High Turndown Crude Oil Flow Meter

[sysengineer]

I have to upgrade some old turbine flow meters used for allocation measurement on a 3-Phase test separator. Currently, due to high turndown, there are two turbine meters for each phase (high and low range). I would like to have a single meter for each phase but the high-turndown limits my choices. Basically for the oil phase I am limited to either vortex or coriolis. For the water phase possibly mag flow and clamp-on ultrasonic for the gas.

Turbine meters seem to be the de-facto standard for allocation measurement but a single turbine meter will not meet the high turndown (75:1 roughly for oil). Can anybody offer any suggestions?

Thanks

 

[zdas04]

With 75:1 turndown you simply do not have measurement, you have numbers. No single meter can do anything close to that.

What pressure is the gas side at? I've had horrible results from real ultrasonic meters in gas below about 150 psig, and clamp-on's never give you adequate repeatability for a test separator at any pressure gas.

If I was designing a test separator today, I'd probably use V-cones for both liquid and gas. On the high turn-down meter I'd probably put a 1-inch, 2-inch, 3-inch, and 4-inch in parallel with control logic that could shift from one to the other at 15% and 85% of range, and could parallel any 2 or 3 or all 4 to get the high end. That is the only way I know of to get your very wide turn-down.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[danw2]

>I am limited to either vortex or coriolis

Forget Vortex. Vortex is a velocity meter and has an inherent low-flow cutoff (that's rule-of-thumb at about 10% of the max flow rate) due to the low number of vortices formed at low flow. It's not like DP where you can try to read low flow if you want to, vortex meters do the low flow cut-off for you.

Notice on the typical vortex sizing chart above that the origin is NOT zero flow, it's 15 gpm. That's where the flow signal output starts from. Any flow rate between 0 and 15gpm is zero. The physical signal (4-20mA, frequency, whatever) can be scaled from zero gpm, but won't be anything but zero until the flow rate reaches a measurable rate.

 

[zdas04]

Since you are talking about a 3-phase test separator, I'm going to assume this is an Oil & Gas industry, Oil field operation. Let me translate, 15.2 gpm is 521 bbl/day. Any number less than that is reported as zero.

The coriolis is about the same magnitude.

David Simpson, PE
MuleShoe Engineering

Law is the common force organized to act as an obstacle of injustice Frédéric Bastiat

 

[sysengineer]

FYI - I have had some discussion with a meter vendor regarding this.

I was quoted for coriolis meters that would achieve 100:1 turndown with an accuracy ranging from 0.44% to 1.96% over the full flow range. This is quite acceptable and for this application each phase will use a coriolis meter to meet the life of field turn down.

 

[danw2]

Coriolis meters do NOT autozero. Be sure that you have positive shutoff so that meter can be zeroed (turning the pump off is NOT a guarantee of zero flow). Any zero shift throws the accuracy claims right out the window.

Please let us know what how well the meter does when it's been in service for period.