Does anyone have experience geting a valve to to stop oscillating? 4

[1BrickShort]

I have a valve core that is oscillating durring a "flowing" condition, I can not have this in my system.

It is obvious that the oscillation comes from the spring characteristics of the valve core.

The question I have is... do all check valves use springs? What other small economical options are there to use a checkvalve design without springs?

Has anyone else ever had this probelm?

 

[btrueblood]

Not all, but those that don't use gravity as the restoring force. Or electromagnetic forces, but that is a solenoid valve, not a check valve. There are duckbill valves that use flexible, typically polymer, strips that bend apart to open a flow passage, but they are still relying on the spring force generated by bending (they also typically cannot stand very high reverse pressure).

Try changing the cracking pressure rating for the check valve (stiffer spring usually, but sometimes a softer spring quiets the valve also).

 

[israelkk]

Can you give more information? Is this a standard off the shelf check valve or a custom designed valve? Flow rate?, fluid? Inlet pressure? valve size? A drawing may help.

 

[FieldTeam]

I once had an accumulator fail in a system that caused noise/vibration in a system, which I isolated to a check valve as the source of the noise. Replacement of the accumulator fixed the problem.

 

[1BrickShort]

I have attached a draft of the valve core. The pressure is applied to the right side of the valve. The pressure pushes on the pin, and opens the cup on the left side of the valve. The spring creates a compression force on the stem which causes it to close when the dP between upstream and down stream is low (i.e. less than 50psig).

From a free body diagram standpoint, I recognize that the spring force is what is causing the oscillation. My question is, how do I get it to stop???

 

[1BrickShort]

Sorry, I forgot to upload the image...

 

[MikeHalloran]

General principle: ANY pressure control valve will sing in the absence of damping. It has to do with the dynamics of the fluid passing between the control element and the seat.

In your case, you're using what appears to be a commodity valve that really isn't intended for continuous throttling service, hence has no intentional damping mechanism.

You might find an interchangeable core of slightly different construction that has enough accidental damping to make a difference.

If that doesn't work, you may need to roll your own. How many do you need, and much money have you got?



Mike Halloran
Pembroke Pines, FL, USA

 

[btrueblood]

Well, I was going to agree with israelkk, but...the dynamics of your valve are influenced by upstream and downstream system influences (the accumulator that FieldTeam mentions, line sizes and volumes...). Change the spring first (I would suggest going softer, i.e. a 25 or 15 psig cracking pressure first - the hope is that the core flips from closed to a fully opened condition, where it's pinned against the stop and so won't rattle...as much). Then change the valve internals, as Mike suggests, if you can.

 

[israelkk]

btrueblood

1BrickShort didn't yet disclose the fluid involved (gas or liquid and if it is a gas what kind of gas. Helium will behave totally different than Nitrogen, etc.).

He didn't mention the downstream conditions. Is it discharged to the amosphere which makes the downstream pressure fairly constant or the downstream pressure flactuates too.

He didn't disclosed the flow rates, working temperatures range and more.

Therefore, I will be more carefull before I suggest any action. I believe that before rushing to machine parts and change the spring, an analysis of the static, dynamic, thermodynamic and fluid flow analysis should be done to build a simulation model of the valve to get the feeling and better understanding of the physical parameters that influence the valve behavior.

From looking at the drawing and assuming it is not just a schematic drawing I can say with good confidence that one problem is the sealing geometry design. But as I mentioned above mauch more info is needed and I do not believe that the problem will be solved in this forum.

 

[1BrickShort]

All,

I do appriciate everyones opion and advise, I understand IsrealKK that the finite details of my situation will not be solved through this or probably any other forum, however I do not believe that is what forums like this are most useful for.

btrueblood and MikeHalloran were both offering advise (with a set group of assumptions, I undertand), that I will take as just that, ways I could change my valve and spring to dampen the system. The math has shown that any spring device in a checkvalve like this will cause oscillation, under the right conditions...

IsrealKK if you are that interested in the exact parameters of my problem send me your resume and we can talk about about your employment. However the lack of critical thinking is not a good start...

I appriciate you btrueblood and Mike Halloran for offering suggestions using critial thinking. Whether I can use that exact advice or further the ideas I am currently pursuing, or open up a new realm of thought on my other solutions.

That is the purpose of this forum is.

For the mathematitian to model the effect...Gaseous N2, Varied temperatures (assume 70F), Flowing from a source into a defined volume (i.e. a cylinder, thus downstream pressure is increasing, part of the problem, also note that the oscillation is a function of flowrate, and harmonics of the spring, so we go through bandwidths where there is oscillation, then no oscillation, then oscialltion, then none...

p.s. The sealing geometery works just fine.

 

[israelkk]

"..p.s. The sealing geometery works just fine." I have no doubt that the check valve seals. I meant that the actual geometry of the sealing parts affects the flow regime around the stem and is part of the parameters that determine if the flow can steadily hold the stem in the opened position when you need the flow into the cylinder.

The flow rate, (size of the orifice) and stem travel affect the chatter of the valve. The natural frequency of the spring-mass is quite simple to avoid using a non-linear rate spring. However, this too depends on the stem travel which depends on the orifice size and the desired flow rate.

Is the cylinder a constant volume (no moving piston) or the cylinder volume changes with time (moving piston).

The 50 psi difference between the upstream and downstream pressure is not enough to define the working condition of the valve. The actual value of the upstream pressure and downstream pressure affects if the orifice flow is choked or not and this too has its effects.

 

[btrueblood]

"Therefore, I will be more carefull before I suggest any action. I believe that before rushing to machine parts and change the spring, an analysis of the static, dynamic, thermodynamic and fluid flow analysis should be done to build a simulation model of the valve to get the feeling and better understanding of the physical parameters that influence the valve behavior."

Israelkk, I don't mean to disparage your comments, they are valid ones. I would disagree with you about the above comment, as, at least in my shop, it's quicker to cut some chips and test it than it is to try and build a model that will accurately describe the conditions that are causing oscillation of a valve. And changing out a check valve for a different cracking pressure usually is a quite simple task. Your methods might work, but will take a lot of time and effort from somebody knowledgeable in fluid systems modelling. My way may or may not work, but nonetheless will yield some valuable information that helps in subsequent analysis, should it become necessary.