11 replies to this topic

[ayan_dg]

How to find Cv of a control valve in two phase flow . Say the upsteam to the control valve is at 10 barg and having 20% steam and 80% water by mass.

[Zauberberg]

The valve Cv is an inherent characteristic of any particular valve, and it does not change regardles of what type or phase of fluid is passing through it. I believe you are interested in estimating the total two-phase flow through a valve with given Cv?

[ayan_dg]

The valve Cv is an inherent characteristic of any particular valve, and it does not change regardles of what type or phase of fluid is passing through it. I believe you are interested in estimating the total two-phase flow through a valve with given Cv?

Yes, that will help .

[Zauberberg]

I think this would be extremely difficult - if not impossible - to calculate because there is dependency between the flow regime (and liquid holdup) upstream of the control valve, and actual mass flow through the valve. I believe there is almost infinite number of solutions covering all the cases between the stratified flow and the mist flow regime in such systems. That is one of the reasons why control valves are basically never employed for flow control in two-phase systems.

I would be interested to hear if someone has an idea how the actual two-phase flow can be calculated through the valve with known Cv.

[sheiko]

I would be interested to hear if someone has an idea how the actual two-phase flow can be calculated through the valve with known Cv.

Hi there,

I 've got a HP magazine article dealing with these issues. Time to pick it home, scan it and i will download it asap.

Edited by sheiko, 20 July 2010 - 05:08 AM.

[chemsac2]

ayan_dg,

I think it is still the old method of calculating Cv considering vapour alone and then considering liquid alone. Control valve Cv for two-phase would be sum of Cv values for liquid and vapour.

Warren controls "Valve sizing and selection technical reference" available online gives the formula.

Regards,

Sachin

[sheiko]

See attached FYI

Attached Files

•  Improved-c.pdf   540.58KB   478 downloads

Edited by sheiko, 20 July 2010 - 05:07 AM.

[cea]

See attached FYI

Dear Sheiko-

Indeed, this is avery good article. Thanks for it.
However, I am unable to forsee situation wherein control valve is required to install in line carrying two phase flow.

In general, I have followed philosophy of installing a control valve so that two phase can occur only at control valve downstream, if two phase is unavoidable.

Especially in steam service, I have never dealt with such situation where control valve is required to put in two phase. As I understand, there are two ways of installing control valve in steam system. 1) Install control valve in steam inlet line to consumer. 2) Install a control valve in condensate line after having steam, condensate separation in condensate drum (This is a typical arrangement wherein sufficient size steam trap is not available). No one can expect that two phase is coming out of consumer consuming steam.

Please indicate the situation wherein installation of control valve in two phase is unavoidable.

Otherwise, my recomendation is to avoid such situation by installing control valve at appropriate elevation.

Regards,

[chemsac2]

cea,

It can not be pressure, level and flow control valve, but can be a temperature control valve. I have seen one such case in which flashing occured across a level control valve and there was another temperature control valve in downstream circuit. Temperature control valve had two phase flow entering it (hydroprocessing plant V-L separator outlet to stripper system).

Regards,

Sachin

[Zauberberg]

Thanks to all participants in this thread, and particularly for sharing interesting technical resources. Certainly when there is no other option, control valve has to be inserted into a two-phase line and used for control purposes - choke valve being one specific but somewhat different example.

I am still of opinion that controlling two-phase flow is impossible in real systems, simply because of high influence of hydraulics (liquid holdup, flow regime) on control valve performance and actual flow (and that was the original question - estimating phase flows through a valve with known Cv) of both liquid and vapor/gas phases through the valve. The same valve, with same characteristics, and with the same % opening will pass different quantities of gas and liquid in all cases, for different flow regimes, which change as the total mass flow through the piping changes.

[djack77494]

I am still of opinion that controlling two-phase flow is impossible in real systems

I totally concur with this viewpoint and would further state that measuring two phase flow is darn near impossible. Even if it were (or is according to some vendors) possible, it is difficult and expensive.

[kkala]

I cannot recall a control valve for two phase flow in refinery service and as a rule we avoid such cases in the design. But in the far past I had sized a control valve (or it was orifice?) by adding vapor and liquid CVs, as Sachim (chemsac2) has mentioned. Books at that time suggested the assumption of homogeneous flow for fittings, so this could be also used for control valve sizing.
Sachim reports an example of TCV on a two phase (apparently condensate) flow. More orthodox practice would be to install a flash tank after the first LCV (but flash steam might be lost to atmosphere) and direct the condensate to the TCV downstream. Yet flashing would also occur downstream of TCV, sacrificing some precision. Original case of no flash tank could be acceptable, since this TCV would try to "correct" heater fluid temperature irrespectively of two face condensate flow pattern. Sometimes control need not be as precise as possible, and a control valve can be better than nothing.
However it seems that the original example of LCV - TCV has a disadvantage, that the two controls can be conflicting: Which controls condensate flow, upstream level or downstream temperature? This issue is alleviated by the flash tank, especially if its holdup time is big.
In brief, control valves can be exceptionally used for two phase flows; design needs more caution in this case and control is not expected to be as precise as in single phase.
For instance, the original case of this thread refers to 20% steam & 80% water by weight; which means that actually steam controls the flow, having much bigger volume. So CV can be determined as above for the control valve. But above assumes more or less homogeneous flow (e.g. not slug or annular or stratified). This can be "helped" by constructing proper upstream fittings / arrangement.