Flow Meter Calibration/Configuration Issues

[nomorenames]

Greetings -

I have an ultrasonic flow meter (UFM) in an 18" pipe that I need to calibrate. I have a NIST-traceable smooth-approach orifice (SAO) that I intend to use for this purpose. I am wondering what's required to assure fully developed/unperturbed flow through each of these devices.

The UFM is similar to this one (see PDF page 1-4, Figure 1-3). The ultrasonic transducers only protrude about 1/2" into the flow area, not much of an obstruction. Can I put the SAO immediately downstream of this, and expect accurate readings from that SAO?

Alternatively, can I put the SAO directly upstream of the UFM, and expect it to not mangle the flow going into the UFM?

A third option involves placing the SAO far upstream of the UFM, such that the UFM has a full ten diameters of smooth pipe ahead of it. This configuration would require the SAO to be open to atmosphere, and I don't have room for a full ten diameters of smooth pipe upstream of that SAO. Is there a generally accepted bellmouth configuration that will allow me to use something less than 10 diameters of pipe upstream of the SAO?

If nothing else, can someone recommend a good book or two on flow measurement that might help me answer some of these questions?

Thanks -
Mitch

 

[zdas04]

Swirl caused by an intrusion in the flow is really hard on both ultrasonic (the swirl is denser than the surrounding fluid so you get a pertubation in both time of flight and doppler shift) and differential producers (the swirling flow will tend to "screw" through the orifice at a significantly reduced dP).

18" pipe (really any pipe bigger than 12-inch) is really crappy at friction damping swirls. 10 pipe diameters just doesn't do much.

I would solve this particular proving problem with a Wafer V-Cone downstream of the ultrasonic. They are self conditioning and don't really need any straight pipe upstream or downstream of the meter. With a big enough β-ratio, the permanent pressure drop is really low. They tend to accomplish the goals of a UT meter without the need to prove them.

As for books, the book for gas measurement is Richard Miller's Gas Measurement Engineering Handbook The third edition was published in 1996, so it is more current than my Second Edition, but I don't know how much has changed.

David Simpson, PE
MuleShoe Engineering

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

 

[nomorenames]

Thanks for the book recommmendation; I have put in an order for a copy.

Unfortunately I don't get to choose the meters; I'm stuck with the UFM and the SAO.

In the normal operating configuration, the UFM has ten diameters of pipe upstream of it. This length of pipe is preceded by a flow-conditioning insert, possibly something from these folks.

I understand what you're saying about such a large-diameter duct doing a poor job of damping perturbations. Would you recommend that I not alter this arrangement upstream of the UFM (10D of pipe + flow conditioner insert), and that I simply remove that elbow and install the SAO (and its own 10D lead-in) upstream of that whole assembly?

 

[zdas04]

I would be really nervous about a flow conditioner that only shows ISO 9001 certification. AGA-3 has a protocol for verifying that a flow conditioner does what it says, and I couldn't find anything on their web page that indicates that they have passed that protocol.

As to the rest of it, gas flow proving is not something that is done that much (that is primarily a concept from liquids). I wrote an paper along with Ron Beaty a few years ago. I don't have a copy of the actual paper on my web site but the presentation we did at Southwest Research is on my Web page

David Simpson, PE
MuleShoe Engineering

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

 

[BigInch]

Rather than use provers for gas, it is more common for each fiscal entity to have their own meters and argue about the differences in readings.

Kind of makes it that much more important to have well conditioned flows. Strive for 20 pipe diameters.

I hate Windowz 8!!!!

 

[LittleInch]

I agree it's not common to do this in the field - I've seen fiscal UT meters sent to national calibrated meter loops to get the official seal of approval, but field proving is not that common.

For me I would go for putting the SAO upstream of the UTM, with the best flow conditioner (plate with lots of holes or set of tubes) you can get which you then leave in. Then you are comparing like with like and even if the flow conditioner creates some sort of disturbance, it will be repeatable across the flow range with or without the SAO meter.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way