Measuring actual flow rate

[sysengineer]

I have an application where a vortex flow meter is installed upstream of a flow control valve which is regulating nitrogen flow through a vent system. The upstream pressure is usually constant but varies between 8 to 11 bara. The downstream pressure is approximately 1 bara.

The meter appears to have been incorrectly specified and is reporting actual flow rate of around 3 m3/h which is outside of the recommended turn down. The meter range is 0 to 20 m3/h and the required flow rate is 12 Nm3/h. Currently there is a dynamic pressure / temperature compensation algorithm to convert actual flow rate to normal flow rate.

A solution appears to be to move the meter downstream of the flow control valve in which case the actual flow rate will be close to the required flow rate at normal conditions.

I am trying to understand how the change of conditions makes this a feasible solution. Presumably there is a change in velocity downstream of the flow control valve in which case the vortex meter will see an increase in volumetric flow rate?

Is there anything else to consider?

 

[xnuke]

Placement downstream of the valve can screw up the velocity profile, and vortex meters can be very sensitive to poor velocity profile.

Some questions:
What meter make and model is it?
Do you have the meter installed in accordance with the manufacturer's directions? Is there enough straight pipe length both before and after the sensor?
What is the turndown on the meter?
How do you know the meter is reading wrong? Do you have a confirmation via another measurement method?
What size is the pipe? Do you have pipe fittings near the meter?

xnuke
"Live and act within the limit of your knowledge and keep expanding it to the limit of your life." Ayn Rand, Atlas Shrugged.
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[sysengineer]

Hi thanks for your reply.

There are a few things I didn't mention in the original post. Firstly, some other options were considered such as replacing the meter with one suitable to achieve the required turn down. As is normally the case the solution I described is preferred since it is a relatively easy to implement. Secondly there is a general reluctance to spend money on fixing this problem and there was a suggestion to remove the meter altogether which obviously suggests this is not a particularly critical instrument. Anyway this is fairly irrelevant I'm more interested to learn about the changes in velocity (if any) and why this solution is considered feasible.

For info, I can confirm based on the manufacturers accuracy vs flow profile the current meter is inaccurate below 5 m3/h. There is no flow reference to compare against but I can confirm the required actual flow rate of 3 m3/h is outside of the recommended turn down limit. It appears that vortex meters become almost useless below some minimum flow since the velocity is not great enough to produce a reliable vortex shedding profile.

Also it is a Rosemount 8800D in a 1" sch 80 line. The upstream / downstream pipe lengths I haven't explored and is something I will try and confirm.

 

[georgeverghese]

For N2 purge gas flow, one approach is to not bother with measuring the flow - the flows are usually small as in this case. All we do is to install a downstream sensing pressure regulator and a fixed RO downstream. Set the PCV setpoint to be higher than 2times the max normal operating backpressure of the vent system. This way, the flow through the RO will always be constant, since the pressure upstream of the RO is greater than critical pressure.

Safeguarding will be to install a PSL downstream of the PCV to alarm at DCS.