Comparing Reg. Orifice Plates to orifice plates w/ flow conditioning

[Siciliano376]

Folks,

I am trying to collect some cost data to compare regular orifice plates to orifice plates with flow conditioning (single plate/multi holes).
I have estimates a reduction in straight pipe lengths of about from 20 diameter (upstream) to 2 d which is giving me considerable saving on large bore lines (24" and 20").
The information that I am missing is:
The additional cost associated with installing conditioning orifices in place of regular orifice.

Can anyone point me to any letterature where I could find this kind of information.

Thank you.

Corradino

 

[zdas04]

Have you even looked at AGA 3 (or ASME 14.3 it is the same thing)? Flow conditioners will both allow you to have shorter tubes (a minor cost savings), but will also reduce the uncertainty on you measurement (a huge benefit forever). They will not allow you to shorten the tube down to 2D, someone is misunderstanding. Maybe they are thinking of the Savant system (the old Gallagher design) that has a short API straightening vane on one end and a perforated plate on the other end of a 2D spool, but that spool is 10D upstream of the plate.

Is your company really allowing someone to design a 24 inch meter station based on hallway rumors? I've rarely heard a scarier concept in gas measurement.

David

 

[Siciliano376]

David,

Thank you for the reply.
The orifice plates will used as part of a process control system both in gas lines and liquid lines.
I picked up the 2D from reading vendors websites. Emerson advertises that their meters only require 2D upstream and 2D downstream

thnx

 

[agameterman]

Corradino, The conditioning Orifice you are talking about requires a small foot print but the accuracy is not good at all. We are currently involved in a program that is replacing them in a steam application because they didn't preform well. I would recommend a cone meter or Venturi as an alternative to an orifice meter as they will be .05%< 1.05%. AGA 3 & API 14.3 recommends upstream lengths and most upstream sections have gotten longer with no flow conditioner @ 145d as well as a Vane now @ 17d-29d depending on allowable Beta. The Gallagher and CPA will allow you a shorter upstream section ( Catch all design ) usually 13D total with .075% accuracy where a conditioning plate accuracy is 5%<40% and is looking for 2d. The cone or Venturi will require 2D to no upstream pipe required the Venturi can be machined out of almost any material unlike the Orifice meter run. Hope this helps.

 

[Bubbarock]

You are willing to throw away accuracy for some savings of pipe?

Blake Patras
Sales Manager
Miquelon Meter Services
Edmonton, ALberta

 

[Niwot]

Keep in mind that there is no governing body that can prevent a company from making flow measurement accuracy claims, whether it be a 4-hole O-plate, cone type, etc. Therefore, you must rely on the manufacturer for proof of accuracy. Has the meter been INDEPENDITALLY tested at an NIST traceable flow laboratory? To what line sizes? How are flow coefficients derived for meters that fall out of the scope of testing? "Educated" guess, extrapolation, etc.?

Also, keep in mind that meters claiming to require no straight-run are dependent on beta ratios that allow for maximum flow conditioning. The data I've seen on the 4-hole O-plate is limited, both in pipe size and beta ratio.

We use API 22.2 flow testing protocol for DP devices to show accuracy variations in baseline testing (long straight-run), swirl induced profiles (2 elbows out of plane), and half moon orifice plates (mimicking a half-open gate valve).

Regardless of what meter you choose, I agree with David's statement above about "hallway rumors". Ask for proof - especially in large gas lines - that an accuracy statement is more than just a manufacturer's claim.

Measure it once, you have measurement. Measure it twice, you have an argument. Since you usually don't have the means to "measure it twice", verify the pedigree up front. I understand that it is not practical to independentally test every meter size (thus, the trust that has been placed in standard orifice plates and their reams of empirical data), but I would also ask the manufacturer to back up a claim of .5% accuracy in a 24" gas line with no straight-run and see what they come back with (especially how their K's are derived).

Just my 2 cents.