Positive displacement pump vacuum suction lift and pump life

[karmavatar]

Hello,
The application is a positive displacement (PD) pump supplying lube oil (mineral oil/turbine oil vg 32) to compressor bearings at 2 bars/30 psig. The reservoir containing oil (atmospheric tank) is built into the baseplate/skid and API 614 allows to have reservoir built in baseplate. Due to height of the baseplate (only 18~24 inches) and limited space inside the reservoir, a submerged vertical pump is not used. The PD horizontal pump is located on top the baseplate/reservoir. The suction piping (as short as it can be routed) runs from side of the reservoir and to the inlet of the pump. When oil is filled up in the reservoir, some of the suction piping gets partially filled with oil up to same level as reservoir. When PD pump is started, it has no issue pulling vacuum/air and in 3~4 secs the oil is suction lifted (through less than 1.5 ft of empty pipe) and pump discharges/runs normally. NPSHa, NPSHr/NIPR calculation is not the problem. Question is not about centrifugal pumps and not about self priming on centrifugal pumps.

However, for those first few seconds of running the positive displacement pump dry (gear,rotors including mechanical seals), is there any literature evidence that this kind of operation decreases the life of PD pump?
Thanks

 

[JJPellin]

I am not aware of any published literature on this specific subject. However, in my plant, we have at least a dozen oil systems of this basic configuration. Some of them have the oil pump sitting on the top of the tank, as you describe. For others, the oil pump is shaft driven, located some distance above the top of the tank. In general, the pumps have very good reliability. I would prefer to have the pumps located below the liquid level in the tank. But, the reliability is still quite good.

Your pump should not have to draw liquid up through 1.5 feet of empty pipe every time it is started. Once the system is primed and flooded, it should stay flooded unless you have some air leaks. Does your pump have a check valve in the suction line or a foot valve in the tank to prevent the line from draining back? This should not be required for the relatively small lift of your system. If it does have a suction check valve or foot valve, I will offer one point of caution. If the drive motor is ever replaced or rewired, be very certain you get the rotation correct. If the pump is started up running backwards with a suction line check valve, you may burst the suction line. These pumps tend to pump very well in reverse and there would generally not be a relief valve in the suction line. We have blown apart at least one suction line because of this mistake.

I assume your pump is motor driven. I assume your pump is a screw pump or gear pump. Please correct me if my assumptions are wrong.

Johnny Pellin

 

[karmavatar]

Yes,your assumptions are correct. It is either sometimes a screw or gear pump and electric motor driven. Yes,reverse rotation against a check valve due to motor rewiring is a reality. But, in this system there are no suction check valves on pipeline to keep the suction line primed all the time. It is difficult to maintenance a foot check valve if installed inside the baseplate built-in reservoir so foot check valve is not present. More or less it is initial cost savings approach not to put the check valves/foot check valves in with relatively small suction lift. But, if pump life is compromised then it is not cost savings anymore.
Thank you for response

 

[TugboatEng]

The load on the pump is very minimal when it's priming so the lubrication requirements are also minimal. The residual oil from the last run is enough to keep it from being damaged while priming.

 

[SwinnyGG]

You aren't going to find 'literature' - i.e. a research paper - telling you if a specific pump can survive a few seconds of unlubricated no-load running.

But whoever makes the pump can very likely tell you.