Double diaphragm pump flow variations

[amar122amar]

We have a VFD (Variable frequency drive) motor driven double diaphragm pump, pumping liquid Teal (sp. gravity 0.829) with 30-37 kg/cm2(g) differential pressure. Pump normal capacity is 10 LPM and minimum capacity 0.15 LPM.
Pump primary capacity adjustment is with drive motor speed variation (automatic) and it has manual stroke adjustment also.
Pump operates well at capacity 2 LPM and above but huge flow variations are observed whilee operating at lower capacity (0.5 to 2 LPM). What may be the probable reasons?

 

[ashtree]

amar122amar

Some diaphragm pumps require a certain minimum head pressure to achieve accurate output. It maybe at 2LPM you have enough back pressure to get the accurate operation but below that maybe not.

I assume you have checked that the motor speed control is working correctly. I assume this is controlled remotely via a PLC and you select the required flow.

It might be worth checking the manual stroke adjustment as well. Set the pump at a particular speed and see if you can adjust the pump output with the stroke length adjustment. Check that the output is linear with stroke length.

I have also seen where diaphragm pumps dont like working at very low speeds. If you must operate at low speed you will probably need the stroke length as long as possible maybe even maximum.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[bimr]

The system head loss may be much lower when the pump is operating at low capacities.

 

[adammal44]

Do you have downstream flow stabilizer (Accumulator or Damper)? It could just be that your cycling of the diaphragms are coming further and further apart to achieve the lower flow rate (i.e., your volute volume in the diaphraphms is fixed and you're asking it for a lower flow, so it just cycles them more slowly, leading to spiky flow that averages out to your desired flow rate).

 

[LittleInch]

I find it difficult to believe you can even get a 5:1 turndown with a pd type pump. Any more than that and you're struggling to get the torque required to pump anything. Hence the pump will jump and stall.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[ashtree]

Littleinch

It will depend on the type of VFD. When used on a PD type application a constant torque drive should be selected, but good point otherwise and i had missed that. 0.15-10LPM is a big turndown for control by motor speed alone. Impossible i would suggest as acceptable motor performance would normally only allow operation as low as about 10 hz. So unless the stroke length is also being adjusted for those very low flows are going to be difficult to achieve.
If that sort of range is required then a different configuration made of multiple pumps is probably required.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[LittleInch]

I had missed the fact that some of the turndown (an unknown amount) is mechanical. However at low speed I still can't see how the torque is "constant".

The other reason at low flow is presumably the stroke length is the smallest you can get and hence leakage or loss of sealing of the outlet check valves suddenly becomes a much bigger volume of the amount pumped.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[micalbrch]

You can easily get a 1:10 or even 1:15 turndown out of a PD pump with a frequency drive. It is just a question of proper selection of motor and VFD.

Back to the origin question: What we are missing is the stroke rate at low speed. The OP wrote it is a double diaphragm pump and I think he is talking about a single acting pump with one pump head with a double diaphragm inside. If I'm right and if the pump operates at rather low speed (let's say 30 strokes/minute or less), the effect of a single acting pump can be seen in the accuracy of the pump. The pump performs one discharge stroke (--> flow) and then one complete suction stroke (--> no flow). At low pump speed this performace has a significant affect upon the measurable flow. A discharge side pulsation dampener might help a bit here but will not eliminate the problem. My recommendation for low flow operation here is to run the pump with the highest possible speed but adjust the stroke length manually as much as possible.

 

[amar122amar]

Thank u very much for your response.

 

[georgeverghese]

Another solution to reduce flow - pressure pulsations at low speed would be to install a recycle line that keeps the pump speed at above a min speed - a min speed recyle line if you like.

So you'd have a a normal capacity resetting speed control to control forward flow capacity

And this min speed recycle line with throttle valve that leads back to the source tank.

A suitably sized discharge pulsation dampener will also help as suggested.

 

[moltenmetal]

George: by so doing, you're turning the pump from a metering pump into a pressure source. In that case, you just run it against a back-pressure regulator at any speed you want and use a control valve and flowmeter to control the flow.