Syphon problem with above ground water tank

[sweetbruce]

I have a problem with an above-ground water tank supplying a pump on the ground, supplying a sprinkler system. When the pump turns off, the water doesn't stop due to the syphon effect. When I first installed the pump, I put in an adjustable sprink check valve after the pump, but it still leaked and the water slowly ran out. I have removed the check valve. I wonder if I put the check valve back in, and after that a tee with a swing-check to allow air to come into the system when the pump has stopped and the pressure is reduced. I consider doing the swing check in a section of pipe that is higher than the pump. The water tank is 5000 liters and about 3 meters tall. The lines are all 2". I am not able to upload my drawing due to poor internet connection.
Any ideas? Thanks!

 

[Q_Goest]

Hi Bruce. welcome to the board. I like your idea of putting in an adjustable, spring loaded check valve downstream of the pump. I would put that back in and adjust the spring load to ensure there was no flow out of it with the pump turned off. If, once the pump is on, the flow rate seems too low, you might need to get a bigger check valve (ie: one that opens more at a given cracking pressure).

I wouldn't do that thing you're thinking of regarding the swing check though. I'm not sure exactly what you're trying to do there but the only other thing that makes sense would be to have a vertical pipe downstream of your pump that went above the height of your supply before dropping back down. Basically, you can either produce a back pressure on the pump with a spring load or a vertical column of water. Either way will work, as will a combination of both, but it makes no sense to do a combination of both IMO. Just do one or the other.

 

[sweetbruce]

Thanks Q Goest, my idea with the swing check was to allow air into the pipe to "release" the syphon "suction". Before I had tightened the adjustable spring check up pretty good, but it still allowed the whole tank to drain (well, not really, because it kept filling up from the well). But I had wondered if the pipe after the pump was elevated to higher than the tank (as you suggest), if that would stop the syphon. I didn't know if the syphon would still occur, due to low point and high tank.
One thing I'm not clear on, if the syphon effect will pull/push the water over one high point (where it now exits the tank) will it not pull/push it over a second high point?
But if the higher pipe will do the trick, I will take that approach, as another concern was the check valve breaking at some point, from the 3 hp pump.
Thanks a lot!

 

[Q_Goest]

Hi Bruce. Maybe I'm just not understanding what you want to do with this second check valve. You say you want to stop the syphon effect. I'm assuming the tank is higher than the pump and the pipe from the tank to the pump is all below the height of the water level in the tank. Is that true or not? I'm also assuming the outlet of the pipe, where it dumps onto the ground (sprinkler system) is also below the height of the water level in the tank. Yes/no?

If the pipe goes above the level of the water in the tank before it goes into the pump suction, you could break the syphon effect there, but if so, you have air in the line before the pump which will make it cavitate and probably not be able to restart when you want it to. How would you restart the pump? How would you get water to the pump suction?

One thing I'm not clear on, if the syphon effect will pull/push the water over one high point (where it now exits the tank) will it not pull/push it over a second high point?

Yes, the water could certainly get drawn through a second or subsequent high point (a point that was higher than the level of water in the tank) if there is sufficient flow. To break this, air has to get up into the pipe, so if it can't get up into the pipe at the outlet, then yes, the water will keep flowing. You could put a check valve in downstream of the pump that allowed air into this high point when the water shut off. Perhaps that's what you meant about the swing check?

 

[sweetbruce]

Yes, the outlet is lower than the tank (outlet are at "0" and bottom of tank is at "0", but tank is often full or filling).
Yes, I meant a "T" with a swing check valve downstream of the pump. The T and check valve vented to the air. When the pump is running the check valve would be pushed closed, when the pump was off (hopefully) the check valve would open from "suction", allowing air in. I'm thinking this T and check would be at the high point (higher than top of tank) of the output pipe. Sorry, I can't upload a drawing (slow connection on an island in Thailand).
Thanks a lot!

 

[Q_Goest]

Yes, I meant a "T" with a swing check valve downstream of the pump. The T and check valve vented to the air. When the pump is running the check valve would be pushed closed, when the pump was off (hopefully) the check valve would open from "suction", allowing air in. I'm thinking this T and check would be at the high point (higher than top of tank) of the output pipe. Sorry, I can't upload a drawing (slow connection on an island in Thailand).
Thanks a lot!

Yes, that should work.

 

[Danger]

This is one of those situations where my lack of education gives me a different approach to things, which is probably wrong. Still, I must mention it just in case.
My thinking would be to replace your check valve with an electric diversion valve (somewhat like the kind used to switch between fuel tanks on a vehicle) which is powered by the same circuit as the pump. When the pump is running, the valve passes the water straight through to the sprinklers. When it's off, the valve diverts the flow to another line which leads back to the tank. The syphon effect shouldn't operate, since the return line will be higher than the outlet. If it does, however, the water will go back to the tank rather than be wasted.

 

[NeuronsAtWork]

It sounds like you've about got it figured out, but I thought I'd throw my two cents in just for good measure. I see no need for a spring check valve, necessarily, as it will add restriction to the pump discharge and decrease efficiency somewhat, and it's just something else to fail. Ideally, just run the discharge 2" line out of the pump, then up to a point above the highest water level in the tank (3.5 meters?), and then back down to the original discharge to the sprinkler supply. At that high point add a tee and install a 'vacuum breaker' open to atmosphere at that point. A swing check might not close immediately and could discharge some water, but a vacuum breaker, which is common and inexpensive, is made for this purpose. When the pump shuts off, it will open and release the suction, allowing air to replace the water in the vertical rise and isolating the pump discharge from the sprinklers. When the pump restarts, it will close, and the pump will push water up over the top and back down to the sprinklers. Hope that helps a bit...

 

[dschlink]

Much simpler would be to install a vacuum breaker at the high point of the line. These are commonly used in irrigation systems to prevent ground water from seeping back into the lines. All they are is a check valve that is held closed when pressure is in the line. When the pump shuts off, the pressure drops and the valve allows air into the line, breaking the siphon. Almost any place the carries lawn irrigation equipment will have the valves.

 

[sketch]

Much simpler would be to install a vacuum breaker at the high point of the line. These are commonly used in irrigation systems to prevent ground water from seeping back into the lines. All they are is a check valve that is held closed when pressure is in the line. When the pump shuts off, the pressure drops and the valve allows air into the line, breaking the siphon. Almost any place the carries lawn irrigation equipment will have the valves.

Or lotech solution - some sort of small 'header' tank above the level of the main feeder tank, thus negating the syphon effect completely...or set it up to be gravity fed with a solenoid operated tap...