Using Positive Displacement Pumps in Series

Question

Using Positive Displacement Pumps in Series

Wendell H (Chemical)

(OP)

24 Mar 21 15:44

Can anyone help me with this problem.

I have an existing RM storage tank system 700 feet from the end user. 3 inch CS transfer line with PD pump at tank.
The raw material is an oil with a viscosity between 1400 and 2000 cP depending upon the temperature.
I have completed my calculations and ran in-plant simulations and am confident my calcs are correct.

The pressure drop needed for a 65 GPM transfer is about 25-27PSI or 175 to 200PSI total.

The existing system PD pump will deliver 120 PSI max at 200 GPM. I can slow the pump down and decrease the flow rate but I still have to overcome the piping pressure drop.

Can I put a second PD pump in the line say at 400 feet? The first pump will transfer the fluid to the second pump and we will see a pressure drop of about 100 PSI. The second pump will then see an inlet pressure of 20-25 psi and finish the transfer to the end user.

On the surface it seems logical but I know I'm missing something huge.

crshears (Electrical) · Answer 1 · 2021-03-24 15:49:26

Insulated pipe?

Insolation / wind chill effects?

Any other local factors that will alter the specific volume of the liquid between two pumps located that far apart?

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

Wendell H (Chemical) · Answer 2 · 2021-03-24 18:42:50

The pipe is either indoors (controlled environment) or traced and insulated. Elevated temperatures of the fluid do help but is not an option.

RVAmeche (Mechanical) · Answer 3 · 2021-03-24 19:05:15

I've never dealt with PD pumps in series but it seems like a potentially bad idea. If their flow rates aren't exactly the same you could create issues. I think I'd be more comfortable with a small buffer tank that fed into the second PD pump, but again that's without consulting a rep.

Is it not possible to retrofit/replace the existing pump so it can develop the required pressure? Is the piping system rated for the additional pressure you're calculating?

LittleInch (Petroleum) · Answer 4 · 2021-03-24 19:30:45

It's generally not a great plan to link two PD pumps together without either a break tank or setting the first one at a slightly higher flow and then allowing a relief valve to return some of the fluid to the tank. That means the second PD pump is never starved of product.

Allow an arrival pressure of at least 10 psi.

What's "RM storage"? not an abbreviation i'm familiar with.

But 1400 to 2000 CP is like treacle...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

crshears (Electrical) · Answer 5 · 2021-03-24 19:32:28

Another thought: install a VFD on the motor for the second pump in the series and have that pump's speed smoothly regulated so as to maintain an even but positive suction pressure on it.

Or even better, put a separate VFD on each pump, and dial an appropriate speed into the lower pump, with the upper slaved to its inlet pressure as described above. This would provide the entire system with operating flexibility in the event of unexpected viscosity changes.

I'd think this would work fine.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

RVAmeche (Mechanical) · Answer 6 · 2021-03-24 19:33:27

I'm guessing RM is raw material. I've dealt with PD pumps for un-loading totes/barrels of very vicious materials like that before, but it was a very short run to the storage tank and the heat traced piping helped significantly.

TiCl4 (Chemical) · Answer 7 · 2021-03-24 20:22:33

Wendell H,

What kind of PD pumps are these? If these are air driven diaphragm pumps, then there is no issue linking them in series as long as deadhead pressure of the series pumps is okay on the piping specification. If they are motor-driven pumps (gear, lobe), I would think twice before placing them in series. Small variations in rotation speed can lead to out-of-sync flow pulsations, which will really spike pressure. How much depends on the type of pump and level of de-sync, but it would need to be evaluated. The likelyhood of popping rupture discs/PRVs is high in this configuration.

Wendell H (Chemical) · Answer 8 · 2021-03-25 12:34:12

The oil we are pumping is a raw material in a storage tank (RM storage). The pumps are sliding vane pumps. I understand the issue with running pumps in series - the second pump becomes a maintenance nightmare - too high suction pressure, thrust bearing issues, etc. crshears thought of running on a VFD is something I am considering and slightly starving the second pump - also not a great option. I am trying to avoid replacing 700 ft of 3 inch sch 40 line.

crshears (Electrical) · Answer 9 · 2021-03-25 13:50:52

Quote:
crshears thought of running on a VFD is something I am considering and slightly starving the second pump - also not a great option.

Not quite; my thought was that if you keep the inlet pressure to the second pump positive enough you won't be starving it at all, matter of fact with some operating experience it should be quite feasible to compare the difference in power consumption between the two pumps and, within limits, tweak the positive inlet pressure of the second pump so as to achieve optimum balance.

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

SwinnyGG (Mechanical) · Answer 10 · 2021-03-25 14:24:29

Can you expose this material to air? If so, just put in a damper tank. Basically create a mid-run lift station.

Wendell H (Chemical) · Answer 11 · 2021-03-25 15:38:26

LittleInch - Good thoughts. The material is very close to treacle. That is a really good comparison.

SwinnyGG - I've used a damper tank in the past - worked well but I do not have the room for a tank. Very tight quarters.

crshears - thanks for the thought.

To all - thanks for the great input. My plan is to replace the pump to generate higher pressure at a slightly lower flow. Pipe should be fine at 170psi. Operations will have to live with 5-8 GPM less - should only add 3-4 minutes to the batch time. I wanted to save the expense of the new pump or piping replacement but two PD pumps in series seems like a maintenance problem child. Again, thank you all.

TiCl4 (Chemical) · Answer 12 · 2021-03-25 15:55:18

Wendell H - Can you post the OEM curve of the pump? I am surprised a vane pump would only be rated to 120 psig. I don't think a VFD would be helpful. If the sliding vane pump relies upon pumped fluid for lubricity, then starving the pump would also cause issues. Things I would explore:

Confirm max head of current pump.
Explore replacing the pump. If you are looking at buying a new pump anyways, just get an appropriate pump for the job. The flow and pressure you are seeking are comfortably in the range of gear and lobe pumps. I remember quoting out a 5 HP, 50 gpm 304 SS wetted parts gear pump a couple of years ago at $8-10k range. The design point on that one was something like 30 gpm @ 5,000 cP. Lobe pumps are another possibility in this flow/pressure range, but a bit more expensive than gear pumps (likely 30-50% more). You'll need more HP than what I required, but a new pump would sure beat replacing 700 ft of pipe.

Edit: It seems my comment was 17 minutes too late. :)

Come Join Us!

Posting Guidelines

Using Positive Displacement Pumps in Series