Discharge Head 3

[Roach]

If I have a dischrge line from my pump that runs up vertically to 102" (elevation), horizontally 60", and then back down to 58" (elevation) to dump at atmosheric pressure, is the discharge head 102" minus friction, or 58" minus friction?
I am wondering if a siphon affect occurs that does not require 102" of head after line is initially filled with water to pump to the 58".


Thanks

 

[JJPellin]

If the vertical leg down the outlet stays liquid full at all times, you recover the head from that drop as a result of a siphoning effect. The total head would be the 58". The limit on recovery of head from this configuration is 32 feet for water. If the vertical drop down the outlet is more than 32 feet, the siphoning effect will create a vacuum at the high point in the line. Since a complete vacuum will only support up to 32 feet of head in water, this is the most that can be recovered. As an example, we have a water pump at the river that pumps to our refinery. The line goes up hill for about 100 feet elevation and then drops back down the other side of the hill 50 feet of elevation drop. The total head requirement of this pump is 67 feet - 100 feet up the hill and 32 feet recovered coming back down the other side.

 

[Roach]

Great explanation.

Thanks,
Roach

 

[CHD01]

Roach:
1. The pump must first be able to overcome the static lift of 102 inches. This is the startup requirement to lift liquid to the highest point of the pipe system.

2. In addition to the static lift, you must ADD (not subtract) the head loss due to pipe friction for the required flowrate.

3. Once the liquid fills the piping and IF up-going and -down-going pipe all remains liquid filled, then the TDH is reduced as stated.

4. BUT the down-going pipe line is only limited by vacuum IF a vapor lock occurs at the highpoint. IF it remains liquid filled completely then it is not limited except by the capability of the pump; and operation will occcur at the intersection of the pump impeller curve and the system curve.

The more you learn, the less you are certain of.

 

[joerd]

I have to disagree with your point 4. As soon as the pressure in the pipe drops below the vapor pressure of the fluid in the pipe (water in this case) at the operating temperature, there will be vapor generation and the siphon effect is lost. In normal ambient operation, the vapor pressure of water is very low, so you can recover up to 32 feet of water, as JJPellin wrote. Note that the vacuum condition at the high point may affect piping and fitting design.

Regards,

Joerd

 

[CHD01]

joerd:

I think we are both correct, the difference is in what the actual operating conditions are. In my case I'm just saying that IF there are discharge restrictions that keep the pipe liquid full, then the pressure in the pipe will exceed the vapor pressure and flow is not limited except by the pump and the system. If there is NO idscharge restriction to do so and the length of the pipe is sufficiently long, then flashing occurs as you describe. Operation should be at the intersection of the pump and system curves. But in the latter case, the system as been split into two; one ruled by gravity fall of the liquid and the other ruled by the ability of the pump to overcome the static lift.

At least this is the way I view the examples. The more you learn, the less you are certain of.

 

[joerd]

CHD01, I agree. The only way to really make sure that there is no flashing is to work out what the pressure will be at each elevation change or valve/long pipe run/restriction, and check if the pressure is above the vapor pressure.
This will be obvious in most cases; you will usually want a liquid full system in any case, and therefore provide a restriction at the end if needed. It may be critical in cases where there are large elevation differences or if you're working with a high vapor pressure liquid. Regards,

Joerd

 

[BobPE]

joerd:
what you are talking about is understanding the hydraulic gradient of the system. The pipes will still flow in vacuum, but when the gradient is suffieiently below the physical pipe to allow vapor pressure to dominate, these sections of pipe will cavitate. If its a pressure system, the pipe will be liquid filled. In some systems, the discharge will allow air to enter the pipe, this increases friction due to the decrease in flow area and increase in velocity, but the system is still under the pressure flow regeim. The restriction at the end serves to reduce of eliminate these headlosses, usually its as simple as an airlock which is an inverted elbow or something similar. All to often I see this detail missed in pump discharge applications.

BobPE

 

[greg87]

A key incorrect assumption that most of the replys make is that "the vertical down line will be full of water". It will not be full of water, unless there is an orifice or valve creating some backpressure at the end of it, which changes the whole hydraulic calc anyway. Remember that for a vertical line, the increase in head due to gravity will be 1 psi every 2.3 feet. The only way that the pipe will remain full is if there is a frictional loss of 1 psi every 2.3 feet of pipe - which may happen at some very high velocity, but not what we are discussing here. The line will only be partially full, with air pockets. There will not be a net siphon effect. The frictional pressure loss might be based on a sort of 2 phase flow; but whatever it would be, it would be balanced by the gravity.

Since we are discussing the subject of full pipe flow, note also that, depending on the velocity in the system, the horizontal section of pipe may not be completely full of water. For instance, in a 3 inch line, the minimum velocity to ensure that the line is completely full (minimum seal flow) is 7.3 feet per second.

The siphon effect should never be counted on for calcs in an open system unless you're sure that there are components that create a pressure loss at or prior to the outlet. In a closed loop system (assuming all of the air is vented) there will be a siphon effect.

 

[BobPE]

greg87:

If the pipe is not full of water, then it is a gravity system and I don't think thats what he is talking about. Look at my previous post and it may help you. Velocity has nothing to do with the problem as the HGL may actually fall below the pipe and will force the system in vacuum conditions. People make incorrect assumptions that air can enter from the end of the pipe, but this affect only extends a minimal distance into the pipe when it does occur and does not generally allow air to enter the system although it could affect the friction loss at the pipe exit somewhat. The voids at the orifice plate at vapor cavities induced by the losses across the orifice, they are not air.

BobPE

 

[abeltio]

Do not forget to put a vacuum breaker and a vent at ALL the the highest pointS in the installation.
The vacuum breaker will avoid crushing the pipe under external (atmospheric) pressure... this is very common in irrigation systems (crushed pipes).
The vent will ensure that the pipe is full during start up and will ensure smooth pump operation.
The pump MUST be able to fill up the piping system completely (i.e. must be able to deliver the suction lift + discharge head + friction losses during start up when the pipe is empty).
HTH
Saludos.
a.