gear pump - strange happenings

[scorman]

first post so I won't try to wander too much

I am attempting to use a hydraulic pump as a gear transfer pump for moving low volumes of veggie oil.

I have had difficulty before with Procon vane pump, Webster HHO gear pump ...especially when temps get low and oil thickens and to the point where all flow stops.

I want to move 55gal drums of WVO from garage to basement at lower level ...the run is 65 feet.

I procured a brand new Parker PGP505 hydraulic pump whose specs are 0.37 in^3/rev. Max operating rotation is 4000rpm up to 3600psi requiring KW power. However, I am using in an open ended environment, so fractional hp is needed. So I got 3/8" thickwalled clear vinyl tubing for output port and two short (12ft) input port configurations ..one is clear poly 3/8"ID and the other is 5/8"ID black poly.

I did several experimental runs, but air bubbles in the output line is the problem.

Initial tests were with the constricted 3/8 feedline.
First, I used a 1725rpm motor and got foam ..should have gotten 2.75 gal/min?

Next I used a geared motor at 230rpm and got steady flow but
air bubbles about every three inches ...oil temp was around 65F. Took about 2 3/4hrs to empty 55gal drum.... about 1/3 gal/min

Next time oil was hotter at around 90F+ and bubbles decreased to about every foot ...took 135min to empty tank and this correlates closely to the rated volume about 0.4 gal/min.

Note: there are no visible bubbles in input line.

Went back to guidelines and they say input velocity stream should be 1/2 output, so I switched to 5/8" input.
Oil temp at about 80F and same result as last try.

BTW, the gear motor runs at 60watts, so 2+ hrs costs about
$0.04 electric ... I was hoping for an rpm that delivers 2gal/min, but that ain't happening. Also, need to run at <60F in future.

My conclusion:
The shaft seals on these pumps cannot handle the internal vacuum generated with oil the consistency of pancake syrup. I have ruled out cavitation and there cannot be enough dissolved air to give the volume of bubbles present.

Any other viable mechanisms that I have missed?
Appreciate any feedback and I can elaborate more if above description is inadequate.
Using 1" input line is NOT an option.

I do have a DC servo motor that can be a variable speed drive to hand tailor the rpm to the given viscosity of the WVO on any particular day/temp, but it seems like overkill.

Stew Corman from sunny Endicott

 

[hydtools]

Put the pump very close to the drums. Put the pump right on the top of the drum. Use the shortest length and largest diamter hose you can. You might see if you can get the pump with a high pressure seal.

Increasing the ouput hose diameter will reduce the pumping load.

Ted

 

[btrueblood]

I'd think you could drop the pump right into veg. oil, no worries about conductivity of the oil is there? Just a wild thought.

 

[FMJalink]

Postscorman,
Did you make a pump calculation of the system? You will probably find that you have a pressure substantially below atmospheric at a viscosity above a specific value (temperature below a specific value). The pump will probably have a single mechanical seal, which is not able to prevent ingress of air at a substantial differential pressure over the seal. This will explain your air bubbles as you already mentioned.

What can you do against it:
• Reduce the suction line length.
• Increase the suction pressure.
• Heat up the oil.
• Reduce the flow as required by the viscosity.
• Add a double mechanical seal with e.g. your oil as sealant.

 

[scorman]

Update:
Thanks again for all the contributions to this thread.

I had the opportunity to modify the hardware for a slightly different application.... as a WVO metering pump to the centrifuge.

"Put the pump very close to the drums. Put the pump right on the top of the drum. Use the shortest length and largest diameter hose you can. "

I created a rigid dip stick out of 3/4" PVC pipe, mounted the pump on top of the barrel, connected with an 18" clear vinyl with an ID of 7/8" as the input side. The exit side is 3ft of 1/4" ID clear poly tubing to the centrifuge.

The motor is now a DC servo which at 15 volts/2amps = 330rpm.
Drum was stored inside garage at ambient (70F), so veggie oil was 15F+ cooler than previous attempts, but viscosity varies drum to drum so remains an unknown.

Conclusion: no bubbles in exit line and throughput = 30gal/hr

At a later time I many recreate the original setup and see how much the rpm can be boosted until air bubbles form.

Followup question:
Why place the pump on top of barrel other than to shorten the input line to the minimum? If it was at the floor level, the added head of oil would be positive pressure on the suction side. As the barrel empties, the pump when mounted at the top now has to suck harder adding to the vacuum.

Stew Corman from sunny Endicott

 

[micalbrch]

If you can place the pump at the floor level with a drum outlet at the same level - perfect. That would be the best. But drums usually have their opening on top. And with an opening on top and with the pump at the floor level you will not only have a longer suction line but - and that is the problem - a suction line like a syphon (up and down). Air will accumulate here and can cause dry running.

 

[hydtools]

Suggesting you put the pump on top of the barrel assumed you could not 1)submerge the motor/pump or 2) you have no bottom tap in the barrel. That gave you the shortest suction line, the other options not being available. And you can easily move the motor/pump to other barrels.

The increased lift with lower and lower levels is within the capabilities of the pump and atmospheric pressure. If you increase flow rate until you get bubbles, increase the size of the suction piping to stop the ingestion of air.

Ted