Siphon systems with pump priming 7

[konrads]

Can anybody clarify the valving/pump arrangement for a pump priming system which then supports the formation of siphon. I am looking for practical examples where this has been employed. We are also seeking advice on the pump type required. The flow rates are i teh order of 75 l/s but with mimimal head less than 6 metres.

 

[BRIS]

Can you clarify. Your question implies that you have a pump at the crest of a siphon and intend the pump to prime the siphon. i.e evacuate air from the siphon which will then flow by gravity (siphonic action). So you want a vacuum pump??

 

[BigInch]

You will want a self-priming pump with the capability to start with a suction head start capability of at least equal to your siphon height over the liuqid level plus some additional height as a safety factor.

You may want to consider a check valve in the suction line to help maintain the siphon flow and also an air vent valve in the discharge line to remove air previously contained in the siphon that will be displaced into the discharge line as the pump starts.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[bimr]

Not sure what you are looking for from the post.

Are you interested in a self priming pump?

http://www.grpumps.com/products.asp?action=display&id=166&level=&kind=&kid=

 

[konrads]

Of particular interest is the valving arrangement used to establish the siphon. I am looking for the pump to draw water from a lake ie it requires approximately 2 metres of suction lift and then for the pump to clear the line of air so that a siphon can be established and then for the pump to switch off and the system run as a siphon. At a first pass I had in mind a pump connected to an intake line with a bypass line the same diameter as the main line fitted across the pump with a check valve to allow flow out of the lake only. With such a bypass arrangement I am not sure whether the siphon would be maintained as the pump is siwtched off and I am looking for any past examples.

 

[katmar]

If you install a non-return valve at the inlet side (in the lake) and an isolation valve at the other end you should need to use your pump to initiate the siphon only once. If you stop the flow by closing (slowly!) the isolation valve at the discharge end then the pipe will remain full of water and you will not need to refill it every time. If you stop the flow by breaking the siphon you will need to go through the whole initiation procedure every time you want to use the siphon.

Any pump will do if you are only going to have to do it once. Install a filling line and an air release line (both with isolation valves) at the highest point, close the discharge valve, and start filling the line and releasing the air. Once the line is full and the air is all gone close off the filling line and the air release line and you are ready to roll.

If you want the pump to re-initiate the siphon automatically every time you will need a much larger pump.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[bimr]

I would not recommend a non-return valve (check valve) on the inlet side because over time, the non-return valve is going to catch debris and plug. The non-return is also not going to hold the water over any length of time either since the very low pressure is not going to be enough to hold the non-return valve totally closed. Water is just going to leak past the partially closed valve and empty the siphon. A non-return valve is nothing more than a check valve.

You should use a full port block valve installed below the lake water level on the discharge side of the siphon. Install a a tee fitting on the siphon and connect the siphon to a self-priming pump, eductor, or vacuum pump. Once you have all of the air removed from the siphon, valve off the pump and slowly open the block valve. You are going to have use a large pump to get this done since the siphon pipe is probably going to around 25 cm or so in diameter.

Not sure why you want to use such an arrangement on such a large flow. Is this a temporary arrangement? Why don't you just install an overfull weir?

 

[BigInch]

To avoid getting wet opening and closing an underwater valve or a long extension handle, maybe a filter screen on the inlet will keep the debris out of the intake and a allow using a check located in the riser. A spring loaded check or ball check will tend to remain closed, at least for a short time.

BigInch-born in the trenches.

http://virtualpipeline.spaces.msn.com

 

[miningman]

You need what is known as an eductor pump. This is a vaccuum pump installed at the highest spot in your system. So long as you remove 90-95% of the air, provided you have enough head differential on your siphons, the liquid flow will displace the remaining air. And yes you do need a valve on the outlet side, preferably close to the discharge elevation, but it does not need to be submerged.

I have run 12 inch diameter siphons with these pumps.

 

[Petroleumman]

How to select the vacuum pump? If we plan to install the vacuum pump over there.

 

[bimr]

The size of the vacuum pump is going to depend on how much air that you intend to evacuate and how long that you want the evacuation to take place. That is going to vary depending on the size (volume) of the pipe and the time available.

Probably the most practical approach is use a small eductor connected to a small water pump to evacuate the air, rather than a vacuum pump.

You can make some assumptions. Call someone like Penberthy to get the particulars on the eductor.

http://www.penberthy-online.com/jet7c.asp

 

[bimr]

http://www.penberthy-online.com/jet8c.asp

 

[SteveWag]

Why not either a suction lift pump (self priming, Gorman-Rupp) just above the water, or a submersible just below the surface (4 -5 feet). Then either a non-submerged discharge (self venting), or an air release at the top of the hair-pin.

Run the pump for 5 min. and turn it off.

No valves at all, let the water go through the pump. The pipe will not freeze in the case of the non-submerged discharge.

Never rely on checks, sooner or later they leak.

Steve Wagner

 

[bimr]

Sure the pump will work, but you are talking about a large pump to get 75 l/s. Plus, you have to get power to the pump. With a small pump, you can use an engine driven pump.