filling the rod end with water. Next, fully extend the piston rod while filling the piston side with water (caution: water will be shooting out of the ball valve). Now both sides of the cylinder will be completely filled with water.

Finally, retract the piston rod until tank water (at the piston side of the cylinder) is at the desired test pressure, then close the ball valve. Provided the piston and rod seals are in good condition, the cylinder will remain hydraulically locked in position, and test pressure should remain constant unless tank leakage is present.

8. May 22, 2017

Nidum

In the UK you would probably not need the 10 metres of vertical hose .

Guaranteed minimum head for water in the domestic supply mains in most areas is 7 metres but it is usually much greater than that . The supply head where I actually live is seldom less than 20 metres .

Connect your tank to the mains supply and check that the available head at the time of the test is at least 10 metres .

9. May 22, 2017

NTL2009

Good point, but I really doubt that air absorption would be significant over a few hours and a small surface area of non-agitated water. The domestic water tanks you mention will have a hundred gallons or so of water a day coming in contact and flowing over a fairly large surface area of air ( ~ 18" D). If the water became saturated with air (which happens at only about 8 ppm at NTP), it is replaced by fresh water over the course of the day. At any rate, simply pouring a 1/4" of cooking oil on the surface of the water in the connecting tube would provide a barrier.
I'm guessing they are not near a municipal water supply?

Last edited: May 22, 2017
10. May 22, 2017

scottdave

I was thinking the same thing about the the method for raising a hose 10 meters off the ground (and filling it). Another thought. I was also thinking about just muicipal water supply as a source of pressure. You could probably build yourself a small expansion tank out of some large PVC pipe, and cap it and mount a connector for the hose, then another connector for a air valve, then just use a bicycle pump to air it up (or a portable air pump that you can plug into your car). Many pumps even have a pressure gauge on it.

11. May 23, 2017

Asymptotic

And you have the better point. :)