Booster Pump for Injection pumps

[maintennance]

There are three water injection pumps ( 2 running + 1 standby) for the oil well and each injection pump suction is fed by an independent booster pump to take care of NPSH requirements.Should there be any reason to have independent booster pump for each injection pump? Why not to have a common suction header for these injection pumps and connect the booster pumps discharge to this suction header. Atleast we can run any two booster pumps to have a flexibility . Similarly in the P&ID it is shown that the MCF line to the suction tank for both the booster pump and the Injection pump are joining in a same line . Is this correct because both pressures are different . Thanks to clarify.

 

[micalbrch]

Independent booster pumps give you independent pump units (injection plus booster pump). That's an advantage in terms of availability.

A common suction header for all three injection pumps and two (one duty and one stand-by) booster pumps only makes sense, if only one injection pump is running at a time. When two are running how do you want to control the flow of the booster pumps? One booster pump will be too small for two injection pumps. Two booster pumps must know how to divide the flow in the common header. So, that will not work.

 

[ione]

micalbrch,

A common booster pump handled with a VFD or a proportional control valve fitted in the system could make it work. On the other hand I concur with you that two booster pumps add flexibility to the whole system.

 

[micalbrch]

ione,

Can you control the flow split to the two injection pumps with one VFD controlled booster pump? I do not think that will work well. I assume the injection pumps are triplex piston/plunger PD pumps. If I'm right, that will be tricky with one booster pump.

 

[JJPellin]

People familiar with this process may know automatically what types of pumps are employed for these services. But, for the rest of us, it would help if you told us what type of pump you are using for the boosters and injection pumps. If the booster pumps are centrifugal, they can be run in parallel into a common header if you take appropriate precautions. And the ability to run any two booster pumps with any two injection pumps would add operating flexibility and system availability.

Johnny Pellin

 

[SNORGY]

I have had experience with systems like this, probably smaller in horsepower but otherwise the same in theory. One centrifugal booster pump per plunger injector pump would be what I would favour, especially if the booster is not on a VFD and has a somewhat flat characteristic to a little bit past BEP. The idea is to keep the suction header pressurized with enough water flow available to meet the volumetric demand of the injector throughout its speed range. The "plug and chug" mentality would be to somewhat oversize the booster so that it could be expected to run a bit left of BEP most of the time.

Two centrifugal boosters in parallel in front of the plungers is another way to flatten out the effective characteristic, but one needs to pay attention to control and shutdown logic to deal with a case in which one of the boosters cuts out.

If indeed the injectors are plunger type, then they might be equipped with belt drive speed reduction between the electric motor and the pump. It takes time - and a certain number of revolutions - to ramp down to zero speed when such a pump trips out. Meanwhile, it continues to demand water from the booster / suction header. Regardless of what else is done, I would be inclined to put a time delay on the booster pump trip for the case when the plunger pump trips; otherwise, the booster will stop first and the plunger will continue to ramp down and create a significant pressure reduction in its suction piping. The subsequent start-up of the booster pump can hammer the system and do serious damage to plungers and cranks. It's better to dead-head the booster for a few seconds before tripping it to avoid this.

Regards,

SNORGY.

 

[maintennance]

Thanks to all for your replies. The injection pumps are Between bearing centrifugal pumps and the Booster pumps are Overhanging pumps. Will provide all other details once I go to office later today.

 

[micalbrch]

O.k. Then I'm out.

 

[maintennance]

We have three (3x50%) Water injection pumps & each pumps are connected with individual booster pump (3X50%) . Booster pump A supplies water to Injection pump A. Similarly booster pump B supplies water to injection pump B. And similarly C.



The injection pumps are multi stage centrifugal API BB5 type pumps. They are rated for a flow rate of 300 m3/hr and differential head of 2600 m approx and 3 X 50% in quantity. Each injection pump is provided with booster pumps having 300 m3/hr capacity and 100 m differential head.



There is no common suction manifold provided for the injection pumps. If a common suction manifold is provided then any two booster pumps can be operated for any two injection pumps operation. Is any flow and pressure related problems foreseen or any operation / control related problems are foreseen in case of common suction manifold?


Thanks to clarify

 

[SNORGY]

Flow balancing will be something to look at if the piping is asymmetric between pumps.

Two boosters in parallel will achieve a somewhat flattened (relative to one pump) effective characteristic upstream of of the injectors, so the system might be a bit more stable in its response to fluctuations in individual injection pump flow demands.

Regards,

SNORGY.