Diaphragm pump calculation for sizing

[jfaucher]

Hi everyones,

When I'm sizing a centrifugal pump for an hydraulic system, I'm looking for a pump for which the head rise at requested flow will meet the sum of suction loss and discharge loss at the same flow (assuming I'm pumping fluid from/to reservoirs opened to atmosphere, at same level).

Now, my question is related to similar calculation for a diaphragm pump. It appears to me that this kind of pump works in 2 strokes sequentially: a suction stroke while a check valve stops the flow in the discharge line, followed by a discharge stroke while another check valve stops the flow in the suction line.

My assumption is that for sizing the pump, the head rise shall meet discharge loss only (without considering suction loss), after what I must verify that suction loss doesn't exceed the suction lift capacity of this pump. Is that correct?

Jean-Pierre Faucher, ing.

http://www.cel-aerospace.ca

 

[ione]

Diaphragm pumps are classified as positive displacement pumps. For these pumps the flow rate at a given speed is not dependent from the discharge pressure. Anyway there’s nothing strange in the sizing procedure.

Consider:

1. Suction side. You need a minimum amount of absolute pressure available to supply fluid to the pump suction. You have to calculate your net positive inlet pressure available (NPIPA) that is the is the total static pressure available at the pump suction centerline, including atmospheric pressure, less all pipe losses at your rated flow, less fluid vapor pressure (practically the same as the NPSHA). Then compare NPIPA with net positive inlet pressure required (NPIPR).
2. Discharge side. You have to intersect your pump performance curve (head vs flow rate) with the system response curve for your rated flow. Be aware that positive displacement pumps have a practically a vertical performance curve, so flow is nearly constant.

 

[jfaucher]

I forgot to mention that my diaphragm pump will be compressed-air activated. Even if this kind of pump is still classified as positive displacement, its performance curve is not vertical at all. Therefore, the resulting flow depends on pressure losses. But should I considere only discharge side loss, since there is no flow in the suction line while the fluid is expulsed from the pump?

Another assumption would be to considere pressure loss in both sides, assuming that the resulting head rise is actually an average of the suction stroke and the dischage stroke...?

Jean-Pierre Faucher, ing.

http://www.cel-aerospace.ca

 

[TenPenny]

If you're talking about a typical AOD pump, most of them are double diaphragm, so one side is in suction mode while the other is in discharge mode.

For the purposes of sizing, you're likely best off to use the same procedure as for a centrifugal, but also look at the suction capability, and also verify that your air supply (pressure and volume) is sufficient to do what you want. Especially when you get up to the 2 and 3" pumps, different brands can have very different air consumptions, which can be a rude surprise.

 

[ione]

You have calculate the total discharge head at your application rated flow. Then from charts supplied by the pump’s manufacturer you can find air supply pressure and air supply flow rate necessary for your application (you’ve now processed your discharge side). Then check whether your pump has got enough suction lift capability (the supplier must provide this info) to match your suction side requirements.