Design for Minimal Flow Orifice for Pump

[RJB32482]

Hello,
I'm looking to design a minimal flow bypass line for a pump. A orifice will be needed to make the flow minimal through this line during normal operation. Here's what I believe the sizing procedure looks like:

1. Look at pump curve to find head at minimal flow for pump (maybe 5-10% safety factor on flow).
2. Calculate pressure drop through bypass line without orifice (will of course be lower than head on pump curve at minimal flow).
3. Design orifice for pressure drop equal to head on pump cureve at minimal flow - calculated drop through bypass line without orifice.

Am I correct with this procedure?

Thanks.

 

[25362]

Visit thread124-107205 and Pump Handbook by Karassik et al., McGraw-Hill.

 

[MJCronin]

RJB...

Some things to consider:

1) Not all centrifugal pumps need a minimum flow line or device. A single stage centrifugal pump may operate without such a device. All PD pumps must have such a device

2) Boiler feed pumps and other notorious high-head, multi staged pumps should have the device that is specified by the pump vendor. There is a large amount of career liabilty in sizing such devices.... this is always best left to "others" IMHO

3) Yarway makes ARC valves that are the power industry workhorses in this multi-stage application.

http://www.yarway.com/control_valves.asp

Please, tell us more about the pump......!???

My opinion only....

-MJC

 

[RJB32482]

The pump vendor did recommend a minimum flow of 29 GPM on the pump. I'm going to size the line at worst case (all valves that pump discharge goes to are shut and all flow goes through minimum flow line).

Thanks for all of the recommendations and post forwards.

I'm just looking for a clarification of the calc method I used above. Looking at pump curve at 29 GPM, I can find the head developed. Then I would need to size the line for min flow to pass 29 GPM at that head. Line would either have to be super small or need a orifice.

So orifice sizing would have a delta P of (Head of pump at 29 GPM - pressure loss of bypass line at 29 GPM without orifice).

Correct?

 

[sbush]

Instead of a fixed orifice design, perhaps you would consider using a pressure compensating, variable area orifice device that can maintain a constant maximum flow rate regardless of wide changes in operating pressures. This is not an actuated valve. We have used them for many years with great reliability. If interested, go to

http://www.griswoldcontrols.com

and look up automatic flow controls.

S. Bush

http://www.water-eg.com

 

[GL431]

RJB, correct. The procedure is OK for run of the mill applications and is cheap.

With the fixed orifice, the min. flow will always be pumped additionally to the required process flow. Can be energy inefficient in some applications. Also watch out for fixed orifices in high deltaP applications. They can be sites at which cavitation can take place.

Some of the compensating variable orifice devices described are nice and can save some energy in the long run. However, they are expensive and often require "clean service". I have heard of some cases, where the "variable" bit was removed after some operation and it was a fixed orifice after all.

 

[pumpking]

We do this quite often when supplying pumps for CIP in breweries etc.

You need to start by calculating what pressure you need the pump to generate, static head, how many 'take-off points do you want to operate at once and friction losses along these stages. Once you know this you can then select the pump.

Once we know the pump, use rule of thumb 10% of BEP on selected Impeller (for single stage centrifugal pump) and work out ring main friction loss and do orifice plate calculation. It will be more efficient also to have a pressure relief valve on return pipe into main storgae tank, therefore, when product is required at a take-off point, PRV closes and makes pump operate at most efficient point.

http://www.cdrpumps.co.uk

Ash Fenn

http://www.cdrpumps.co.uk

 

[davefitz]

pumpking is correct - the use of a pressure relief valve can also be used to advantage in maintaining the pump outlet piping system at a lower design pressure, if the relief valve is code rated and isolation valves are not provided.

gl431 is correct- cavitation is a big problem if you are using a single orifice and the pressure drop is high. For such a small flow of 29 gpm, you might want to consider the use of a SS "capillary tube" in lieu of an orifice.