To Calibrate or Not to Calibrate a fieldbus transmitter? Dale Perry Pressure Marketing Manager Rosemount

Transcription

1 To Calibrate or Not to Calibrate a fieldbus transmitter? Dale Perry Pressure Marketing Manager Rosemount

2 Introduction Digital Transmitters are here Wireless Fieldbus Same architecture as Smart-HART transmitter but without the A/D component Traditional methods of calculating accuracy apply although most vendors don t explicitly explain it Calibrating Digital Transmitters is easy 27-Jun-01, Slide 2

3 Agenda Digital Transmitter Theory of Operation Accuracy Types Performing the calibration New at the Foundation 27-Jun-01, Slide 3

4 Basic Block Diagram of fieldbus Transmitter Ff Electronics Resource Sensor Correction & A/D Sensor Sensor Transducer AI Function Block Function Block 27-Jun-01, Slide 4

5 Sensor Correction is Called the Characterize/Verify (C/V) Process CHARACTERIZE Set Temperature Apply pressures Store Data Set Temperature Apply Pressures Store Data. VERIFY Set Temperature Apply pressures Compare Characterize data Set Temperature Apply pressures Compare Characterize data. Sensor Characteristic Data Sensor Correction 27-Jun-01, Slide 5

6 A/D Converts a Measured Variable to a digital value Converts an Analog signal infinite number of values To a digital value Discrete Number of values Number of digital values determine sensor resolution Pressure Lower Range Limit A/D Upper Range Limit Pressure PSI Infinite Values Discrete number of Values 27-Jun-01, Slide 6

7 A/D is a Generic Term H L C 3 C 4 C 1 C 2 C/D Micr Capacitance to Digital V/D Micr Voltage to Digital 27-Jun-01, Slide 7

8 Output of Sensor Transducer Block is Channeled to an AI Block Ff Electronics Sensor Correction & A/D Resource Sensor Sensor Transducer AI 27-Jun-01, Slide 8

9 Agenda Digital Transmitter Theory of Operation Accuracy Rangedown Accuracy Types Calculating Accuracy Total Probable Error Performing the calibration 27-Jun-01, Slide 9

10 Knowing When to Calibrate Requires a lot of Information Accuracy Reference Accuracy Linearity, Hysteresis, Repeatability Accuracy type Accuracy at rangedown Total Probable Error Accuracy, Temperature effect, Static Pressure effect Long Term Stability Accuracy Tolerances Test Equipment 27-Jun-01, Slide 10

11 Accuracy is Typically Stated in One of Three Categories 10:1 100:1 %% Full Scale Accuracy wholly dependent on Full Scale Value Error is significant at low end Inexpensive transmitters 27-Jun-01, Slide 11 10:1 100:1 % Calibrated % Full ScaleSpan Accuracy dependent on Range Error gets smaller at Rangedown Standard Rosemount accuracy Standard high end units 10:1 100:1 %% of Full Reading Scale Accuracy dependent on Reading Reading gets smaller, Error gets smaller. Commonly used in calibration equipment Very accurate at low end

12 Reference Accuracy Includes Linearity, Hysteresis, and Repeatability Pressure Range Upper Range Limit Sensor Output Lower Range Limit Accuracy Linearity + Hysteresis + Repeatability Pressure 27-Jun-01, Slide 12

13 Calculating the Rangedown is Simple HART/Analog Sensor Rangedown is the Ratio between the sensor Upper Range Limit and the calibrated Span Upper Range Limit Calibrated Span = RangeDown Digital Transmitters Sensor Rangedown is the Ratio between the sensor Upper Range Limit and the calibrated Range Upper Range Limit Calibrated Range = RangeDown 27-Jun-01, Slide 13

14 Calculating the Rangedown is Simple Sensor Rangedown is the Ratio between the sensor Upper Range Limit and the calibrated Span 3051S Range 2 Upper Range Limit Calibrated Span Example 250 InH InH20 = 100 Span= 0.5bar = RangeDown Rangedown 100: bar 0.5 bar = :1 27-Jun-01, Slide 14

15 Accuracy Statements are Typically Only up to 10:1 Calibrated Range Typical Accuracy Statement: ±0.055 % of span. For Spans less than 10:1 URL ±[ %( )]% of span Span 10:1 ±0.055 % of span 1: in H20 62 mbar -250 in H bar 27-Jun-01, Slide 15 See full range but accuracy Depends on last calibration in H bar

16 Three areas need to be investigated to determine Total Performance Total (Installed) Performance Accuracy of measurement at installed conditions Reference Accuracy Temperature Effects Line Pressure Effects Long Term Stability Accuracy of measurement over time Dynamic Performance How fast the measurement tracks process changes 27-Jun-01, Slide 16

17 Installed Performance can be Calculated by Using the Total Probable Error Equation Accuracy at Calibrated Range Installation effects On Accuracy 27-Jun-01, Slide 17

18 Example: Total Probable Error What is TPE in psig, given: Accuracy = ± 0.1% of span Total temperature effects = ± 0.5% of span LP effects = ± 0.3% of span Calibrated span is 5 psig TPE = = ± 0.59% of span TPE = x 5 psig = ± 0.03 psig 27-Jun-01, Slide 18

19 Determine Calibration Frequency needed to Maintain Required Accuracy Maximum Error Long Term Stability Long Term Stability Accuracy Total Probable Error Published Accuracy Time 27-Jun-01, Slide 19

20 Agenda Digital Transmitter Theory of Operation Accuracy Performing the calibration New at the Foundation 27-Jun-01, Slide 20

21 Digital Transmitters Require One Calibration Ff Electronics Sensor Correction & A/D Resource Sensor Sensor Transducer AI 230 inh20 27-Jun-01, Slide 21

22 Calibration Process Select Accurate Test Equipment ANSI Suggest that Calibration Equipment be at least Four times More Accurate than the Device Calibrate at Temperature Perform Zero Transmitter (optional) Perform Lower Sensor Trim Should be at or below typical lower end of operational range Perform Upper Sensor Trim Should be at or above typical upper end of operational range 27-Jun-01, Slide 22

23 Sensor Zero Trim Zeros out small linear offsets due to mounting or static pressure effect. 27-Jun-01, Slide 23

24 Sensor Trim (Lower & Upper) Sensor Trim is a two-point sensor calibration where two end-point pressures are applied, and all output is linearized between them. Lower Sensor Trim is an offset Shifts entire sensor characterization curve up or down. Upper Sensor Trim changes the slope of the sensor characterization curve. Pivots on the low trim point. Note: Always perform the low trim first! 27-Jun-01, Slide 24

25 Sensor Trim (Lower & Upper) Start 150 Reading 0 0 Pressure Input Low Trim Reading 0 Actual Your test equipment/ideal 0 High Trim Pressure Input Reading 0 27-Jun-01, Slide 25 0 Pressure Input 150

26 Calibration Documentation Last Calibration history stored in transmitter Maintenance system As Found As Left Calibration History 27-Jun-01, Slide 26

27 Agenda Digital Transmitter Theory of Operation Accuracy Performing the calibration What s New for Calibration 27-Jun-01, Slide 27

28 New for Calibration of Ff Sensors Transducer Block Standardization Human Centered Design User Interfaces 27-Jun-01, Slide 28

29 New at the Foundation Sensor Transducer Block Standardization Easier to make Ff calibrators. FF-903 Pressure FF-904 Temperature FF Positioner 27-Jun-01, Slide 29

30 Traditional Technology Driven User Interface Block Mode Parameter 27-Jun-01, Slide 30

31 Human Centered Design User Interfaces Task Based 27-Jun-01, Slide 31

32 Human Centered Design All Calibration Information Accessible 27-Jun-01, Slide 32

33 Summary Digital transmitters do not have an analog output that has any correlation to the sensor input Digital transmitters use the Calibrated Range when determining rangedown Look carefully at the reference accuracy statement Total Probable Error should be used when determining calibration frequency The simulator used for calibration should be at least four times better than the accuracy of the transmitter Human Centered Design User Interfaces makes calibration of Ff sensors easy. 27-Jun-01, Slide 33

34 27-Jun-01, Slide 34

35 Where To Get More Information American National Standard ANSI/CSL Z Calibration Laboratories and Measuring and Test Equipment PlantWeb University Emerson Process Management Education Services 27-Jun-01, Slide 35