Pump Suction Uptake Vortexing 2

[PuffJr]

I have some pumps that are under-performing, and my local Viking guy has suggested that this is the result of vortexing inside the tank I am pumping out. We are seeing about one-half the expected flow. (I don't know how the pumps sound; they pumps are at a remote location, and I am getting somewhat poor support from my guys out there.)

To facilitate more complete emptying of the tanks, my suction nozzles turn down toward the floor of the tank. The gap between the opening of the uptake and the tank floor is such that the cylindrical surface traced by projecting the pipe inside circumference onto the floor equals the inside cross-section of the pipe. My Viking guy says that the gap should be a full diameter to avoid vortexing. Wouldn't vortex formation be governed by flow, viscosity, and specific gravity, such that the critical gap would be different form fluid to fluid, and from pump to pump?

Any other ideas on what might be causing this problem?

(The pumps are Viking internal gear pumps, and the media is motor oil and motor oil components.)

 

[dcasto]

Install a vortex breaker in the tank.

http://www.maverickinspection.com/rvprocessvess.html#image22

Here is a picture of a simple one. I've even had success with one that was inserted through the flange and held in place like an orifice paddle with the "X" welded to the paddle and inserted in the nozzle of the tank.

 

[BigInch]

plus... keep more fluid above the inlet level and/or extend the inlet farther away from the tank edges (and bottom if you can).

http://virtualpipeline.spaces.msn.com

 

[dickon17]

Viking guy is wrong to say one full pipe diameter gap. Your calculation for area equivalence is correct. This usually results in a gap of about 1/3 pipe diameter. What is intake velocity into suction pipe? If pump flow only reduces when tank gets close to empty then this might be your problem. However, if this is a general problem then I would look to NPSH as a first guess. The fluid viscosity might just be too much for the speed of the pumps.

 

[rzrbk]

Is this a new installation, or have these pumps been running for a while? Did they ever pump the expected flow?

Does flow improve if level is raised?

Do pumps run in parallel or one at a time? Any chance for reverse flow back to tank?

There are too many causes for low flow to diagnose without more information.

 

[PuffJr]

This is a new install.

We have not observed that tank level has a dramatic impact on flow.

Each tank has one dedicated pump.

There is a path for fluid to return to the tank, through a PRV. I doubt that this is happening, because that would require a full stoppage, and I am certain that this does not exist.

 

[dickon17]

In this type of application suction pipe velocities are usually low and cavitation happens before vortexing. Get the Viking guy to calculate NPSH available/required. You may need a bigger diameter suction pipe.

 

[rzrbk]

In addition to NPSH, I would also suggest several other checks including:
-motor speed and gear ratio against flow curve (measure actual shaft speed if possible)
-leakage thru piping psv back to tank
-setting of integral pump psv (sometimes factory set to minimum)
-calibrate/check method of flow measurement
-suction/discharge pressures against predicted

 

[Artisi]

To say it is vortexing without any evidence of the fact sounds like crystal ball gazing. you need to properly analyse what is happening and if you aren't getting much help from site you have a problem on your hands - rzrbk and others have given you some good pointers which you need to address.