Essentials of PRV Stability Webinar

AIChE Academy Webinar Presented by ioMosaic

Attendance to this webinar is free.

Could undetected pressure relief valve (PRV) installation instability be putting your facility at risk? Watch this webinar to find out.

In this 60-minute ioMosaic sponsored webinar, gain a better understanding of the risks associated with potentially unstable PRV installations. Then, investigate how to perform an engineering analysis when PRV instability is suspected. Mr. Houston shares how to evaluate historical data and inspection records and presents screening methods for evaluating instability. Because speed of sound estimates are critical, he addresses how to improve your estimating accuracy. Wondering how engineering analysis is being successfully applied in the real world? This webinar includes case studies that will show you while demonstrating state-of-the-art developments in PRV stability.

 


Our Presenter

Casey Houston

Casey Houston
Senior Partner, ioMosaic Corp

Mr. Houston is a Senior Partner at ioMosaic and brings over 15 years of engineering and process safety experience to his role as a leader of the firm’s Relief Systems consulting group. His work is focused on managing and executing large-scale pressure relief and flare systems design projects for reactive and non-reactive chemical, petroleum and pharmaceutical systems, as well as providing technically sound analysis and documentation for existing process and reactivity hazards. Read more


Webinar Q&A

1. Can you give an example on which fluttering is acceptable?

I would take the fluttering example and look more at the type of fluid of the pressure that I am looking at to really try and translate that into a consequence, and ultimately a risk. So, it might even be that maybe I predicted fluttering for a scenario that is a very low likelihood event and it’s a clean fluid that doesn’t pose any toxic or flammability risks. Therefore, I would have to look at that whole profile before I really make a cut and dry decision on whether or not fluttering would be acceptable.

 

2. Is there any way to test a PSV for a leak at pressure other than the release pressure so you can detect a leak prior to putting the valve in service? We recently installed a conservation vent that passed all factory leak tests but had a leak that was detected with LDAR monitoring. Is there a way to avoid this?

I am not aware of a way and that’s a little bit outside the PRV stability. Sounds like a lot of low pressure vents need to be handled pretty carefully and the DIERS group has had a good presentation on handling and maintenance of low pressure vents in particular with respect to the LDAR and meeting fugitive emissions requirements.

 

3. How does the pipe flexibility affect speed of sound impact and how will you use this in a design?

Our pipe flexibility tends to decrease the speed of sound, so the speed of sound we would measure using the fluid properties is just the speed of sound in the fluid. Whereas the flexibility of the piping and piping support and the rigidity of that system can tend to decrease the speed of sound. So you saw during our presentation that our Δ P wave is proportional to the speed of sound in the system and there were a few (we went through it a little quickly during the presentation) but there are a few different correlations for whether the pipe is anchored at an end, or both ends, or rigidly supported throughout and how to estimate the impact of that on the speed of sound of the fluid.

 

4. Do these equations work for liquid too?

Yes. These are good for vapor, liquid and two phase systems. And the important part is really making sure you have a good estimate for that speed of sound in the fluid.

 

5. Can you expand on the wave pressure loss?

The wave pressure loss is really the two parts of that equation – the fluid hammer term and the fluid inertia term. And the impact it has on the valve is just the wave component of that pressure loss. Which is why we have the Τ term and so that accounts for only the pressure losses from our source for our reflection point, if we have a large pipe diameter change or sudden large pipe diameter change that would be a reflection point where the pressure wave comes from our source to our PRV. And we want to quantify by the pressure losses associated with that wave traveling through the linear length of the pipe.

 

6. If one has 2 different relief design scenarios, one chooses the biggest one but that may cause chattering for other relief scenarios. What would you recommend?

That is a very typical case and if we do have that sort of installation we really look for multiple valve installation with one valve that is perhaps smaller and set at a lower pressure and then a larger valve set at a much higher pressure. We also want to be careful to avoid acoustic interactions between the two because we don’t want them fighting each other and causing each other to be instable so we want those set pressures to have a decent margin between them. We will usually have a smaller lower set valve to account for those smaller scenarios and then a larger valve to account for our design or controlling case.

 

7. If an existing valve is not stable and less than 3% ITD – what are some cheap solutions to fix it?

It would really depend on how the existing installation is and the details of that installation as to what we can do inexpensively. A lot of industry tests and valve manufacturers have shown that pilot operated valves are a very good solution for avoiding chattering, especially a pilot operated modulating relief valve.

 

8. How are acceleration losses due to flash and flow handled in the analysis?

I believe that the answer to the question is that they are ignored and that the delta P wave term is simply the math flow rate. And what you would need to do is get the average of your fluid densities and your speed of sound due to that flashing and flow – is how that would be captured. The acceleration would still be – or recovery on the other side would still be ignored because the frictional component of the wave is still just the frictional losses in a piping system. So that Δ P wave term only includes the velocity head loss effect and the Τ.

 

9. What stability effects could be expected when the inlet piping has a concentric reducer?

This is similar to the previous question that we would be accelerating the fluid through that concentric reducer. What we’re seeing is that larger diameter piping will actually act as a reservoir and so that might have beneficial effects if we have that larger diameter piping. Regardless of the historical thought that is lowering the frictional losses but it might also give us a little bit of the damping of the fluid as it travels to the valve.