NATURAL GAS MEASUREMENT

[VictorMVA]

Hi,

I´m going through a gas flow measurement loop and am a bit confused about the way it´s configured. The flow is measured with an orifice plate (Pd = 8 kg/cm2 man, Td= 37.8°C)and the associated transmitter outputs the flow in ft3/h; the actual pressure and temperature are measured with other instruments and the corresponding units are kg/cm2 man and °C. The flow conditions differ from plate design conditions and the final units for flow must be SCFH. In DCS, the only formula is

F = Fact * SQRT[(P_act/P_d)*(T_d/T_act)]

where F = flow in SCFH
F_act = flow in ft3/h from transmitter
P_act,T _act= pressure and temperature at actual conditions (6.5 kg/cm2 man, 30°C)
P_d,T_d= design pressure and temperature for the orifice plate

However there is no values for standar pressure and temperature (14.73 psia and 60°F) considered in the formula. What would be the right sequence of calculations to get the correct flow in standard conditions?

 

[zdas04]

I would be more concerned that there is no dP or compressibility. The flow factor probably takes care of the conversion. I don't like to see it done this way, but it more or less works.

David Simpson, PE
MuleShoe Engineering

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

 

[VictorMVA]

Correcting the above information, the transmitter output is DP and the extraction is made in the DCS, then the flow is used in the formula.Seems were not considered the conversion from actual to standard conditions after correcting the flow which needs another calculation, is this correct?

 

[zdas04]

I'm thinking that if the transmitter is doing the heavy lifting, then the "D" subscripts are not "Design" but "Designated" or "Standard". Without the SQRT (and with a compressibility term) then ACF times P_actual/P_std*T_std/T_actual would give you flow rate at standard conditions. With the square root it gives you junk. You need to be really careful of units (14.73 psia is 1.03562 kgf/cm²) but without the square root it is nearly right.

David Simpson, PE
MuleShoe Engineering

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

 

[VictorMVA]

David,

I think that the formula is for compensating the diference in operating conditions respect the design T and P for orifice plate calculation. If so, then the compensated flow must be "standardized" with the formula that includes the compressibility factor.

 

[zdas04]

If you know, why ask? If the device calculates ACF, then no compensation for deviation from design is required, desirable. If it is calculating something else then maybe. Who is the software vendor for your transmitter and DCS?

David Simpson, PE
MuleShoe Engineering

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

 

[RPilch]

The transmitter only puts out a signal proportional to the delta P across the orifice plate. The delta P is converted to a gas volumetric flow in the DCS (normal, standard or actual) based on an orifice plate design calibration. In the flow correction equation you provided, the square root of the pressure and temperature ratios (compared to the design conditions at which the flow calculation was calibrated) is used to account for the change in gas density at the meter (based on the ideal gas law). The problem I see is that the pressure and temperature ratios inside the square root must use absolute pressure and temperature. Assuming MAN means manometer or gauge reading, then atmospheric pressure must be added to the measured and the calibration pressures. In the same way, the appropriate temperature conversion factor must be added to both temperatures.