Suction Pipe Smaller Than Suction Nozzle on PD Pump

[Fooooks]

All,
I have a positive displacement pump I am considering to unload rail cars with. The area is tight and we have an 8" pipeline that we can utilize for our rail track. The 8" pump we first considered will only give us approximately 1150 GPM, we are looking for more. The 10"x10" positive displacement Viking RS4324A that we were quoted can get approximately 1550 GPM @ 105 psi with an NPSHR of 22 ft. This pump would be side suction/side discharge.

We will have a VFD on the pump that will slow the pump down when the rail cars get low on product (lower NPSHA). We would also be able to install about 10 ft of 10" suction piping before reducing down to the rest of the 8" suction pipe. I ran some calculations on this and the velocity entering the pump is approximately 6.5 ft/s, the worst case NPSHA (when the rail car is near-empty) is 22.65 ft'.

From the numbers ran in AFT Fathom, it seems like this pump will work in this scenario, but I have never used a suction pipe smaller than the suction on the pump. The pump vendor said this would not be a problem, but I'd like to double check where I can. Are there any concerns I should take into consideration that I may have overlooked?

I am in the USA with an liquid storage facility, the product is vegetable oil @ ambient temperature.

Any additional information would be great. Thanks,

EDIT: we are unloading from the bottom of the rail car with 4" hoses going into the 8" suction line.

 

[KoachCSR]

When you are piping up to the pump, be sure to utilize an eccentric reducer such that the "flat" part of the reducer is on the top - this will prevent an air pocket from forming in the pipe, which may lead to air entrained issues.

 

[Fooooks]

Thanks, KoachCSR, I had not thought of that, but it makes sense. I do imagine this pump will see some harsh operational tendencies. It may pull air from an empty tank if we do not close a valve in time or other potentially dangerous operations. We have smaller pumps on the rail track, but they are the 8" pumps and they have been working well in the application already.

 

[bingopin]

You did not mention the length of the suction line. You might want to google "acceleration head".

 

[Fooooks]

Bingopin,
The length would vary with relation to where the rail cars are spotted, the closes railcar would be within 40 feet, but each additional railcar it pulls from would be approximately 75 feet away.

I am unsure if acceleration head is calculated in AFT Fathom, which is the software I used to model this scenario. If it becomes an issue, we would be able to slow the pump down with the VFD automatically if is loses pressure on the suction of the pump.

 

[bimr]

One would doubt that you will be able to get 30-40 ft/sec through a 4-Inch hose. Hoses do not have the same hydraulic characteristics as pipe. You may destroy the hose at that velocity. You may also collapse the hose with the pump suction.

 

[Fooooks]

bimr,
Maybe I did not clarify, but if we are unloading only from one hose, then we would not try to achieve 1550 GPM. That would only be if we have, say, 4 railcars or more unloading simultaneously.

I believe we would be closer to 350-400 GPM through a 4" hose at any given time (9-11 ft/s). This is much more realistic than 30-40 ft/s.

 

[LittleInch]

The pump would appear to be a gear pump which helps as there is only a small amount of pulsation, at least compared to a piston pump.

I had a quick look and from what I can see the NPSHR of the pump at max rpm is only 13.5 ft ( see page 15 of linked doc ). 22ft for a PD pump looked high

http://vp.salesmrc.com/pdfs/Sec_430.pdf

Just try and maximise the length of the larger size of your inlet pipe (can you make that 10" length 12"?) but if you've calculated the NPSHA and you're ok then just monitor the pump for strange noises and pulsation of flow which is indicating cavitation and vacuum / vapours forming.

At higher rates you may well find that the railcars offload at different rates.

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

 

[EdStainless]

Are you going to build the interface so that instead of selecting a speed the operator will simply enter how may 4" lines are in use?
You need a pressure sensor at near the pump inlet tied to the VFD, to slow down or shut down when the suction is starved.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, Plymouth Tube

 

[Fooooks]

EdStainless,
Yes the VFD will have a pressure sensor on the inlet of the pump, which we have a company come out and set up for us. They have done the other pumps on other railcar applications. I do not know what the setpoints are for the pressure at the inlet. I do not know if they have an option of inputting how many 4" lines are in use, but it is a good idea that I will explore.

LittleInch,
Yes it is a gear pump, and I had received that NPSHr from the vendor, maybe they accounted for a safety factor or a recommended NPSHr and not the actual NPSHr from the data you linked.
We are also considering the HXL10E Sliding Vane pump from Blackmer, which we have used in other applications for rail unloading. The NPSHr from the vendor on this is approximately 8.7 ft.

We could also increase the suction side to 12" if needed to slow down the velocity more. The scenario I ran in AFT Fathom shows that the closest railcar would actually unload up to 525 GPM (13.2 ft/s through the 4" hose). Does anyone foresee this becoming an issue?

Thanks for the confirmations in this thread, all. I hope this is helpful to anyone else who may run into a similar problem.

 

[Fooooks]

LittleInch,

Just to update you. Viking updated their document recently, and the pump quoted is the RS in Table 5, pg. 12 of the attached file.

 

[LittleInch]

Fair enough.

Maybe time to choose a different pump....

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

 

[Fooooks]

Agreed. The other pump manufacturer we are familiar with here for Positive Displacement pumps is Blackmer, their sliding vane pump HXL10E has NPSH of around 8-9 the vendor tells me. The only problem is the price point is much higher. We could also try a screw pump, which would be cheaper and not use a gear reducer. But we only have one in the terminal at the moment so making the switch would be difficult at the moment.

Thanks for all the responses.

 

[MortenA]

progressive cavity pumps also have many advantages in this kind of application especially since they should be well suited for food services (if it is for food). Seepex pumps are one manufacturer that i know.

Best regards, Morten