Suction line Air-Elimination

[PieterStander]

I am decanting hydrocarbons from rail cars into storage, the suction line is 4", into a 4" flow meter with an air-eliminator downstream of the pump which has a feed to a downstream valve that will close when air is detected.

The problem I have is that my pumps are higher than the lowest part of the rail cars and when the rail car get empty my flow rate drops below the meter's designed min rate and then the presets stops the transaction.

Do any of you know of a way to eliminate air safely and efficiently in the suction line?

 

[BigInch]

It sounds like its doing what it is supposed to do.
You say, "When the rail cars get empty my flow rate drops below the minimum rate". Your meter should not be operating below the minimum rate, so as I see it from here, that's exactly what you want to happen.

So, the problem is not really that you have pumps higher than the rail car bottom, its that you get air in the suction line. But what do you expect to happen when the system is shut down ONLY when air is already in the suction line. That's kind of silly.

That shutdown method needs to be changed. Preferably it should be working on the fluid level of the railcar. Maybe that isn't practical, so what about setting up a stop based on suction pressure. Set a pressure switch in the suction to stop the pump when the pressure reaches a value that indicates the railcar has just run out.

I would also move the valve you have downstream of the pump to somewhere between the railcar and the air eliminator. The idea being that the air eliminator should only release, if something unusual happens, like air getting in the line by accident of some kind. It should not be used as a standard operating method to prime the pump each time air is "purposely" introduced in the line when the railcar finishes.

So, now run the pressure switch to the pump and the valve. The pump should stop and the valve should close at the same time and no air gets in the line. If at times a little air does get into the line, then the air eliminator should release it.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[PieterStander]

I might not have been clear enough, I need to drain the whole tank car, I go into "low flow stop" on the preset when there is still 10 000 liters left in the tank car.
I should have said "as the rail car gets emptier"

This is not the only solution I'm working on, one of the other things might be that the pumps is not speced high enough to pump the 15m head (I am still waiting for the pump's specs)

I did not design this depot and inherited it, I would have done a lot of things differently.

Another problem I have is that we are also decanting from road trucks, through the same suction pumps 9they just swing valves, the suction pumps from the rail is +/- 20m from the furthest coupling point, for the road this distance is 100m, in the opposite direction, so I have double the U bends T's in the line, here my max flow rate when, the receiving tanks are empty and truck is full is 150l/min, versus, the rail's 1000l/min.

http://www.dlm.co.za:

Over Pressure Protection; Flow, Level & Density Measurement; Automation and Control.

http://www.metermatic.co.za:

Electronic Presets; Additive Systems.

 

[BigInch]

That's a better description. Your problem is pretty assuredly not air in the pipeline. I think its low pressure vaporization and probably cavitation at the pump too.

The only solution I can see is to rip that out and start over.

Sounds like you have the wrong pump, its too high and too far away, with too many fittings in between. Kinda' hard to wave the magic wand and make that one work. I suppose its a PD pump too.

**********************
"The problem isn't working out the equation,
its finding the answer to the real question." BigInch

http://virtualpipeline.spaces.live.com/

 

[JJPellin]

You didn't mention why air needs to be eliminated or if your pump is centrifugal or positive displacement. We have a few services that are comparable. I will describe what we do and some of this may be applicable to your situation.

We have one chemical unloading facility that involves a chemical that is extremely valuable. We are required to remove all of the chemical from the truck. We built a special unloading stall for the trucks which is on a slight ramp. The unloading nozzle is at the rear of the truck. The centrifugal pump is located below the bottom of the truck and is run until it trips off on low amp draw. This has been very successful. Obviously, this won't work with rail cars.

If air is a hazard in your process, then I would probably put an inert gas blanket on the top of the car so that no air can be drawn in. We usually use nitrogen. If air is a problem only because it causes the pump to loose prime and fail, then BigInch is correct and you just need to position the air eliminator valve so that it removes the air to reestablish prime when the next rail car is connected. Of course, you could just use a manual repriming process for your operators.

It sounds like you should have two sets of unloading pumps so that one set does not need to draw through such a long suction line for trucks.


Johnny Pellin

 

[katmar]

Pieter, how much pressure can your rail cars take? It seems the basic problem is loss of NPSHa as the level in the rail car gets low. If the product can tolerate air you could blow the product out (or at least boost the NPSHa) with air. I suppose you could use nitrogen, but that would be expensive. I guess you only need a pressure equivalent to a few feet of liquid head.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com