Crude Oil Terminal - Brine Pump Sizing

[PamelaS]

I am working on a crude oil terminal in salt caverns. The caverns are either filled with crude or with brine. The brine is pumped into the cavern to displace the crude. The brine water has a SG of 1.2 and the crude oil 0.86. The brine pumps are about 6000 ft from the cavern with about a 0.3 psi/100 ft pressure loss (equals ~20 psi line loss) to the cavern. The line that enters the cavern is 6500 ft (about another 20 psi line loss) down into the cavern. Assuming the cavern is full of crude oil, what pressure will you need to pump the brine? The brine is much denser then the crude, therefore, I assume once it is over the edge of the cavern, the brine weight overcomes the crude and not a lot of pressure is needed. Is my assumption correct? I am having a hard time putting this into my pump calculation program. Can I assume the cavern is at 0 psig at the entrance of the brine side? Also, the pressure the crude needs to come in would be line losses plus 6500ft*(74.88 lb/ft3-53.6 lb/ft3)/144in2/ft2=960 psig. Does anyone have experience on this topic?

 

[georgeverghese]

You've got to build up the line hydraulics model right up to the point where a known pressure value is being maintained - when the displaced crude comes up to the surface, does it feed a tank at atmospheric pressure ? The cavern is just an intermediate point in the model.

If so, any static pressure gain down the cavern with brine will be countered somewhat with static loss with crude coming up to the surface through the shaft annulus. You've got to develop the model to include the frictional head loss for the brine going down and the crude coming back up through the annulus(?).

This is an unsteady state operation. Crude coming up will initially be at a low flow, while that coming up later will be at a higher flow, due to the change in elevation of the brine - crude interface. So what is your controlling case flow; do you have some minimum flow demand at the start of this operation or some other scenario?

 

[LittleInch]

George is correct that you need to consider the whole system.

The other factor you need to look at is what pressure does the cavern need to remain at in order to stay as a cavern and not simply crumble. You need this as your start point as a pressure to not drop below or pressure above as the aim is to keep the pressure below that of the local hydrostatic pressure to make sure that if anything water comes in rather than oil out.

You need this before you go any further...

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

 

[PamelaS]

Thank you.
Assume crude fills the cavern (8MMBBL volume starting at 4500 ft under the surface to a depth 6500 ft in the ground). Later, the brine is pumped in at 285 kbpd to displace the crude out at the same rate with another suction to a crude pump close to the caverns. The pressure from the ponds is ATM and then the pressure at the ship where the crude is being pumped is assumed around ATM (a lot of unknowns right now). The crude pump near the cavern will pump crude 12 miles to the ship crude pumps. The pressure losses getting to the suction of the ship crude pumps is about 250 psig line losses.
Brine is only pumped in the cavern to lift out the crude. The pressure in the cavern stabilizes and the brine is at atmospheric pressure at ground level, but the pressure of the brine changes as the depth increases. The static head is negative since multiplied by -6500ft? That tells me if I get brine to the surface at 0 psig, then once to the edge, it will flow to bottom of cavern and lift the crude? I just have a lot of line losses to consider.
I noticed the crude pumps that inject into the brine when the cavern is full of brine are much greater in pressure then the brine pumps because of the density differences at 6500 ft.
I also noticed brine pumps were much lower in pressure and they commonly use HDPE pipe for the brine, which can only withstand about 250 psig.

 

[georgeverghese]

Would then suggest we develop the model to the suction of the surface crude ship transfer pumps to enable derivation of the head requirement for the brine pumps, with the brine - crude interface at 4500ft below the surface.

The suction press for these shipping crude transfer pumps would obviously be that corresponding to a flow of 285kbd.