3051S ASP Pressure Pulsation

8/12/2019 3051S ASP Pressure Pulsation

1/4


2/4

White paper: Pulsation Induced Flow Measurement Error DiagnosticJanuary 2008 Page 2

www.rosemount.com

BackgroundEmerson developed a new generation ofDifferential Pressure transmitters with a unique,scalable architecture. This new pressuretransmitter platform, called the Rosemount

3051S, provides best in class performance andreliability and also features advancements inpower management that allows the easyaddition of advanced functionality to the basepressure transmitter. This functionality,embodied in a feature board, can be ordered aspart of the transmitter or simply added to anexisting transmitter already installed in the field.

Making use of this advanced functionality,Emerson has developed a unique patentedtechnology that provides a means for earlydetection of abnormal situations in a processenvironment. The technology, called StatisticalProcess Monitoring (SPM), is based on thepremise that virtually all dynamic processeshave a unique noise or variation signature undernormal operation. Changes in these signaturesmay signal that a significant change in theprocess, process equipment, or transmitterinstallation will occur or has occurred. Forexample, the noise source may be equipment inthe process such as pumps, agitators or thenatural variation in the DP value caused byturbulent flow or any combination thereof.

The sensing of the unique signature begins with

a high speed sensing device such as theRosemount 3051S Pressure Transmitterequipped with patented software resident in aHART Diagnostics or FOUNDATION

fieldbus

Feature Board. This powerful combination hasthe ability to compute statistical parameters thatcharacterize and quantify the noise or variationand represent the mean and standard deviationof the input pressure. Filtering capability isprovided to separate slow changes in theprocess due to intentional setpoint changes frominherent process noise which contains thevariation of interest. The transmitter provides

the statistical parameters to the host system viaHART or FOUNDATIONfieldbus communicationsas non-primary variables. The transmitter alsohas internal software that can be used tobaseline the process noise or signature via adefined learning process. Once the learningprocess is completed, the device itself candetect changes in process noise and willcommunicate an alarm via the 4 20 mA outputor alert via HART or FOUNDATIONfieldbus.

With the design of this transmitter andfeature board complete, Emerson initiated atest program to determine if this technologycould be applied to the prediction of square

root error in DP flow applications.

Test Program DescriptionThe ability to predict pulsation inducedmeasurement error was evaluated usingthree data sets consisting of simulatedmodel-based data, data taken at the MicroMotion facility in Boulder, Colorado and dataobtained using a flow test bench at theEmerson facility in Chanhassen, Minnesota.

The simulated data set was created using asinusoidal model allowing for the inclusion ofup to two noise inputs being able tomanipulate the frequency and amplitude ofeach source over various flow rates. In thismodel, the DP was sampled at 45 msintervals (22 Hz) to mimic the output of atypical transmitter.

The 3051S DP transmitter is bandwidthlimited both mechanically and electronically.Inherent to its design consisting of an oilfilled three spring system, pulsations aboveapproximately 25 Hz are damped out.Configuring the transmitter to sample at themaximum 45 ms rate limits DP sampling to

below 22.22 Hz. It is important to note thatin almost all cases, regardless of thefrequency, the DP transmitter has the abilityto detect pressure pulsations and willtranslate this into changes in standarddeviation.

On December 4, 2006 testing wasconducted at Micro Motions (MMI) inclinedflow test stand. A 405P 0.65 beta orificeplate was used with water at mass flow ratesvarying from 170 to 500 lb/min. Each testrun was performed for a minimum of 60

seconds. The DP signal was measuredusing a 250 Rosemount 3051S DifferentialPressure Transmitter. The transmitter hadFOUNDATIONfieldbus outputs and wasconfigured to output pressure data at themaximum rate of 22 updates per second.The data from the 15 test runs wererecorded using standard PC FOUNDATIONfieldbus tools and analyzed post test usingstandard Excel spreadsheets and MiniTab.


3/4


www.rosemount.com

Additional data were collected on April 23, 2007,at the Emerson facility using a Daniel 0.67 betaorifice and a 3051S DP transmitter configured asdescribed above. A total of 17 test runs were

conducted with flow rates ranging from 3.4 to22.6 lb/min of water each lasting a minimum of60 seconds. As before, the data were collectedusing PC FOUNDATIONfieldbus tools and lateranalyzed with Excel and MiniTab.

Test Results and AnalysisData from all three sources, the processpulsation model, the Micro Motion inclined flowbench and the Rosemount flow loop, wereconsolidated and plotted in Figure 1.Regardless of test setup, flow conditions ornoise characteristics, all data lie on onecontinuous curve. This confirmed relationshipwould allow for the prediction of %SRE using thestandard diagnostic feature board output ofcoefficient of variation.

As shown in Figure 2, the relationship betweenthe coefficient of variation and the AGA-3

pulsation limit value is clear and virtuallymimics that of the coefficient of variation and%SRE. In both instances, pulsation inducederrors are related to an increase in noiserelative to the fundamental pressure signal

(the coefficient of variation) and have beendemonstrated to be an excellent tool torepresent pulsation induced errors.

ConclusionPulsation induced flow measurement errorsare a concern for any flow operation,particularly custody transfers where flowaccuracy is critical. Emerson has identifieda means by which to provide two uniqueindications of process pulsation, %SRE andthe AGA standard for acceptable pulsationenvironment while continuing to output ahighly reliable and accurate pressuremeasurement. This technology is residentin HART and FOUNDATIONfieldbus versionsof the Rosemount 3051S PressureTransmitter, and can provide early warningof process, equipment and installationproblems related to process pulsation.

ReferencesBell, Sr., George. "Lost & Unaccounted Natural Gas Lost & Unaccounted Natural Gas. TheEffect and Control of Pulsation in Natural Gas Measurement." pgiint.com.http://www.pgiint.com/pdf/Lost&UnaccountedNaturalGas.pdf

Durke, Ray G. et al. An Assessment of Technology for Correction g Pulsation-Induced Orifice

For Measurement Errors, PCRC Technical Report No. TA 91-1, 1991.

Edlund, et al. U.S. Patent 4,654,813. "Electronic square root error indicator." 1987

Ifft, Stephen. Test of 8 1595 Conditioning Orifice Plates 2D Downstream of Centrifugal Pump.Emerson Process Management Dieterich Standard, Inc. February 23, 2006.

Orifice Metering of Natural Gas and Other Related Hydrocarbon Fluids, AGA-3 2000, Part 2Specification and Installation Requirements, Fourth Edition.


4/4


www.rosemount.com

Figures

Coefficient of Variation [%]

%SRE

35302520151050

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

MMI

Model

RMT

Source

Scatterplot of %SRE vs Coefficient of VariationGrouped by Data Source

Figure 1: %SRE Correlation with the Coefficient of Variation

Coefficient of Variation [%]

AGA-3PulsationEquation[%]

35302520151050

5

4

3

2

1

0

Scatterplot of AGA-3 Pulsation Limit Equation vs Coefficient of Variation

Figure 2: AGA-3 Pulsation Equation Correlation with the Coefficient of Variation

Documents

3051S ASP Pressure Pulsation