PRV on firewater pump discharge header

[akupikree]

I have a submersible firewater pump (offshore) with a rated capacity of 1050 m3/hr@12barg. There is a pressure relief valve (PRV) located at the discharge header just before the check valve and the valve is sized based on NFPA 20.

As per my transient study, during pump start-up (pump is running at the full speed), there is high pressure surge occurred in the discharge header. the PRV which is initially closed is too slow in dumping the overpressure. what is the typical time constant of the Firewater PRV to open? I'd appreciate if someone can suggest a solution for this issue.

Thanks in advance.

Fikree

 

[MortenA]

It depends on the actual size of the valve. Larger valve are slower. I assume its a spring loaded valve? I have used 1 sec minimum traveltime from fully closed to fully open (for any size - although i sespect that small valves would be faster). This is where it get tricky. I interpetate this is the time that it would take the valve to open fully IF the pressure jumped above SP instantaneously (that is in one simulation timestep) and stayed there.

I have never been able to confirm this numebr, and no PSV manufactor that i have approached would say anything to the subject.

Best regards

Morten

 

[akupikree]

Thanks MortenA.

Yeah, manufacturer can only say that the valve can open fast. But the question is how fast?

Back to my case, it is actually a self modulating pressure relief valve and it's an auto pilot valve. The size is 8". Ive tried the time constant of 1second but the surge is still high. Is it possible for the valve to response less than 1second once the pressure exceeds the SP?

I am still new in designing firewater system, and wondering what makes the pressure so high during pump start-up?

 

[stanier]

Please provide graphs of pressure versus time at the pump and at the check valve discharge.

Is the pipe from the pump to the check valve partially full whereby the friction is reduced and the pump flow is more than for steady state?

Why not start the pump against a closed isolation valve and slowly open the valve?

Pressure relief valves generally do not act quickly enough to mitigate surge pressures.

It could also be how you are modelling the check valve. Have you used the deceleration vs reverse velocity characteritics to determine check valve opening time?

It could also be your software. Some products are better than others.

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[akupikree]

thanks Stanier,

From pump discharge to the check valve is a dry section and the PRV is located just before the check valve. During pump start-up, the pump flow will be higher than rated flow (steady state).

As per my understanding, when the pump start it will fill-up the column (since this is a submersible pump). Once the column is full, the large amount of water will hit the check valve and PRV which are in a close position. this makes the pressure instantaneously increase above the SP of the PRV.

Im using PIPENET version 1.6 for my simulation. I dont know anything on how to set the check valve properties. it just a built in valve and i just drag the symbol into the simulation.

kindly refer attachment for the graphs.

 

[gerhardl]

I might be in 'deep water' here, but I have one question.

If I understand the situation correct the pump is starting to fill an air filled stretch of pipeline, with a T-section with a pressure relief valve, and in the mainline hitting a checkvalve, closed by an unknown pressure column on the other side.

The situation will in a simulation program give an inacceptable pressure peak, as the simulation program says that the PRV will open to slow (at too low capacity) and that the checkvalve does not contribute to a (fast/large enough?) relief, and that the cehckvalve in the simulation program is treated at standard (and unknown) opening time/relief time?

Correctly stated?

If yes:

a) It is difficult to advice on the simulation program, but it could well give an incorrect picture, as both program and realtime parameters are unknown.

b) There is a number of precautions, as indicated by previous posts, that could be take to avoid the pressure peaks.


What type description and details are available on the checkvalve?

I have a notion in my mind that the correct type of checkvalve, correctly dimensioned, combined with a correct layout of the upstream pipeline could at least minimize your problems.

 

[akupikree]

Thanks Gerhadl,

Yes, your explanation of the situation is correct. And i believe this kind of situation of pump start-up is typical for all firewater system in offshore platform where we have the pump outlet with ARV, PRV and Check Valve in sequence.

in PIPENET, there are 3 types of check valve and the properties for input are as below:

1) Inertia Check Valve
- Cv value
- Mass of check valve disk
- spring stiffness
- damping value
- inclination

2) Check Valve
- Cv value
- trigger option (flowrate or pressure trigger)
- trigger set point
- time constant

3) Non-return valve (this is the one that i used)
- no need to input any data.

Do you know which one to use?


most projects are using the typical layout/arrangement at upstream of check valve the same as mine. so i guess (hope so) there is nothing wrong with the arrangement isn't it? But still the transient study needs to be done as a prove that the system is safe.

 

[stanier]

Thorley suggests that an air release valve be installed before the check valve. The air valve is to have a controlled release such that the rising column of water is slowed before hitting the check valve.

This sort of analysis is not for someone who just plugs numbers into software. The use of option 3) means that you are not taking into account the dynamic behaviour of the check valve and will result in errors.

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[akupikree]

The firewater pump discharge is already provided with ARV but it doesnt help much in dumping overpressure during pump start-up.

I've also tried to use a proper check valve setting but still, it wont open fast enough. Plus, in case of blocked discharge and there is high pressure from the other side of check valve, the check valve will not open and the result is worse.

After some research, the only solution is to add an overboard dump valve which is set to be initially opened. Total flow from the firewater pump will be dump through this valve before its slowly close to control the pressure. This valve also can act as a minimum flow valve.

 

[stanier]

This pressure control valve should be used in conjunction with an air release valve such as the Ventomat RBxb.

http://www.ventomat.com

"Sharing knowledge is the way to immortality"
His Holiness the Dalai Lama.

http://waterhammer.hopout.com.au/

 

[Baram]

You should have pilot operated pressure relief valves or surge anticipating valve which install on by-bass not in the main line. this valves are adjustable by pilot and normally closed valve you should adjust the pilot with 1 bar higher than the dynamic pressure in your mainline.
Also this valve outlet you can take to tank or drain line.
The by-bass diameter which the valve will be installed should be designed that take flow rate 30% from the mainline flow rate for example if you have mainline 8" size that 3" pressure relife valve is good to eliminate the pressure positive wave.

There is no typical time to let pressure relief valve or surge anticipating valve to open, it depends on the pilot pre-adjusting by the end user.


I think you have pressure sustaining valves or Pressure reducing valves Or pump control valves because pressure relief valves should be installed in by-bass not in the main line.


You can visit the website for singer valve:

http://www.singervalve.com

Baram

 

[Baram]

By the way the surge anticipating control valve will be more quicke response than pressure relief valve, it will be opened at very less positive wave.

Baram

 

[Pumpsonly]

When you said submersible pump,is it a submersible motor pump or a engine driven line shaft pump commonly used in Off shore fire system.

AS per NFPA 20...PRV is only required for engine driven pump and the purpose is not for surge pressure relief.The main purpose of the PRV is to protect the fire main from possible damages cause by the over pressure in case the engine over speed.