Pressure surge following a fast opening of a line size ball valve.

[figure8]

I'm familiar with the equations used to determine the forces resulting from fast closure of a valve in liquid service, however, I'm trying determine the instantaneous flow of a gas when a quick opening valve opens.

I'm pressuring the section upstream until I get a 1.0 bar differential over the valve (this is a 12" line size ball valve that opens in 5s). The flow in to pressurise the line upstream only needs a small Cv to pressurise the line in about 1 minute (circa 100). Due to the quick opening valve characteristic (and the Cv of a 12" ball valve being +22000) there will not be an opening sufficient to restrict the flow as soon as it opens.

As I understand it, following the quick opening on the valve, the pressure energy building up behind the valve will be converted in velocity as the fluid start to move and there will be a pressure surge. Is there anything i should be using to determine the increased flow at this instant vs the flow into the section that was pressurising it.

I have a feeling that it will require specific software to do the analysis but I'm just looking for a worst case flow at the moment of opening as i suspect the FIs on the line would not detect this.

Would a worst case evaluation based ideal gas equation to determine the number of moles in the section upstream and then determining the moles that would need leave the section to drop the pressure back to a point where the pressure driving force falls back to approaching zero then assuming that passes through the valve in the first second be a completely ridiculous way to determine this.

 

[LittleInch]

I've read this a few times and I'm confused. How is flow going backwards (up stream?

But basically I've never bothered doing a gas surge pressure analysis as the impact is negligible and the gas compresses so that it essentially "cushions" any impact.

you can get very high peak local velocities, but that's not surge.

A drawing and sketch or series of sketches to show your planned operation would help enormously.

 

[figure8]

I've read this a few times and I'm confused. How is flow going backwards (up stream?

But basically I've never bothered doing a gas surge pressure analysis as the impact is negligible and the gas compresses so that it essentially "cushions" any impact.

you can get very high peak local velocities, but that's not surge.

A drawing and sketch or series of sketches to show your planned operation would help enormously.

Sorry the upstream sentence isn't clear. I meat the gas it flowing and the section upstream of the XV is being pressurised from flat to 20bar. I've attached a sketch.

It's a line with Oxygen from to a burner. I need to make sure there isn't an significant change in flow from the 10000Nm3/h being used to pressurises the section when the XV first starts to open.

I didn't think of adding a sketch sorry.

I'm also reasonably sure with it being a valve opening rather than closing and a gas rather than a liquid thee effect is going to be relatively small.

 

[LittleInch]

All helps to understand.

So you're opening a valve with one bar DP with a relatively small volume upstream?

Any impact is going to be minuscule.

There will be a small burst of gas but how much friend in the volume in your downstream system.

Or open the valve more slowly?

Why not open at a lower DP if you're worried by this?

But this is a CFD type of thing. There's no formula

 

[figure8]

All helps to understand.

So you're opening a valve with one bar DP with a relatively small volume upstream?

Any impact is going to be minuscule.

There will be a small burst of gas but how much friend in the volume in your downstream system.

Or open the valve more slowly?

Why not open at a lower DP if you're worried by this?

But this is a CFD type of thing. There's no formula

Thank you for your time and response on this. I think you are correct that it's probably a CFD evaluation required. I agree the effect would be miniscule however even a small amount of addtional oxygen at the burner would result in elevated flame temps and therefore increased pressure due to expansion downstream which I'm trying to determine worst case for.

You are correct it's a relatively small upstream volume and relatively low pressures in a gas system. If the valve could be opened slowly that might help. Also yes I'm hoping to use the findings of this to push for a smaller d.p over the xv to reduce the risks but was hoping to quantify the reasoning.

 

[Snickster]

The pressure will not "surge". The pressure on the upstream and downstream side of the XV will equalize when the XV opens. Basically the equalized pressure will be based on the ideal gas law and the volumes of each pipe segment upstream and downstream of the XV, assuming the flow in from the FV into the piping is exiting through the burner nozzles at the about same rate. In this respect the pressure downstream will spike above 19 bar but not due to any water hammer type surge, but just due to mass flow from upstream of the XV in accordance with PV=mRT.

 

[georgeverghese]

If this were to be a liquid, then you would have surge as the momentum of the liquid, once it hits the XV, creates a pressure wave travelling backwards. But this is compressible gas.
So you are throttling the FV for about a minute to get the pressure upstream of the XV to 20bar. Keep line velocity as low as possible, this being pure O2. There is a low risk that pressure upstream of the XV could rise up well beyond 20bar if the FV is not closed in time, so a manual vent valve in this interspace may help to g et rid of excess pressure.

 

[figure8]

The pressure will not "surge". The pressure on the upstream and downstream side of the XV will equalize when the XV opens. Basically the equalized pressure will be based on the ideal gas law and the volumes of each pipe segment upstream and downstream of the XV, assuming the flow in from the FV into the piping is exiting through the burner nozzles at the about same rate. In this respect the pressure downstream will spike above 19 bar but not due to any water hammer type surge, but just due to mass flow from upstream of the XV in accordance with PV=mRT.

Thank you. I think this is what I was thinking, something akin to determining settle out pressure. My only concern was that its a relatively short length between the final xv and the burner tip and a significant volume upstream and weather that burst of pressure actually changes oxygen to the burner. In all honestly I'm not expecting it too but will hopefully be able to show that.

 

[figure8]

If this were to be a liquid, then you would have surge as the momentum of the liquid, once it hits the XV, creates a pressure wave travelling backwards. But this is compressible gas.
So you are throttling the FV for about a minute to get the pressure upstream of the XV to 20bar. Keep line velocity as low as possible, this being pure O2. There is a low risk that pressure upstream of the XV could rise up well beyond 20bar if the FV is not closed in time, so a manual vent valve in this interspace may help to g et rid of excess pressure.

Yes thank you, and you are right, with this being oxygen I'm also needing to consider the velocity doesn't exceed allowable velocity limits as a result of any of this in the oxygen pipework.