Relieving Capacity for a PSV due to Pressure Control Valve Failure.

[asade]

don1980 advised that I should start this question on a new thread.

Let us consider this scenario: A PV is installed on pump discharge line to regulate the discharge fluid pressure from 526 psig to 360 psig with a throughput of 70,000 bopd. If for any reason, the PV fails, it is expected that excessive pressure would be experienced in the downstream pipeline. To avoid this, a PSV would be installed.

What condition/scenario should be considered to determine the relief flow rate to size the PSV? The purpose of this is to protect the downstream pipeline from excessive pressure because it has been derated to 360 psig.

Someone suggested that the relieving flowrate should be the total throughput from the valve. Is the condition right?

I am what I am by His grace

 

[Bill3752]

Don, here is the approach I would usually take. There are three levels of conservatism:
1. Most conservative - determine the letdown flow through the CV, using the MAWP of the downstream piping as your downstream pressure (probably the set point of the RV). No credit for normal flow.
2. Middle of the road - determine the flow using the OVP as the downstream pressure (110% of MAWP), no credit for normal flow.
3. Same as 2, but take credit for normal consumption. To take this credit, look at the distribution of normal flows and pick one that makes sense. If this is constantly used, use the normal rate. If the system is used off/on, credit = 0. If the flow has a range, use the lower end of the range.

 

[don1980]

asade - thanks for moving the post. That keeps things more orderly for forum readers.

There's a very straighforward answer to this question. TD2K was explaining this in the other thread.

The solution is easy to visualize if you look at your pump curve. Specifically, look at the relationship between pressure (head) and flow.
1) Draw a horizontal line on the pump curve to represent the max head allowable before the pipeline is overpressured.
2) Mark the point where this horizontal line intersects your impeller curve.
3) Draw a vertical line downward, to the x-axis (flow).
4) If we ignore friction loss in the pipe, that's you're answer.

As long as you have that amount of flow, or more, the pressure will be acceptable.

That's the technical answer, but now let's look at this from a the perspectives of safety and practicality. If you use a PSV, where are you going to send the discharge? This is going to be a big flowrate and you can't just go to the ground with it unless we're talking about water. Also, you're probably looking at installing several big parallel PSVs. Consider alternate methods for limiting the pressure during this emergency. One option is to use a HIPS to trip the pump when the system pressure exceeds the max value. Or, use a HIPS to open a recirculation valve. You might also be able to greatly reduce the required relief flowrate by cutting down the pump impeller. If you can live with a lower dP across the control valve you'll be able to reduce the impeller size. That will also save energy.

 

[asade]

don1980,

The discharge of the PSV is proposed to be connected to the inlet line to the oil surge tanks. A recirculation valve is installed on the system, which is controlled by the LIT connected to the pump tank. PSH is install on the discharge line to trip the pump in the event the pressure available exceeds it maximum set point.

The PSV would be installed as a safety backup and to protect downstream pipeline.

On the steps you provided, the pump curve provided does not have its impeller curve on it. Drawing a horizontal line from the Y-axis (head), the maximum allowable pressure before the pipeline is overpressured into the chart does not touch the head vs capacity curve at all. It is far below it. Check the attached pump curve. How can I apply your recommendation with this curve.

Bill3752,

The normal flowrate through the valve is 70,000 bopd (2,042 gpm) at upstream pressure of 526 psig and downstream pressure of 280 psig. The specific gravity of the crude oil is 0.865. The MAWP of the downstream pipeline is 360 psig, which may be used as the PSV set point.

Based on your recommendation, I used HYSYS to check if I can determine the flowrate through the control valve at different downstream pressure. I am having the same flow rate across the valve. The significant changes I noticed is on the calculated control valve Cv values generated by HYSYS sizing handle. I dont know if I am missing something or overlook a point. The MAWP is 360 psig. Based on your recommendation 1, Cv = 276.0 and on 2, CV = 311.8.

Using Cv = Q * Square root(G/dP), the Cv for the control valve at normal operation was obtained as 121.10. When I used the Cv = 121.10 at new downstream pressure of 360 psig and dP = (562 - 360) = 202 psi, I obtained a new flowrate of 1,692 gpm (58,011.4 gpm). Can I use this as the PSV relieving flowrate?

In the HYSYS model, what could be responsible for no change in the flowrate across the control valve when the downstream pressure changes?



I am what I am by His grace

 

[don1980]

asade - there was no file in the link.

When I said "impeller curve" I was referring to the head curve (pump curve) for the specific impeller installed in you pump.

Maybe I'm misunderstanding the relief scenario. You have a pump and a downstream pressure control valve. So, the piping on the downstream side can be overpressured if that control valve goes full-open. Are you talking about a zero flow condition? The valve is open and there's no flow, aside from what gets relieved through the PSV? That's what I inferred but I'm not sure that what you meant.

 

[asade]

don1980,
Am sorry for the error. Find the attached file in the link.

I am what I am by His grace

 

[asade]

don1980,

The CV is installed downstream of the pump discharge line. It is meant to maintain the pressure across the downstream pipeline by reducing the upstream pressure of 526 psig to 280 psig. The scenario is when the CV fails and couldnt regulate the pressure, thereby causing excessive pressure across the downstream pipeline. How do I determine the relieving capacity for the PSV that would be installed?

I hope I have explain better.



I am what I am by His grace

 

[asade]

Bill3752,

I had used InstruCalc to determine the flow rate through the control valve based on the first two conditions you proposed and I obtained the following results:
1) Flowrate through the control valve at MAWP of downstream pipeline = 58,708 bpd
2) Flowrate through the control valve at 110% of MAWP of downstream pipeline = 51,953 bpd

Would I be correct to use case 1 as the PSV relieving flowrate?

I am what I am by His grace