12 replies to this topic

[Rush123]

I have come across an interesting case.

 

On one of our facilities, there is a level control valve which takes the water from the prodn sep to the water treatment skid. The Prodn sep operates at 1300 psi, and the water treatment skid operates at 50psi. There is no choke / RO in this line. In other words, all the pressure drop occurs across this valve. The flowrate is quite small as well. (circa anywhere between 2 - 20 bbls / day) The valve internals need to be changed frequently, which is not surprising due to the large DP across the valve.

 

The valve was originally sized for a much larger flowrate. The existing valve has a CV of 0.35 and when we did a fresh sizing case we determined that the valve may need a much smaller CV of ard 0.002 in its maximum case.

However, when we looked at the process trends, we noticed that the valve position actually stays closed for some time, then swings 100% open…stays open briefly, then closes again to allow level to build in the separator. While this may be fine for some methods of level control, I am not sure if this is the best approach to have a valve opening and closing against such a large pressure drop. It may be better to have the valve operate between a more reasonable range and not fully open / fully close often (or at all.) i.e have the valve between 20% - 80% open.

 

At the moment, I am not convinced that changing the valve to a smaller oner will fix the problem of having to change internals annualy.  According to me, whatever valve is used the method of control (i.e full open / full closed) will not be changed by changing the valve. The onyl difference with a different size valve will be the time the valve stays open / closed.

 

In summary:

This is a two part problem. On one hand you have a valve that may be too large for current service, and on the other hand the control method may need to be adjusted. We should try to fix / tune the exisitng valve and monitor operation before buying / changing a valve...correct?

 

My question: How can we ‘tune’ the level control valve to have it operating more in a modular manner rather than this on/off manner? BTW the controller is a PID  controller.

[tushar.patil84]

You have to put orifice plate inline before control valve to reduce the preesure on the valve Disc & it will solve your problem maintaince of youe valve.

[Zauberberg]

Valve behavior seems a bit strange. I could understand quick closing action because the valve is - as you say - oversized, and it passes much more flow than required. But on the other hand, I do not understand the quick opening action. You might want to look at the controller tuning parameters as well.

[curious_cat]

What are the settings of the PID controller? Are you sure it is acting as a PID controller?

[Rush123]

I am in some agreement with the possibility of putting in an RO / choke valve to ensure the pressure drop doe snot occur across the valve alone. It is better to try and play within the exisitng system boundaries before making any modifications to the facility though. For this reason, I am trying to determine what parameters may need to be configured to tune the valve.

 

This is out of my forte totally. An instrumentation engineer will lead the effort on tuning the valve (with me present) sometime next week. From now till then, I'll google what I can to try and educate myself of such things. If all goes well, the valve may have a more 'steady state' of operation after the exercise. 

 

Curios_Cat: 

The tuning parameters are:

Gain - 1.4

Rate - 2

Reset 120

 

These values are almost greek to me. Hopefully someone else on the forum may be able to translate / shed some light.

[Zauberberg]

Troubleshooting the problem together with Instrument Engineer sounds like a good first step. I have seen the derivative action applied for level controllers quite rarely (if all), but again, this might be a different application.

 

There are some good online articles about tuning of level controllers:

 

http://www.engr.ucon...l/pdf/cep08.pdf

http://www.isa.org/F...ticesTuning.pdf

http://www.picontrol...cal Process.pdf

http://home.hit.no/~..._controller.pdf

[fseipel]

Staged control valves such as Fisher notchflo will be quite resilient but expensive -- pressure drop doesn't occur all in one stage.  If Cv is 175 times that required, a lower Cv is worth pursuing.  You can simply auto-tune the valve, but if it's Cv is so oversized, it is likely going to operate nearly fully closed after tuning, where erosion issues will be worst -- which may well be worse than slamming open and closed.  An orifice plate will reduce the pressure drop across valve and reduce the wear and is also relatively inexpensive.

[Babu Prasad]

Generally Angle valves used   for high pressure drop, if you believe control valve is over sized then try to throttled upstream block valve of the level control valve which help some extent to have smooth control over level. Tuning of control valve will not help unless this control valve floats between 20 to 80% range. In your case valve is over size and pressure drop also more hence you need to restrict the flow. Temporally throttled the upstream block valve of the control valve and carry out the tuning.

[curious_cat]

I'm not so sure your problem is fixable by tuning. Say you had a very oversized valve on a 6" trying to fill a tiny testtube: No matter what control algorithm you use the situation seems problematic. 

 

I think you need to fix your valve to the correct size. 

[Zauberberg]

There is no doubt about that, if the difference between actual vs. design Cv is as stated earlier. However the valve behavior does seem strange, and if the case, tuning can be of no harm.

 

I would say a better option would be to reconfigure level control mode from modulating to step control. That way the same valve can be used in future, if flow gets closer to the design value, for any reason.

[curious_cat]

There is no doubt about that, if the difference between actual vs. design Cv is as stated earlier. However the valve behavior does seem strange, and if the case, tuning can be of no harm.

 

Agreed. 

[Rush123]

Thanks for the input. 

I'll update this topic later this week after we try to tune the valve.

[Hamedcheng]

Hi

 

How can indicate bitumen or vacuum bottom  Level  in bitumen tower (blowing tower)?

 

( bitumen in blowing tower is hot)

 

Thank you