PIPEING PROBLEM, NEED HELP 3

[NTME76249]

I am need some help trying to figure if this setup is going to work. i am not as skilled in fluid dynamics as i should be. any help i can get would be welcome. thanks

 

[MiketheEngineer]

What are you trying to accomplish??

 

[NTME76249]

Sorry
I meant to post some questions with it.

How does the oil on top of the water effect the head?
The riser pipe is under the top oil level, how does this rise and drop effect the suction need by the pump other than normal flow loss?
They have been burning up the pump and not sure what they should do.
Would a 6 in Line help?
lowing the Riser pipe?

Thanks
Michael

 

[pennpiper]

Please explain the need and purpose for the 20'-6" high point in the suction line

 

[77JQX]

If they've been burning up the pump, I think they have air trapped at the high point of the riser, resulting in lower NPSHA. Is there any way to check that?

As to your questions:

How does the oil on top of the water effect the head? The oil adds to head according to (ft of oil)*(S.G. of oil).

The riser pipe is under the top oil level, how does this rise and drop effect the suction need by the pump other than normal flow loss? As long as the riser remains fully flodded with water, then normal friction is the only loss.

They have been burning up the pump and not sure what they should do. Would a 6 in Line help? Some, but not much - A 6" line would gain you about 2 feet of NPSHA.

lowing the Riser pipe? If it's kept flooded, only by the friction due to length of suction pipe removed.

 

[Duwe6]

Try adding a vent to the top-of-pipe in the vert loop. Keep the vent tall enough to not allow slop to be forced out from the tank headpressure. Since it is just an air vent, it can be 3/8" tubing and swagelock fittings.

 

[curtis2004]

NTME76249

I would join pennpiper to ask why you need this high point?
Another question: How pump operates? Constantly, or intermitent? How level of fluid changes inside tank during operation?

Regards,

Curtis

 

[NTME76249]

I talked to one of our field guys and got some clarification on the hole thing. This is a Gun Barrel with a riser to set the interface level between the water and oil. The 200 GPM is the inlet. the problem has been is that the outlet is not letting flowing quick enough such that the water level starts coming out the oil out let. I Believe the Water outlet is big enough they just need a bigger interface. they said they are currently running about a 2" interface. if the lower the riser down to create a 2 ft interface than they should be OK, just not sure how to calculate the effect on the flow rate out of the tank that the riser will have. i calculated at the water outlet to have a 1200 gpm flow rate

 

[zdas04]

This device is called a "gun barrel", it is an atmospheric process vessel to separate oil and water. The drawing is missing the oil outlet which is a skim line at 24 ft. The height of the weir piping is based on the hydrostatic head of the oil plus the water and it is always some distance above the design gas/water interface (if it is too high, everything goes to oil, if it is too low then everything goes to water). It is based on a set oil SG and a set fluid temp (we often don't manage the fluid temperature as well as the designers had hoped we would)

In normal conditions, the oil is at the oil outlet and the left hand leg of the weir piping is always full to the overflow. If a liter of liquid comes in, then a liter of liquid leaves out some combination of the oil outlet and the weir piping. The weir has to be vented to prevent siphoning the whole tank into the water system--the vent either just isn't shown or the gun barrel was designed to fail.

With this design, I'd figure on the centrifugal pump lasting a week or so. The only way these things work (and they can work amazingly well) is if both the oil and water go into atmospheric tanks (and you have the pump from them controlled with on/off floats). Trying to pump out of the weir piping is ALWAYS a disaster.

David

 

[BigInch]

Could be that your pump draws too much flowrate from the downstream 4" piping when and the feed stream coming across the gooseneck can't keep up, breaks and thereafter flow does not easily recover.

Don't suppose you have a suction gauge reading?

Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[btrueblood]

Zdas,

Can you explain the purpose of the riser?

 

[zdas04]

The riser allows you to maintain a volume in the gun barrel. As I said, the vessel is designed to be exactly full to the oil outlet and the top of the weir riser all the time. If the riser height is too low, then the oil surface never gets to the oil outlet and everything coming in goes out the water side. If the riser is too high then you don't have enough head to reach the top of the weir and everything goes out the oil.

If the riser is not there, you have to use an oil/water interface level controller and a dump valve. This works better with light oil than heavy, but even in light oil it is still at risk from leaking or hung dump valves.

There are several ways to skin this cat, I've attached an extract from my 5-day course to show some considerations. On the second page there is a picture of a couple of kinds of gun barrel. They do the same thing--the external weir like we're talking about here is much less expensive, the internal weir is much more forgiving.

This "simple" process is really quite complex. I always design the weir with multiple overflow heights (with block valves on them) so that if the SG of the oil changes I can shift the weir height.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

"It is always a poor idea to ask your Bridge Club for medical advice or a collection of geek engineers for legal advice"

 

[NTME76249]

THANKS TO EVERY ONE FOR THEIR ADVICE.
HERE IS WHAT I CAME UP WITH.

 

[Duwe6]

And a star for the man from Muleshoe! I learned a lot. Thanx.

 

[zdas04]

Been a while since I saw an analysis laid out like that. What I always ask students to look at is a final reality check.

Your inlet pipe is the same size as the water outlet pipe. Inlet velocity (disregarding the Cv as a first approximation) is q/A (the velocity equations you are using are generally used to determine a flow rate from pressures, you know the flow rate) or 5.57 ft/sec. That is the combined stream flow rate. Part goes into the water pipe and part goes into the oil pipe. Logic says that the water outlet velocity must be less than 5.57 ft/sec (since it is a lower mass flow rate), not 34 ft/sec.

I can't comprehend why you stated oil density at 140F and water density at 52.8F. The inlet stream is mixed and very likely at the same temperature. It is crucial have the two densities at the same temperature. I'll assume that the tank is heated and that everything is at 140F, so the density of the water (assuming SG=1.0 and density at 60F is 62.4 lbm/ft^3) is 61.34 lbm/ft^3.

Your weir height is pretty wrong. It is not a function of velocity, it is a function of hydrostatic pressure. Weir height is always greater than the height of the oil/water interface and less than the oil outlet. The height is:

h_weir=([ρ]_water*h_water+[ρ]_oil*h_oil)/[ρ]_water

Which works out to 23.55 ft. I will generally try to get a bit more definition and in this problem I'd plan on the interface being at 15 ft instead of 20 ft. That works out to a weir height of 22.96 ft which is only 7 inches lower, but that can make a big difference.

David

 

[BigInch]

It would seem that many think these systems are dynamic, but in actuality they are semi-static. It can take a lot of time, relatively speaking, for systems that are driven by slight imbalances in static pressures to stop oscilating and reestablish themselves at new flowrates whenever their usual trickle flow patterns are interrupted by valve openings and highly dynamic pumps that bring on rapidly changing fluid levels.

Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso

 

[zdas04]

The flow within the tank itself qualifies as "creep flow" with velocities on the order of 0.002 ft/sec. The outlet has to go go a tank so that you can accumulate enough fluid to pump.

David

 

[BigInch]

Ya. If that gooseneck dropped down into an "accumulator" tank, then going on to the pump, it might work like he expects .. given that those critical elevations are readjusted. Or putting a high vent on that gooseneck would isolate the separation tank enough from the pump dynamics, but then there wouldn't be enough water in the remaining 4" going to the pump to bother putting a pump there.

Only put off until tomorrow what you are willing to die having left undone. - Pablo Picasso