Excess line pressure when pump is shut off

[mechem03]

Hi everyone,

I am wondering if anyone has had any experience with a piping system becoming overpressurized when the pump is shut off. Here is my scenario:

Fluid: No. 2 diesel fuel
Pump: 2 hp, 1800 rpm gear pump
Flow: 30 gpm
Design pressure: 30 psi

There is a 12 ft negative suction lift at the pump. Downstream of the pump we have: internal pressure relief valve set to 50 psi, pressure gage, butterfly & check valves, 420 micron strainer, oscillating piston flow meter (which requires the 420 micron strainer), and an automatic solenoid valve. After this point, the fuel is discharged into a storage tank. A float switch in the tank shuts off the solenoid valve when fuel is no longer needed which then sends a signal to the PLC to shut off the pump.

When the pump is operating at steady state, the discharge pressure gage reads at 15-20 psi. When the pump is shut off, the pressure gage initially reads 5 psi, but after several minutes, it creeps up until it reaches 50 psi. After draining some of the fluid out of the line, the pressure falls back down to 5 psi, but then creeps back up to 50 psi. After draining the system again, the gage stays around 5 psi.

I've been puzzled by this all afternoon. Anyone have any opinions or advice? I'd appreciate any thoughts on this. I don't think it's due to thermal expansion, but if anyone disagrees, let me know!

 

[Latexman]

Can you eliminate the blocked in volumes by removing the check valve or car sealing the butterfly valves open. What you are after is the fluid upstream of the solenoid can expand back to the supply tank and the fluid downstream of the solenoid can expand forward to the discharge tank. The closed solenoid (fail closed, right?) may make the check valve redundant.

Good luck,
Latexman

 

[Thealanator]

Can you do away with the solenoid valve and just have the float switch turn the pump off? Perhaps you could lower the float switch to compensate for additional fuel that could drain into the discharge tank from any high points in the piping.

I am not familiar with an alternating-piston flow meter, but it almost sounds like it could act as a line block at some point.

 

[mechem03]

Latexman, my understanding is that fuel is being routed from the relief valve into a drip pan rather than the suction side of the pump so the fuel would not make it back to the supply tank. Changing that would be an option, but my understanding is that an internal relief valve on a pump is only meant as a temporary solution to overpressure - I believe it's better to have an external relief valve for this section of pipe.

Alanator, the fuel forwarding system serves multiple tanks and the tanks do not all call for fuel at the same time so we aren't able to remove the solenoid valve. It would defintitely help if we could, though!

 

[hydtools]

Add another solenoid valve, normally open, to connect the trapped fluid back to the pump suction when the delivery tank float switch closes the feed solenoid valve and shuts off the pump.

Ted

 

[mechem03]

Hi Ted. I'm not sure that this would work well in this case - there are multiple tanks that require fuel and if one tank is off, but another is on, the line with the shut-off tank is going to be sending fluid back to the pump and basically recirculating the fluid.

 

[hydtools]

Wire the bypass solenoid valve to each demand system. When any one tank requires fuel, the bypass valve will close.

Wire it to the pump so that anytime the pump is running the bypass valve is closed.

Ted

 

[Latexman]

Most gear pumps with internal relief I've seen discharge back to suction. That's a first for me.

In my previous post I tried to described how to eliminate the hydraulic relief scenario. It'll work for multiple destinations too. No check valve and only one closed solenoid valve (we use automated block valves) at each destination.

Good luck,
Latexman

 

[Latexman]

Continuing on, you can solve this by adding a hydraulic relief PSV, or by eliminating the check valve and eliminating or administratively controlling the manual valves. Your choice may be influenced by past practice, policy, the unit’s layer of protection analysis (LOPA), your boss, your co-workers, etc. I am an advocate of the KISS approach (Keep It Safe and Simple), so I’d recommend working towards eliminating both the check valve and manual valves. I really cannot think of a good reason to keep the check valve. Automation has made it redundant, thus creating the thermal expansion hazard (no good deed goes unpunished). Your operations and maintenance folks may have good reasons to keep the manual valves, so have Plan B (administratively controlling the manual valves) in your back pocket.

During this entire post there has been something nagging at me in the back of my mind, and it may be the difference in the industry you work in and the chemical industry I work in, or it could be a company-thing, region-thing, or culture-thing. Anyway, is this small bore piping you are dealing with? I think of small bore as < 1 inch nominal diameter. Using a solenoid valve directly in line (which is what I have interpreted you have) suggests small bore piping, but using butterfly valves suggests large bore piping. I don’t believe my company would use a solenoid valve directly in the fuel line, especially if it is large bore piping. We would use a quarter-turn valve with spring loaded actuator for fail safe positioning in the fuel line. Usually a 3-way solenoid valve in air service would be used to drive the actuator. Anyway, I was just wondering what you were dealing with and compare that to my experience. That’s all.


Good luck,
Latexman

 

[mechem03]

Hi Latexman, thanks for your thoughts. There are 1" solenoid valves and 3" butterfly valves at this project (3" header, 1" branches to each day tank). There are only electric & manual actuators, no pnematics, and there is a back-up power supply. I appreciate that you've related your experiences - it's good to know what has and hasn't worked for others. I'll keep it in mind if I ever have to work on a similar design!