instalation of psv 2

[fane14]

hello to all,

I have a little problem.

I have to install 4 PSV for 4 gear pumps from discharge in suction.
The pumps are sucking heavy oil from trucks, each pump from 2 trucks and on the suction lines are installed 2 big filters.
The pumps are discharging with 12 bars and the PSV are setting up to 17 bars.
The question is where is properly to route the dischrge of the psv, before or after the filters or is better to route directly to truck???

thank you in advance,
fane14

 

[JohnGP]

Hi fane14,

It is best to route the piping from the PSV discharge back to the suction side of the pump, after the filters. The oil has already been through the filters once, so doesn't need filtering again. It's also an unnecessary complication to route it back to the truck, requiring more connections and hoses.

Cheers,
John

 

[fane14]

Thank you JhonGP,

my query is if i we will route after the filters(between the outlet of filters and suction nozzle), the fluid will not pass to truck(i can not find the sense of flow in the filters if is only one direction)and/or will damage the filter or will create some problems/turbulence inside.

If no will recirculate the fluid till the ps(at 20 bars) will trip the pump.

thank you again
Fane14

 

[JohnGP]

Maybe I don't fully understand your setup.

There should be no backflow through the filter to the truck if the pump goes into recirculation through the PSV. If it continues like that then it may start to get a bit hot. The PSV should only relieve if the discharge line becomes blocked, or a valve in the discharge line is shut.

The PD pumps I am familiar with had an integral relief valve to allow recirculation within the pump itself, so if you have an external pipe from a PSV to the pump suction, then that is just an extension of that principle.

Cheers,
John

 

[micalbrch]

Why do you think that the oil from the PSV will be at 20 bar? I think that is your mistake.

JohnGP's advice is correct: From the PSV back to the pump's suction between filter and suction nozzle.

 

[fane14]

yes is correct if the filters are not allow the back flow(if they will work like a check valve)mean that will be only recirculation and in my opinion if the blockage in the discharge will continue the temperature will be increase(the pump resist till 200 Celsius degrees) and the pressure will be increase. I have installed a pressure switch on the suction line,that will trip the pump at 20 bars.

In my opinion is if the filters allow the back flow I don't know what will be the effects inside the filter and in the truck - maiby some turbulence.

best regards
fane14

 

[JohnGP]

No, the flow will just circulate from the pump discharge through the PSV and back to the pump suction. There is no excess flow to go back to the truck.

Additionally, you need to install the pressure switch on the discharge side of the pump to detect the increase in pressure. As micalbrch indicates, the pressure can only get to 20 bar on the discharge side, and there will be a lower pressure after the PSV to the pump suction.

Actually I would set the pump trip set point at just below the PSV set pressure, and just use the PSV as a secondary control.

Cheers,
John

 

[don1980]

Something else to consider for this case, and all liquid service PSV cases, is that liquid PSVs don't reclose as quickly as those in vapor service. In liquid applications, the pressure must drop (blowdown) by 15-20% in order for the PSV to reclose. Vapor PSVs typically reclose after 7-10% blowdown.

You have a gear pump, so be aware that the head curve is flat. In other words, the discharge pressure is relatively constant over a wide range of flow rates. So, what will happen if/when this PSV opens? .....It might not reclose until you shut off the pump.

Why do liquid PSVs require more blowdown? It's because the spring must overcome the momentum (mass*velocity) of the moving column of liquid. You can easily see that a moving column of liquid has much more momentum (force) than a moving column of vapor.

This is something you always have to consider with liquid PSVs, especially when they're routed to the atmosphere. Be sure you have enough operating margin (set pressure - operating pressure) for the valve to go back closed.

 

[sheiko]

Reading this discussion, I have a few comments:

- I would rather pipe the PSV discharge directly on the source truck in order to avoid the risk to upset the pump's ability to pump.

To don1980:
- Could you please share with us the source of this (blowdown in liquid service)?
- This sound strange to me because a blowdown of this level (ie 15-20% of set pressure) would almost mean there is no risk of PSV chattering (rapid and intempestive opening and closing) in liquid service because the closing pressure would be very low with respect to opening (set) pressure...This would also mean no need to limit inlet line pressure loss to 3% (which normally correspond to about half the "normal" blowdown)

"We don't believe things because they are true, things are true because we believe them."

 

[fane14]

Hello to everybody,

Thanks to all for cooperation.

yes Jhon I was wrong I have a PS on the discharge setup to 20 bars and another PS on the suction for LL flow both will trip the pump.

My problem is please convince me that no back flow will pass the filters to the truck.
The suction pressure is 0.2bars and the truck is atmospheric.
The PSV will discharge with 17-20 bars- How are you sure that I will not have back flow in the filters?

thank you again
fane14

 

[fane14]

For sheiko,
Please my friend can you give me more details where did you route the psv discharge directly to the truck?
Usually the trucks have it available nozzle for discharging?
What pressure have you on the your Psv discharge?
will not be problems turbulence in the truck for big pressure(17bars)?

thanks again
fane14

 

[don1980]

Sheiko......good questions. This is something that I though was well documented, but based on a quick google search I'm not sure it is. The source of my information about liquid valve blowdown is from many years of discussions with manufacturers. Liquid trim valves were introduced ~1985 as a result of a change in ASME Sec VIII. That change required the rated capacity of liquid valves to be determined at 10% overpressure. Prior to that, liquid valves needed more overpressure to reach the same flowrate. They also needed even more blowdown (as much as 30%). Manufacturers responded to that code change by introducing liquid trim valves which open faster - they get to a higher flowrate at 10% overpressure. These "new' valves also have less blowdown, typically < 20%.

I'll continue to look for a reference source, and I'll come back to report it if I find one.

If the blowdown for liquid valves is much higher than that for vapor valves, does that mean that liquid valves have less risk of chatter? Intuitively, one would think so, but inlet pressure loss is just one cause of chatter. For example, chatter can also occur in liquid systems if the acoustic frequency of the inlet system matches the natural frequency of the valve. Valve stability is not as well known in liquid systems as it is in vapor systems. Long inlet lines increase the risk of chatter in liquid valves, even if the pressure drop is low. The explanations, as I understand it, is that this is caused by acoustic resonance.

The bottom line is that we can't say that a liquid valve won't chatter, just because the inlet loss is less than the blowdown.

 

[JohnGP]

fane14,

The PSV will only lift if there is a restriction in the discharge piping, if for example a valve is shut. It will only pass as much flow as the pump is able to pump at the pressure for which the PSV is set. That flow when returned to the suction side of the pump will be drawn back into the pump, otherwise there is less flow going through the pump and hence back to the suction.

The pressure after the PSV will not be 17 bar. The PSV will only open to the extent that the differential pressure across it matches the differential pressure across the pump (assuming the PSV is spring loaded). The only way to get flow back to the truck is if you have a separate discharge line all the way there. Even then you only get the flow that the pump can produce.

Cheers,
John