How to determine vena contracta pressure for a control valve? 1

[jrjones]

I am working on a failure investigation of a control valve in my plant. The valve type is a Fisher 461 Sweep Flow Valve. The valve has been cavitating badly and has caused damage to the valve and downstream piping.
What I want to do is to determine at what feed conditions (rate, upstream pressure, downstream pressure, etc) flashing starts, ie when the initial pressure drop to the vena contracta drops below the vapour pressure before recovering to the downstream pressure.
The reading I have done allows you to calculate choke flow conditions etc, but this is based on upstream and downstream pressure, not really looking at the vena contracta. Help!

 

[vzeos]

jrjones, here are some resources that might help you out.

http://www.fisherregulators.com/technical/sizingcalculations/#FlashingandCavitation

http://www.documentation.frco.com/groups/public/documents/book/cvh99.pdf

http://www.ajdesigner.com/phpcavitation/cavitation_number_equation.php

 

[davefitz]

Also, the "ISA handbook of control valves" has a chapter on cavitation , and prediction of the onset of cavitation.

 

[jrjones]

Thanks folks...looks like the ISA Handbook has what I was looking for...will give a try when I get back to the office.

Cheers

 

[Scotsinst]

The valve you mention is specifically designed with low Fl factor to flash/cavitate/mix more than a typical angle valve. It is used for dirty flashing liquids.
It is used as a visbreaker valve and is useful for medium pressure drop applications.
If your pressure drop is beyond the 461 capabilities then consider a Fisher DST-G or Masoneilan 77000 series which will reduce vibration and cavitation while handling the flashing.
You need to size upstream piping large to keep the liquid above bubble-point with a long reducer just upstream of the valve and 5 pipe diameter straight run upstream of the valve then swage up immediately downstream of the valve.

 

[JimCasey]

Good point, Scott. To elaborate: a sweep-angle valve is designed to avoid damage from FLASHING service, thus it has a large bowl upstream of the seat ring, and a venturi discharge. In some designs the venturi is superhard alloy or ceramic, extended the full length of the tailpiece. If the valve is installed in a CAVITATING service the only part of the previously mentioned modifications that help are the superhard trim.

Cavitation is caused by excessive velocity: Bernoulli tells us that as the velocity increases the pressure decreases. Decrease the pressure to Pv and the fluid flashes. Slow the fluid down a bit and the pressure recovers. The bubbles collapse. This is cavitation.
Cavitation PREVENTION trim has many stages incorporating changes of section and direction. These cause the fluid to be throttled WHILE CONTROLLING THE VELOCITY. Your sweep angle valve does nothing to reduce velocity and has a single stage of pressure reduction between the plug and seat.

Side note: there are also cavitation CONTROL trims that actually allow cavitation but force the bubbles to implode in free volume where the possibility of damage is eliminated. Drilled cage type trims with radial inward flow operate in this manner. The bubbles crash into each other in the center of the cage and no damage is done.

If you have a clean service, cage anticavitation trims work. Example: CCI DRAG(tm) trim. However these make highly effective but difficult to clean strainers. So if there are any solids in the stream, an anticavitation design such as the Masoneilan LincolnLog should be considered.

So the moral of this story is: you probably have a valve that is misapplied. Get all the information together about your application and contact your most competent valve vendor for some applications help.

 

[jrjones]

Thanks for the replies.
Really I am trying to assess just want you pointed out, is this the right valve for the application?
More info for you:
Fluid: Vaccuum topped bitumen
SG: ~0.818
Pv: ~0.5 psia (very low I know)
Valve: 8x10 with 3.5" seat.

What is happening is that when we run on this valve (level control valve) the piping system vibrates and has managed to crack numerous small bore vents from the line. This only happens at the higher (desired) feedrates. But when we switch to the bypass valve (globe) the vibration goes away.

Now we have done a vibration assessment of the piping and have modified the support system slightly, but the problem still prevails. When the system is vibrating, the valve sounds like it is cavitating. If you talk to the valve vendor...cavitation cannot cause vibration. However, I look at a pretty simple cause and effect relationship here and say that cavitation must be able to cause this type of vibration.

Thoughts?

 

[jrjones]

OOps, should have also said that upstream pressure is 330 psig, downstream ~85 psig (calculated from vessel level tap).

jrjones

 

[vpl]

jrjones

Whats the flow rate in the piping? We've seen a lot of flow-induced vibration problems that have resulted in small bore pipe cracks and failures.

The cavitation noises you're hearing may be another symptom rather than a cause....

Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[JimCasey]

You did not state the Viscosity or flowrate, but I guessed at some reasonable values for those parameters and when I ran the numbers through the Flowserve sizing program I get severe cavitation. Yes cavitation CAN cause vibration.

The math in any sizing program assumes pure chemicals with well-defined properties. With Bitumen or most distillates you have sixteen dozen specific chemicals and they all have slightly different vapor pressures. When you get a little pocket of flash it forms a bubble and launches a blob of tar downstream. This would be sluggy flow and that causes real shock to the piping.

I wouldn't use an anticavitation valve with the tarry product. I don't think you are seeing cavitation damage inside the valve. You seem just to be describing inertial effects from the sluggy flow.

If you can put the valve directly on the nozzle for the vessel into which it discharges you may find that the vibration is attenuated. I think a lot of what you are experiencing is because of the way the stuff flows in the downstream piping. If you get rid of the downstream pipe you get rid of the impulse mechanism.

Moving the valve seat closer to the nozzle would also help but it would mean you need to replace the valve. I'm thinking an eccentric rotary valve such as the Valtek Maxflo3, Masoneilan Camflex, with the seat at the outlet end. Hardfaced trim, probably reduced capacity. Go for the cylinder actuator option as it is stiffer and would be more stable with this stuff that's flowing nonuniformly through the valve.

 

[jrjones]

You guys caught me...sorry. The flow rate is roughly 65-95 kbpd. The viscosity is 1.4 cp.

The vibration problem only seems to occur at flow rates >65 kbpd and really gets bad around the 85+ range.

The arrangement of the piping system has the control system at the ground with the bypass loop and the nozzle on the vessel is at an elevation of ~200 ft.

We had the control valve rolled out once and did see minor indications of cavitation damage on the seat retainer (diffuser cone area).

One of the biggest problems I am having is convincing the powers around here that cavitation can cause the high levels of vibration we are seeing. The difficulty is that the vibration does have a steady (although quick) oscillation to it while it is felt that cavitation would be quite dissonant.

As yet we haven't got vibration data...working on that this week.

Thanks for the continued help.

jrjones