Nitrogen pressure increase on dead end systems.

[Carlloss]

Hi all,

Not sure if this is the right place for this post but could not find anything closer (feel free to point me into different group if you think there is one).

I have recently faced a situation which is beyond my knowledge so I hope having other user view may help with solving it. I work for company who design and manufacture pressure regulators. Two months ago we supplied one of our high flow units to customer who design and fabricate fire fighting systems. System idea is fairly straight forward, when there is a need for it to be on following happens, 8 Nitrogen 300 bar cylinders are activated at the same time to deliver Nitrogen, pressure is reduced on first valve to 30-50 bar Nitrogen then goes via our regulator with further reduction to 6-8 bar, after this Nitrogen is delivered to water tank and push it via series of pipes.

Problem we are seeing is that regulator pressure should stop at the preset point (say 7 bar) once reached but what is does it goes far beyond in and activate PRV. This only happens under dead end scenario, where the water tank downstream valve is closed and we are actually not pushing anything. The pressure regulator is leak free, it maintains preset pressure under flowing condition, it did maintain set point during factory test (we were flowing much less Nitrogen and were filling empty cylinder not water tank), it is dome loaded with external reference line to sens pressure outlet pressure, flow values are well with it capacity and we supplied numbers of dome and spring loaded version of it in past. Line size is 2" and flow via regulator is in a region of 25 000 SL/min.

I do not know if the system conditions/parameters affects Nitrogen and we face situation when there is some pressure recovery??? Have no idea why this happens.

Any ideas, thoughts are highly appreciated. Thanks.

Karol

 

[LittleInch]

well not a bad start but can you describe how:

what happens on water stop
How long does it take for the pressure to exceed the 7 bar set point?

Most importantly how do you for sure ( i.e. 100%) that your regulator valves are not passing on no flow??

Are you getting very low temperatures at the outlet due to JT cooling down from 300 bar?
Is this damaging the sealing material?
What does the regulating valve look like?
Is the water freezing during operation?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[LittleInch]

See this for more info on N2 cooling

https://www.eng-tips.com/viewthread.cfm?qid=374882

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Carlloss]

Thanks for replies and link.

Water stop is closed and tank is filled would say 80-90% own capacity (not sure if I answer it fully?)

It takes around 50s, initially pressure drops (which is expected), then gradually increases towards set point, fluctuate a bit around it and on similar rate carries on. Do not know at which point it would actually stop as due to PRV.

I do not know it for 100%. I only know that valve is 100% seat leak tight.

The temperatures drops down but we do not have any freezing due to JT, nothing being damage, no marks on O' rings, diaphragm etc.

From my perspective I have either a case under which our pressure regulator dose not close and remains open allowing pressure increase or it dose close and further pressure increase came from system?

We had one success with activating only one cylinder instead of 8 and setting regulator at 4 bar.

Valve picture attached.



Karol

 

[LittleInch]

Do the first stage regulators also show some drift? i.e. does the pressure rise above 50 barg?

Without some sort of diagram we can't see your system so my guess at this point remains that during operation the seats loose a bit of their sealing capability.

Looking at the diagram is the plug simply seating metal to metal?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Compositepro]

My guess is that your regulator is way oversized for the volume that it is required to fill. What is the actual gas volume in your tank when it is filled with water? If you are trying to control the pressure in a one cubic foot volume with regulator valve that will flow 200 cfm, you will get to your set pressure in a fraction of a second. The valve will not be capable of closing fast enough to prevent overshoot.

 

[Carlloss]

I will see how much of the PI&D I can obtain as this is not under my jurisdiction, same applies to obtaining volume info. will need to check with customer.

The seat seal is metal against soft material (tried PTFE and PTCFE both gave same results). The first stage regulator is not really affected, it has big droop initially as it's set to 50bar but once activated goes to ~30 bar. Other than this nothing else is really noticeable with this unit.

I will need to dig out the sizing calculation as it might be that those were done taking under consideration only full flowing conditions.

Will back as soon as I get some extra information.

Thanks.

Karol

 

[Carlloss]

Hi,

Attached it is a simple PI&D with some volume information. The tank volume is 16400 L and it is filled with 15000 L of fresh water as an operating level so we have a bit of volume.

Regulator was sized to work around 40-50% of it's flow capacity.

Still have no idea why I see this pressure increase.

Karol

 

[LittleInch]

The HF250 manual

http://www.pressure-tech.com/files/18/HF-250 Series - O&S Manual.pdf

section 3.2 appears to say that a Dome loaded control an increase in pressure results in an increase in downstream pressure.

Looking at your diagram this seems to be correct. If the sensing line goes into the top of the dome then an increase in pressure will result in pressure / force down on the seat an hence opens up the incoming plug.

Spring loaded would seem to work as per your description.



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[gerhardl]

My guess is that LittleInch is probably right. Trouble with this type of regulation comes in 90% of the cases either from getting the pilot pressure from a unstable/unsuitable point, or from unstable/undulating pressure in the system.

Your systems to the nozzels are probably filled with partly air on higher points and water on lower points. This might create a system with water hammer and differently undulating pressure systems in the different pipelines to the nozzels???

If yhe system is designed for 10 bar try the system with PRV at 10 and system at 7.

 

[Carlloss]

Ok. Thank you for all your answers and inputs. Really appreciated.

I did on site testing and went through quite few combinations, tested even a spring loaded version. Unfortunately,non of it get me to finding solution but will try to sort it out.

Thanks again.

Karol

 

[LittleInch]

Ok, Please let us know if you find a solution.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Carlloss]

Hi all,

I have no solution at this moment but will let you know if I manage to get something. Thank you for all your help and info.

Karol