Orifice Plate measurement error - entrained liquid

[invalauble]

I'm measuring a natural gas stream with an orifice plate dp meter. Occasionally, slugs and gurgles of water will come through the meter (not on purpose, but it happens). We separate the liquids from the gas at the other end and quantify them - I know how much water came through.

I know these liquids are throwing off the true gas quantity in the meter, but I don't know how to correct for them. Is there an industry standard or an accepted correction method? I've heard:

100 bblswaterperday = 728000 scfd,

but that seems too static to be true at all flowing pressures. Thanks in advance for your help.

 

[zdas04]

Squared-edged orifice meters are "inferential" devices. They allow you to infer a volume flow rate from unrelated (but easily measurable) parameters (P, dP, and temp),

Don't ever assume that this inference can be extrapolated. An infered measurement that has a different fluid than you've input to the calculation will simply be wrong in some direction for some magnitude (neither the direction or magnitude can be assumed for a given event). You cannot, ever, do the obvious procedure of: (1) convert the indicated volume flow rate to mass flow rate; (2) subtract the mass of the liquid; and (3) convert the remainder back to volume flow rate. This would be wrong.

There is no way to estimate gas flow in the example you gave. Any technique you chose to use will be wrong. The more complex the method, the more effort you have put into a wrong answer. The technique I used when I had responsibility for EFM editors is to delete the section of the "chart" that had known slugs and fill in the missing data by "connecting the two ends". This result is wrong, but pretty easy.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[m777182]

If water is present as moisture in the gas then there is a way to measure the moisture content of the gas and make a correctionof the flow; with fine droplets of water it will not work.
m777182

 

[sdl]

Invaluable:

This is a tough one. It almost sounds like the meter is at the wrong place. Water will accumulate at the plate and will always create problems with the metering. Is it possible to move the meter downstream after the liquids are separated?

If this is not possible, then there are some things that you can check with the installation:

Are the sample lines sloped towards the process? There should not be any pockets. If it is a tapped flange, the sample lines should come off the top (with the meter being higher than this elevation). Hopefully the area is warm enough to prevent freezing. If all else fails, you can put in knock-out pots in each sample line and have an operator blow them down daily (or more often if you are getting that much liquid). This could be automated if you have the money. This is not a fix for the water problem, but it will help with the metering issues.

The only other alternative is a mass flow meter or something similar.

sdl

 

[invalauble]

Thanks sdl. The meter is downstream of a 3rd party separator. Our problem is liquid carryover from the separator. In essence the meter shows more gas than they actually sent, since some of the time liquid droplets and slugs were going through with the gas. They don't maintain their separator very well.

Sounds like short of scrubbing the 1-second EFM data by hand (it's not a chart recorder) or shutting in the flow when the separator carries over, we'll just have to tough it out.

 

[sdl]

invaluable:

I know your exact situation very well. A couple of questions for you:

Do you have control over the third party separator? If so, then it is simple, shut them in on high level. I am also assuming that you are the processor, so you should have an agreement in place to have them shut in themselves. It sounds like the separator is not large enough. The outside producer should be limiting the flow when they pig (I am assuming that this is where the slugs come from), to give the slug catcher time to do its job. This is best done upstream through a manual valve (HIC or manual control) or could be done with a flow (with pressure override) control loop.

The real fix is a new separator (either replacement or series) to prevent the liquid carry over. I really don't know of a way to measure gas flow accurately with liquid in the line. A different style of meter might help.

The liquid carry over cannot be good for the downstream process. The liquids can cause hydrate problems as well. Personally, I would go back and review the agreement and force proper pigging/level control procedures on them.

Good luck!
sdl