Pressure Relief Valve Chattering 2

[JAlton]

Can anyone offer an opinion based on experience or physics which support the position that there is more tendency for SRV in Liquid to service to Chatter than SRV in Compressible Fluid Service? Thank you.

JAC

 

[JLSeagull]

Cosnider replying in safety relief valve engineering to eliminate double threads.

 

[MortenA]

Chatter occurs because the pressure drops when fluid starts to flow (when the PSV opens). The liquid has a higher density and acceleration head is thus higher?

Best regards

Morten

 

[proinwv2]

Oversizing the valve will cause a rapid pressure drop when the valve opens, and it will quickly close.

Subsequently, the pressure rebuids in the vessel and the valve starts to open, and before it is far from the seat, it closes due to the rapid decrease in pressure. Ad infinitum.

Paul Ostand

http://www.ostand.com

 

[MortenA]

He didt ask for the reason for chattering (i think he knows that) but why its more likely to occur in liquid servce than in gas service.

I gave i shot but im not sure thats why.

Best regards

Morten

 

[gerhardl]

A SRV for gas is most often (or should be) especially designed for gas with a special designed (and sometimes adjustable) reaction ring. This reaction ring (formed as an extra 'skirt') has the purpose of gathering and leading the gas flow to open the SRV smooth and fully, and avoid chattering.

The reason for the skirt is 'mass flow' necessary to keep the disc lifted, and as such, I believe, partly answer to your question.

But answer for physics backed by correct design: Properly designed SRVs (different designs for incompressible and compressibel fluid) should at least be somwhat equal compared to shattering tendency.

Answers for other than physics: As shattering occurs when operating pressure is reoccuringly too near lifting pressure or just above lifting pressure and repeatedly relieved by small outlet amounts, this is either caused by selecting a too large valve or wrongly selected or adjusted set pressure compared to process operating pressure. Human fault presumed equal for both types of fluid! ;-)

 

[MortenA]

Gerhard

API 520 states that 3% max dP in the inlet should be designed for. It dosnt state this clearly - but I was under the impression that this was in order to avoid chattering. Is this not the case?

Best regards

Morten

 

[JAlton]

ASME Sec. VIII, Div. 1, Non-Mandatory Appendix M, Para. M-6 (a)states that "all nonrecoverable inlet losses shall not exceed 3% of the valve set pressure".

But my concern is really whether liquids have more propensity to chatter than gases given the same inlet piping configuration.

Thanks,
AC

JAC

 

[maocba]

Hello.
Somebody has written about a reaction ring. I think it's one of the factors which has influence in chattering (or fluttering). It deals with the concept of "blowdown" in a PSV. Depend on how you adjest the blowdown ring you can modify the pressure for close in two ways: more early o more late.

 

[MikeHalloran]

I read somewhere that _any_ pressure control valve needs to have damping to avoid chatter, and I have seen chatter eliminated by adding damping features to hydraulic valves.

The geometry would presumably be a little different to do the same thing for a gas valve.



Mike Halloran
Pembroke Pines, FL, USA

 

[jistre]

I have do nothing scientific, have done no testing, and have no numbers to back this up, but I've been thinking about this, so I thought I'd pass my thoughts by you guys and see what you think. What can I say? I'm a fan of Einstein's thought experiments.

I have two identical SRV's on two separate, but identical, piping systems and they're set to pop at the same pressure. The pressure profiles in my two piping systems are identical. In fact, everything about the setup is identical between the two except that one of the fluids is completely liquid, and the other completely gaseous. No two-phase flows.

Okay, so we ramp the fluid pressures up to the set point and the valves begin to open. As the liquid valve opens, a small amount of liquid comes out, the pressure drops quickly, and the valve closes. However, the gas should be a bit different. As its valve opens, a small amount of gas begins to leave the valve, but as it is a gas and enjoys its space, it begins to expand when it encounters the lower external pressures. It seems that more volume has to go through the valve, so it would tend to stay open a bit longer than the similar liquid valve. Since the period that the valve is open is longer, the chatter appears to have a lower frequency.

Your job is to see the angles I missed in my thinking. One possible angle is that the local pressure of the gas at the disc face drops quickly due to the expansion through the valve and the disc closes, only to encounter the pressure rebound caused by the system pressure, which quickly opens the valve again. I suspect there are a lot of little factors that I missed that would be helpful to know. Maybe we'll all come out of this learning a bit more on the way to figuring out what's going on 4 NoSoup4U.

Now, I have to ask: we are talking about chatter and not simmer here, right?

 

[JAlton]

Yes, we are talking about chatter. Not simmer. My experience is in hands-on Pressure Relief Valve repair and testing, not engineering. I can set the adjusting rings in a two ring design PRV to make it chatter due to increased pressure drop in the huddling chamber. I cannot force a single ring PRV to chatter due to the restricted adjustment available with one ring (i.e. uper ring is replaced by "skirt" whichhas fixed geometry. I can understand the reasoning regarding the lack of a skirt in a liquid service design PRV. No expansion of liquid, no need for skirt. However, in a single ring design where ring adjustment in my 34 years experience testing PRVs cannot induce chatter, a one ring design SRV which may be placed in Liquid or Gas service has only one reason to chatter, inlet piping loss. This is either because of excessively long inlet piping. excessive bends in inlet piping, or reduced inlet piping. I know and can demonstrate the results of all these configurations. However, I cannot prove that Liquid will have a greater propensity to chatter. Given Jistre's scenario, I have no basis for stating that the liquid system will suffer chatter more frequently or more seriously than the gas system. I am interested in Morten's comments regarding density and acceleration. This is the area where I do not have the education or experience to determine the effects. Thank you for your comments. I appreciate everyone involved.

JAC

 

[israelkk]

I had the privilege to design a safety relieve valve for aerospace use for gas at 1500psi. To avoid chatter we actually designed the outlet orifice and the sealing poppet geometry such that the calculations clearly show that even after the valve is opened the fluid drag force can hold the poppet in the opened position against the return spring. The trick was to find the correct sealing geometry.

One more difficult issue is to make sure that the poppet will seal again when the pressure drops after a minimum specified pressure difference below the crack point. I have seen bad designed safety valves that seal back after cracking in a pressure lower than 80% of the crack pressure.

 

[mbt22]

Perhaps a naive answer, but a very slightly compressible liquid will need a relatively small change in volume to drop the pressure compared with a gas. Is it just that it's rather easier to oversize a liquid RV?

 

[JimCasey]

A star for Jistre. Compressible flows expand and chatter is avoided for the reasons he said. Incompressible flows must accelerate to flow, must have adequate flow to keep the valve pressurized and open, thus have more mechanisms to enable chatter.

Size your liquid PRV carefully

 

[JAlton]

JIm Casey, thanks for putting succintly.

JAC

 

[cb92]

Nobody mentioned water hammer. A damper(accumulator)would probably help.