Control Valve Minimum allowable Pressure Drop

[Justice100]

Hi there,

A problem I've come across a few times and had a few different opinions so I thought I would asked here...

Typical rule of thumb for allowable control valve pressure drop is 30% of the total frictional losses or a minimum of 0.7 bar. How critical is the minimum value... I have a system with quite a low total pressure drop so I can achieve the 30% criteria (using the typical line sizing velocity/DP criteria) but not the 0.7 bar. Is the minimum of 0.7 bar critical? In this case I have a flashing liquid (1 barg) upstream going to a tank at 0.1 barg using a 200m line with a 10m drop. It will be used very infrequently so trying to minimise investment without having to rely on an operator to control it locally by hand.

 

[Latexman]

Are you controlling a level in the receiving tank or the flow rate of the flashing liquid?

If a level, can you just do a deadband control with an actuated ball valve? It's cheaper than a control valve.

Good Luck,
Latexman

 

[Justice100]

Hi Latexman,

I don't think that would work for me. The LCV is controlling the level in the upstream vessel and there's only a few mins holdup at peak flow. The upstream vessel is filled from elsewhere with a flow that ramps up over time (10 hrs) and then stops. This is repeated a couple of times a year.

To add detail, the reason I am looking at this is the associated gas blowby case. The smaller line size would help the gas blow by case but not sure if it would be to the detriment of the control valve being able to control adequately. The increased DP for the smaller line size does not appear to impact the LCV Cv since it is choking but helps a lot with reducing flow for gas blow by flow.

 

[casflo]

I think that the rule of 30% control valve pressure drop, is applicable to liquids, but in this case, as you say, there is flashing in the valve and in the downstream line 200m long.
If the line and the valve completely open, has capacity to discharge the peak flow, the design is correct.
Keep in mind that the valve has choking conditions and the valve flow is limited to the critical flow. The downstream line must have enough size to discharge this critical flow.

 

[Justice100]

Thanks Casflo.

Yes, I think that is a good point. The 0.7 bar min is probably based on not having too big a control valve but in this case due to the choking it doesn't matter since it does not affect the control valve Cv until you get to a very low DP, in this case ~0.1 bar. Making my downstream line size larger does nothing to the Cv unless it causes the control valve DP to be less than 0.1 bar.

 

[Compositepro]

It depends entirely on how sensitive you need you control system to be. Sometimes on/off control is adequate. For other application you you may need to control flow to +/-1% of setpoint.

 

[shvet]

Rule 30% and dP 10 psi rule are both just rules of thumb and have a lot of exceptions.

30% related to pressure rise of head curve of a conventional centrifugal pump so in your case % can be calculated based on actual pump head curve and sensitivness required.

dP 10 psi related to sensitiveness of a conventional globe control valve. So as Compositepro said - dP can be decreased depending on sensitiveness or construction of control valve. For example I used control valves (with expensive special internal devices) with dP 1-2 psi at normal point for both gas and liquid services with no problems reported.

If flashing is a problem you can located a control valve just downstream of a pump (or any another source of pressure) in a point of hydraulic profile where total pressure is max so way it will be affected by high outlet pressure caused by flow friction. In this case flashing will occur only at start of pumping when flow friction is 0.