Booster Pump to PD Pump

[Zyzz123]

I just wanted to double check something. If I have a booster pump (centrifigul) upstream of my pd pump, and the booster pump has a lower discharge flow rate compared to the discharge rate of my pd pump it will create cavitation and vibrations. Is it true that the bigger the difference the stronger the vibrations?

 

[micalbrch]

What PD pump do you have, a piston pump? You will not only create vibrations. The higher flow of the PD pump might even destroy the centrifugal pump. I suppose this system is not in Operation yet because the PD pump can not have more flow than the booster pump provides but the problem is: It will try - and that will destroy your centrifugal pump.

 

[SNORGY]

Yes, the greater the disparity, the worse will be the problem.

Th PD will demand flow from the centrifugal. The effect on an undersized centrifugal will be that it will be forced rightwards along its head-capacity curve in an effort to provide the flow demanded by the PD. This will occur at lower differential head (hence delivery pressure) from the centrifugal to the PD. if the equilibrium point is far to the right along the centrifugal's head-capacity curve, the suction pressure to the PD can be reduced to the point where cavitation will occur. Indeed, cavitation might also occur in the centrifugal, but possibly of more concern will be its sustained operation at run-out conditions that, in addition to vibration, might overload the motor.

 

[Zyzz123]

Thanks for the quick reply,
Just a heads up, I'm kinda new to the industry so my next question might seem a bit dumb.

micalbrah: The pd pump is a piston pump. It's being in use right now and there's lots of problems in the building.

Snorgy:If my centrifigul pump is set at a certain flow rate, how will it know it needs to increase the flowrate?

 

[MikeHalloran]

You have to study the set of curves provided for the particular centrifugal pump to begin to understand what's going on, because centrifugal pumps' flow rate is not fixed relative to its rpm, like a piston pump. Even if the centrifugal has a nameplate that says 'X gpm', it will almost never run at that exact flowrate.

The easiest condition check you can do is install a compound vacuum/pressure gage between the discharge of the centrifugal pump and the suction of the piston pump. If the pressure stays above the NPSHR of the piston pump at all times, you don't have a problem.

If the inter-pump pressure goes below the NPSHR of the piston pump, then you need a bigger or faster or rebuilt supercharge pump, or you need to clear the clog from its inlet.

Also note that centrifugal pump curves are typically provided for water. If you're pumping something else, you need to correct the curves or the math. Check the detailed data sheets for the centrifugal pump, and look up 'affinity laws'.







Mike Halloran
Pembroke Pines, FL, USA

 

[SNORGY]

The systems I have seen - particularly if there is also a filter or strainer between the booster pump and the piston pump - usually start having trouble at a piston pump speed of about 250 RPM coincident with a booster pump delivery pressure, measured close to the discharge of the booster pump, of 200 kPag. If the fluid is hot or has a vapor pressure higher than water at a given pumping temperature (depending on a few other things like Vg/Vf ratio), the problem is worsened.

 

[SNORGY]

Also, if the booster pump is "set to a certain flow rate", then does that mean that there is a flow control valve or loop in the system? If so, where is it? Ordinarily you would control the flow from the PD pump using a VFD or recycle back to suction and let the booster just ride wherever it needs to on its characteristic in order to keep the intermediate piping pressure above that corresponding to the PD NPSHR.

With respect to your question, the centrifugal pump will only operate at a given instant in time on one point along its characteristic head-capacity curve, and that point will be where the system head curve intersects the pump characteristic. Thus, when the PD pump begins to demand more flow than can be delivered by the centrifugal pump corresponding to that point, the intermediate pressure falls and the centrifugal pump will automatically deliver more flow at lower head - in other words, the operating point shifts to the right to a new intersection point. The reverse is true when the PD pump demands less flow - the centrifugal pump operating point shifts to the left to an intersection corresponding to less flow at higher head. It's actually the hydraulic resistance in the piping system between the pumps that sets the flow from the booster.

Hence my question about flow control. Be careful how this is set up.

 

[micalbrch]

The piston pump can not get too much flow from the booster pump. The booster pump must be sized for the max. flow of the PD pump plus - and that is very important - an adder which compensates the non-uniform flow of the piston pump. If your piston pump is a triplex pump (three pistons) the booster pump (as a rule of thumb) must have 1/3 higher flow than the piston pump.

 

[Zyzz123]

Snorgy: Why is it that having a filter along with rpm greater than 250rpm creates problems? Right now both these criteras are met.

 

[SNORGY]

Zyzz123,

Again, *my experience has been* that plunger pumps like to see about 25 psig at their suction flange; otherwise, vapor pressure, line losses and acceleration head *I have observed* contribute to the onset of cavitation due to the separation of the plunger face from the incoming fluid on the plunger's receeding stroke. You also need to make sure the pulsation dampeners are set up and tuned correctly.

Your experience might be better than mine. I am only relating what I have seen.

It does cause damage over time, such as was the case several years ago when I was called to the field to look at a triplex in water service that had blown apart, shattered all three plungers and expelled the crank from the casing.