Orifice flange taps drilled at more than 1" distance from the flange face

[svi]

A pair of 4" class 1500 RTJ WN ASME B16.36 Orifice flanges, the tap location was drilled in error at 34 mm vs the 24 mm required from the flange face. The orifice plate is held in a RTJ carrier. The machinist tried to center the tap over the flange thickness. The Orifice meter is to be used to measure the flow of injection water. The metering vendor has said the change in dim from the B16.36 std will affect the calibration of the orifice meter. I could read off the net that the 1" dimension from the orifice plate has come about from the study done by AGA and ASME.

With today's availability of CFD and related software tools, is it difficult to correct the calibration for this additional 12 mm. From generic pressure profiles across an orifice plate the d/s pressure recovers towards the end of the flange hub.

 

[danw2]

I don't know, but it appears that someone did a study on it:

Link
Abstract:
The effects of errors in the location of the flange taps on the discharge coefficient of orifice plates have been invesgated experimentally. A specially designed fixture by which it was possible to change upstream and downstream tap locations independently was used to study the effects of errors in each location separately. The effect of plate displacement was investigated by making appropriate changes in both upstream and downstream tap locations. Three orifice plates of β = 0.4, 0.6 and 0.75 were tested. The study covered the effects for 51 mm (2 in) and 102 mm (4 in) meter runs. Full-scale tests were conducted for the 51 mm case while model tests with a scale ratio of 1 : 2 were conducted to simulate the 102 mm (4 in) case. Calibrations were done against a dynamic weigh tank system while differential pressures were measured by a number of D-P cells interfaced with a laboratory computer. The results indicate that the effects of tap location errors exceeding those allowed by the standards are negligible in some instances.

Is the cost of a copy of the study less than replacing the flange(s)?

 

[svi]

When an orfice meter vendor does a sizing, do they use only emperical calculations? Are mathematical simulations such as CFD, off limits to them. What kind of uncertainty does the 2 have vs the experimental study.