Ultrasonic Flow Meter/Crude Oil 1

[DaEarl]

Greetings,

I am currently working a project where flowrate needs to be measured. My question is whether anyone has any experience with ultrasonic meters. I have heard many different opinions on these meters. Some of my colleagues are sure of their reliability and others are very skeptical. Are ultrasonics reliable for old pipes that are used for transporting crude oil? Measuring flow in older pipes worry me. I am more worried about repeatability rather than accuracy. My other option would be to install an insertion style vortex shedding meter.

Kyle

 

[25362]

Fot the peculiarities of these meters, see, for example,

http://www.omega.com/prodinfo/ultrasonicflowmeters.html

 

[zdas04]

Ultrasonic meters do a pretty good job of calculating mass flow rate in liquids. Oh yeah, you're selling volume. No problem, if you assume that the density is constant then the conversion is simple and accurate. If the make-up of your crude can change from moment to moment then your volume flow rate will never be "perfect" and often it won't even be close.

I don't often recommend ultrasonic meters for applications with varying density (like most raw "crude" streams), but the only thing that really works is tank gauging and run tickets to measure volume and changes in volume directly.

David

 

[jmw]

Depends on the accuracy you need.
Ultrasonic were promoted as a concept meter by GE many decades back but it is only recently that the technology has caught up with the application.
If you are simply wanting a rough and ready flow rate measurement then you could use a clamp on meter and try before you buy.
I am assuming, from your comments about the old pipe that this is your intention, if you want more accurate measurement then you will have to cut into the pipe and fit a meter spool piece.

For fiscal measurement you may need to consider a multi-chord system which enables the measurement to compensate for the flow profile.

Some systems claim to measure density but you can always use a vibrating tube or fork density meter to measure the density independently.

JMW

http://www.ViscoAnalyser.com

 

[JLSeagull]

The accuracy may be between one and two percent. The transit-time ultrasonic meter turndown is high, about 100:1. Two different technologies exist with nearly mutually exclusive applications. You are likely looking at transit-time a.k.a. time-of-flight. These meters work well without any bubbles or particulates. The other technology is Doppler that requires bubbles or particulates. As suggested the representatives may use a portable meter to assure that the technology selected works on your stream before buying. Vendors include Siemens Controlotron, GE Panametrics, ...

 

[TD2K]

Wouldn't the accuracy be better than that for a multipath flow meter? They have now been approved for custody transfer applications.

 

[JLSeagull]

Two common accuracy statements exist. One is percentage of full scale. This is common for head type meters with a square-root relationship between flow rate and differential pressure. The other is percentage of flow rate.

I lack custody transfer experience with ultasonic meters. My applications were the external clamp-on non intrusive meters. Two percent of anything whether full scale or rate is reasonable with such an installation. However, non head type meters are normally stated in percentage of actual flow rate. This is a function of the high turndown as compared to a dp transmitter across an orifice plate. Orifice plate technology is used for custody transfer because it is well understood - not because it is precise.

 

[MortenA]

There is an article on the subject in the most resent number of "Hydrocarbon processign" (august 2006).

Best regards

Morten

 

[monaco8774]

My company has a lot of these in crude oil service here is our experience:

They are not accurate and only allow trending if you get a signal (see below)
They do not work (no signal) when the pipe doesnt run full if Froude number is below 1
If you get the time of flight type, they do not work (no signal) if you have to much gas in the line downstream of bends or after control valves or with any heavy crudes where you get moose carry under
If you get the doppler type they dont work (no signal) if you dont have enough particulates or gas in the flow and they dont work if you have too much.
The only place I would use an ultrasonic if I had any choice would be on flare gas as you need so much turndown, but even then I wouldnt rely on the reading and ensure you get P,T compensation and ensure the guage is absolute.

 

[chemeng1357]

these fowmeters make error if there is pulsation in flow,
temperature variations and for low boiling point liquids,
we have trouble with methylene chloride line

 

[Ashereng]

The clamp-on style of ultrasonic also has one more inherent problem that is sometimes overlooked. If the pipe wall deteriorates, so does the signal.

For crude oil measurements, my clients typically use coriolis meters. We have installed coriolis meters for both pipeline/transmission and booster station applications.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[JLSeagull]

Coriolis meters are well suited to small line sizes. Accuracy is excellent.

Other technologies including ultrasonic transit time meters are a reasonable consideration for larger lines such as 24 NPS pipe. The permanent meters are better than clamp-on meters. The external clamp-on style meters have their place. Some rare fluids are so corrosive that only external measurement will work. Some temporary measurements are required where the installed stuff is broken or a flow is required and the line cannot come out of service. Any ultrasonic meters that might be used for custody transfer would surely be permanently installed. Orifice runs and most flow technologies are affected by changes in the pipe diameter due to buildup, corrosion, errosion, etc.

Plenty of newer head type meters are worthy of consideration. Examples include the V-Cone or FlowPak.

 

[Ashereng]

We have installed coriolis meters for custody transfer on 16" transmission lines. We have 2 4" coriolis meters in 2 separate parallel meter runs at the station. We have a third run build, but blinded, for future expansion.

Recently, I have noted that E&H has a 10" coriolis meter. That is a big coriolis meter. Not sure how much that would cost.

I would think a couple of those would be good for 24" P/L.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[unclesyd]

Here is another ultrasonic meter company you might give a call to.
We used a couple of their meters with good success, not in crude. I have a old colleague who trying out the water in oil meter.

http://www.eesiflo.com/