Mass flow through Venturi tube 3

[flinana]

Hi,

I have a Venturi tube installed on a superheated steam line. In order to calcula the mass flow I have a couple of questions.
1. does the transmitter have to extract the square root or linear and then which formula would need to be applied to read mass flow and not volumetric flow?


thanks

 

[zdas04]

A Transducer is sensing a physical parameter and transmitting it. To use a Venturi meter you will have to process static and differential pressure in an equation that happens to include a square root. You still have to do the math.

The equation that allows you to infer a flow rate from the differential pressure across a known restriction provides volume flow rate. To convert that to mass flow rate you only have to multiply times density. Not terribly difficult.

David

 

[flinana]

Ttanks, The transmitter provides the signal with the square root extracted so that the do is proportional to the flow. This flow needs to be compensated wiith P and T since it is superheated steam. As far as I understand you are saying that all I have to do is multiply the signal from the transmitter by the density which constantly changing. It might sound silly but this is the part where I get confused since I guess I need to somehow introduce the steam tables right?

 

[Thealanator]

You might check out the Contrec Model 415 Gas Flow Computer. It supposedly contains equations for saturated and superheated steam. There may be other flow computers besides Contrec, but I am using this model for gas flow measurement.

 

[flinana]

The problem is that on our Steam Generator there is a mass flow meter installed which does not coincide with the venturi tube installed on the HP header. For example, when the SG is measuring 199 tn/h the venturi measures 215 tn/h.

The only difference is that the mass steam flow calculation for the Venturi is done with a DCS function attached while the SG has an orifice with a transmitter ehich corrects the steam flow for P and T.

So basically I woul like to make sure that we are using the function correctly.

 

[hacksaw]

go to your steam tables, use the internet, or talk with you mechanical engineer, to give you the steam density used for the original meter calculations, and the steam density for the new case

you may need to consult the plant record books for the current design basis

according to the writeup you posted, the scale factor should be multiplied by sqrt(rnew/rold) or simply replace r by the new but actual steam density in units required by your computation block.

updating meter factors for changed plant conditions is pretty normal stuff

since the venturi is reading high, it sounds like the operating pressure is lower than originally expected, or the steam temperature higher

 

[flinana]

thanks.
I am still not sure whether the fscale value i use is correct, the venturi datasheet says a max flow of 250 tn/h and i use it a fscale, the rest of variables are ok.
I use the datasheet parameters on the formula. The densiy I calculate it using a macro in excel called water97,

http://www.cheresources.com/content...-and-transport-properties-of-water-and-steam,

to calculate density at operating point.


Please find attached the venturi datasheet. Could you please let me know whether I am using the datasheet correctly?
Does it really matter what the max flow is s long as the max dp in venturi is changed accrodingly? Should I be using operating values inthe formula or maximum values?

thank you very much.

 

[hacksaw]

The spec sheet is for an orifice plate with flange taps as opposed to a venturi.

I get virtually identical results steam data, discharge coefficient, etc. for the given condtions. What you have to confirm is the current pressure and temperature (at least the average readings). The get a reading of the actual transmitter differential in mmH2O and the indicated flow in t/hr at the same instant.

 

[flinana]

sorry you are right, we use an orifice on the HP header and a Venturi on the LP header.

The transmitter gives an output in mmH20, the DCS does the conversion to t/hr.

For example with these values,
T= 376,9 ºC
P= 99 barg
dp= 155 kPa

The DCS function outputs 214 tn/h while te steam generator which has a rosemount Annubar flowmeter outputs 199 tn/h (there are two steam trains each showing 28 kg/s, so approx 199 tn/h)

This difference is what I dont understand.
I also attach the calc for the steam generator annubar.

 

[hacksaw]

You might need to confirm that your XMTR span is 0-22000 mm and not 0-25000 mm



FLOW METER SIZING CALCULATION

Tag no
Service
Fluid Steam
Meter Square Edge Orifice Plate
Connections Flange Taps


Pressure 99.000 Barg(100.013 Bar(A))
Temperature 378.0 Deg C
Ambient Pressure 1.013 Bar(A) ( 0 Meters Above msl)
Base Conditions (STP) 1.013 Bar(A) , 0.0 Deg C
Specific Volume (Vf) 0.0248 M3/kg
Isentropic Exponent 1.271
Viscosity 0.025 cP

normal | maximum
Flow 196000 | 250000 kg/hr
Pipe Reynolds No. (Rd) 12115205 | 15453067
Mean Fluid Velocity (Pipe) 32.4 | 41.4 m/s
Line Loss 1.27 | 2.07 (Bar)/100 m
Friction Factor (f) 0.0138| 0.0138
Meter Loss(Installed) 0.61 | 1.00 Bar
Meter Velocity 58.3 | 74.4 m/s


**** METER CONSTANTS ****

Meter Bore (d) 6.7587 @ 20 Deg C
Pipe i.d. (D) 9.0618 Inches
Beta Ratio (d/D) 0.7458 Inches/Inch Caution
Metal Expansion (Fa) 1.0126 (316 Stainless Steel)
Expansion Factor (Y1) 0.9946
Reynolds Factor (Fc) 0.9593
Drain/Vent Corr. (Fh) 1.0000
Discharge Coef. (Cd) 0.59547
Spinks Factor (S) 0.41553


**** TRANSMITTER DATA ****

normal | maximum
Meter Differential (h) 13448.0 | 21878.9 mm w.c.
Flow Rate 196000 | 250000 kg/hr


W (kg/hr) = 8.121 * S * D^2 * Fa * Fc * Fh * Y1 * sqrt( h / Vf )

 

[flinana]

The span is correct and set to 22000 mm w.c.

With a value of 15809 mm w.c. and your formula I obtain 215 tn/h so it seems to coincide with my calculation. I guess maybe the error is in the pitot on the Steam generator?

What is the difference between the value in **** TRANSMITTER DATA **** of 21878.9 mm w.c. and the span of 22000 mm w.c.? I guess none.

I need if possible the formula for the LP flowmeter, this is the next strange thing, is that the flow in HP is 215 tn/h and the LP is 183 tn/h, how can there be so much loss in steam between Hp and lp.

Op T: 325 ºC
Op P: 16,4 barg
Attached is the Venturi data sheet.

What SW are you using for the calculation and to derive the formula? Is it Flowcalc32?

btw: Remind me to include you in my Christmas cards lists!! you are awesome, thank you.

 

[hacksaw]

Meter reconciliation is always tough, you have to check all the details, including the actual installation

You haven't lost steam, unless you've an extraction turbine, usually it is the changed process conditions. Often the LP steam will have wildly varying temps depending on the turbine load, with the higher temps at reduced load

the sw is custom developed, but used in lots of meter surveys to sort out the non-sense and keep suppliers in line. Uses the equation of state for steam/condensate.

hopefully with attachments

 

[hacksaw]

FLOW METER SIZING CALCULATION


Tag no
Service
Fluid Steam
Meter Venturi W / 15 Deg Exit Cone


Pressure 17.000 Barg(18.013 Bar(A))
Temperature 378.0 Deg C
Ambient Pressure 1.013 Bar(A) ( 0 Meters Above msl)
Base Conditions (STP) 1.013 Bar(A) , 0.0 Deg C
Specific Volume (Vf) 0.1623 M3/kg
Isentropic Exponent 1.289
Viscosity 0.024 cP

normal | maximum
Flow 183000 | 200000 kg/hr
Pipe Reynolds No. (Rd) 5230769 | 5716687
Mean Fluid Velocity (Pipe) 39.7 | 43.3 m/s
Line Loss 0.11 | 0.14 (Bar)/100 m
Friction Factor (f) 0.0121| 0.0121
Meter Loss(Installed) 0.02 | 0.03 Bar
Meter Velocity 87.5 | 95.6 m/s


**** METER CONSTANTS ****

Meter Bore (d) 13.6450 @ 20 Deg C
Pipe i.d. (D) 20.2618 Inches
Beta Ratio (d/D) 0.6734 Inches/Inch
Metal Expansion (Fa) 1.0126 (316 Stainless Steel)
Expansion Factor (Y1) 0.9918
Reynolds Factor (Fc) 1.0000
Drain/Vent Corr. (Fh) 0.0000
Discharge Coef. (Cd) 0.99349
Spinks Factor (S) 0.50554


**** TRANSMITTER DATA ****

normal | maximum
Meter Differential (h) 1920.3 | 2293.6 mm w.c.
Flow Rate 183000 | 200000 kg/hr


W (kg/hr) = 8.121 * S * D^2 * Fa * Fc * Fh * Y1 * sqrt( h / Vf )

 

[hacksaw]

FLOW METER SIZING CALCULATION


Tag no
Service
Fluid Steam
Meter Venturi W / 15 Deg Exit Cone


Pressure 16.400 Barg(17.413 Bar(A))
Temperature 325.0 Deg C
Ambient Pressure 1.013 Bar(A) ( 0 Meters Above msl)
Base Conditions (STP) 1.013 Bar(A) , 0.0 Deg C
Specific Volume (Vf) 0.1527 M3/kg
Isentropic Exponent 1.293
Viscosity 0.022 cP

normal | maximum
Flow 183000 | 200000 kg/hr
Pipe Reynolds No. (Rd) 5740138 | 6273375
Mean Fluid Velocity (Pipe) 37.3 | 40.8 m/s
Line Loss 0.11 | 0.13 (Bar)/100 m
Friction Factor (f) 0.0121| 0.0121
Meter Loss(Installed) 0.02 | 0.03 Bar
Meter Velocity 82.3 | 89.9 m/s


**** METER CONSTANTS ****

Meter Bore (d) 13.6450 @ 20 Deg C
Pipe i.d. (D) 20.2618 Inches
Beta Ratio (d/D) 0.6734 Inches/Inch
Metal Expansion (Fa) 1.0106 (316 Stainless Steel)
Expansion Factor (Y1) 0.9920
Reynolds Factor (Fc) 1.0000
Drain/Vent Corr. (Fh) 0.0000
Discharge Coef. (Cd) 0.99354
Spinks Factor (S) 0.50557


**** TRANSMITTER DATA ****

normal | maximum
Meter Differential (h) 1813.1 | 2165.6 mm w.c.
Flow Rate 183000 | 200000 kg/hr


W (kg/hr) = 8.121 * S * D^2 * Fa * Fc * Fh * Y1 * sqrt( h / Vf )

 

[flinana]

It is a 50 MW Siemens extraction turbine. Do you sell that sw by any chance? Its amazing.

 

[hacksaw]

there you have the disappearing steam, i'll give the notion some thought

by the way, good luck on sorting the meters out, it is always a lot of hard work and fun

 

[flinana]

Thanks, please if you ever consider selling your program, let me know.

 

[flinana]

btw you mention that Drain/Vent Corr. (Fh) 0.0000, if this is true then the equation is 0.

 

[hacksaw]

thanks, should be shown as 1 minus the correction

 

[flinana]

Btw, in case you are interested this has to do with 3 50MW CSP solar power plants.