Circulating liquid full system 1

[EngineerRvB]

Hello all,

I'm wondering what would happen in the following situation:

A hot oil system has a centrifugal pump and is flow controlled by a spill back.
The pump suction pressure is controlled by a split range controller and a buffer vessel, supplying more N2 to the vapor head space when the pressure is too low and venting of some of the pressure in the head space when the pressure is too high.
The flow to the users is flow controlled.
There is a small circulation flow from the discharge of the pump to the buffer vessel, to keep the buffer vessel warm and to vent low boilers. This circulation flow is small compared to the pump minimum continuous flow.

What would happen if the connection between the buffer vessel and the suction line becomes blocked (completely or at least so that the flow is lower than the flow of the small circulation line)?
I can image that there will still be flow via the small circulation line to the buffer vessel, slowly filling up the buffer vessel, but what happens in the main circulating system? In my mind the removed liquid cannot be compensated by vapor (from open tank or blanketing). Will a vacuum be created in the circulating system?

Thanks and regards.

 

[LittleInch]

At T=0,nothing will change, but over time, which could be seconds or minutes, the volume in the loop will reduce as the small bleed line continues to take some liquid off, though if the line is blocked this will rapidly reduce to zero. The pump suction pressure will fall and presumably the gas pressure will increase to maximum pressure to try and maintain pump suction pressure.

But sooner rather than later your pump will start to cavitate / the liquid will boil creating a gas and liquid fluid and flow will fall off, but would continue in some form as long as the pump is the lowest thing in the system

This would probably continue until the pump starts to break up or someone realises its making a rather odd noise....

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

 

[Snickster]

I basically agree with LI but I will explain it in a different way.

If the buffer tank outlet line is not blocked then there will be equal flow in to the tank from the system and equal flow out with a constant controlled pressure in the buffer tank. There will also be constant level since there will be constant volume in the system, which is the volume that the system was started at and the level of the buffer tank that the system was started at. I assume that with this close system all users don’t consume any water but just use the water for some function like a shell and tube heat exchanger heat transfer fluid.

OK now suppose the outlet line of the buffer tank begins to be blocked up but very slowly, for instance if there is some kind of build up from the chemicals in the liquid like minerals in hard water or whatever.

What will happen is this:
At the same pressure in the buffer tank but more restriction in the outlet pipe, there will be more pressure drop in the outlet pipe for the same flow so the flow out will want to decrease. If the flow out of the buffer tank decreases and the flow in remains the same, the pressure in the suction line will begin to drop since it is starting to suck more fluid out of the suction line than is returning. Think of this as a very slow quasi static process where everything is happening real slow and it will be easier to visualize.

However, since the pressure control is on the suction line and not on the buffer tank vapor space, then the nitrogen blanket pressure will increase to maintain the flow at a higher pressure drop until the flow out equals the flow in. so that the pressure at the pressure sensor remains at set point. I image this will continue until the maximum pressure available from the nitrogen is reached considering what safety controls (relief valves, maximum nitrogen available pressure, or high pressure shut down on nitrogen line, etc.) limit the maximum nitrogen pressure in the buffer tank.

There may come a point where the nitrogen pressure available reaches its limit (there should be some set limit as noted above or the vessel may fail) so the flow out cannot remain constant and keep up with the flow in as flow path narrows and narrows and becomes very blocked. So when the pressure limit is reached, the flow out of buffer tank will decrease to be less than flow in, volume in the pipe system will begin to drop, and so will the pressure, with the lowest pressure in the system being in the suction line, and in particular at the suction of the pump.

There will reach a point where the suction pressure reaches the vapor pressure of the liquid and the liquid will begin to vaporize forming gas bubbles in the liquid creating cavitation of the pump. Under quasi static conditions the suction pressure would remain constant at the vapor pressure. This is because as the more net liquid leaves the system, more volume of vaporized gas replaces it thereby maintain the pressure constant at the vapor pressure, and pump keeps cavitating. Eventually the pump will destruct through cavitation or stop pumping because so much vapor in the system will cause it to stall and be unable to pump anymore. Centrifugal pumps will stall and not pump with a certain percentage of gas in the suction liquid.

Another thing to consider is that at some point net flow into the tank will cause the discharge pressure to be felt by the tank so eventually the buffer tank vapor space will disappear and be under the full discharge pressure of the pump. This will push any nitrogen remaining in the tank out, further adding gases to the piping and further tending to stall the pump.

So in the end you will have the buffer tank full of liquid, all nitrogen that was in the tank now in the piping system, the system suction operating at approximately vapor pressure of the liquid so a lot of liquid in the piping system vaporized, pump cavitating and likely stalling.

 

[georgeverghese]

This is not a good design scheme. All return flows should go back to the expansion drum, including min flow control. The suction line may become vapor trapped in your scheme, reducing pump flow.