![Campbell CR800 Series Скачать руководство пользователя страница 352](http://html1.mh-extra.com/html/campbell/cr800-series/cr800-series_operators-manual_489099352.webp)
Section 8. Operation
352
• Estimating Measurement Accuracy for Ratiometric Measurement
Instructions.
The following topics discuss methods of generally improving voltage
measurements. Related information for special case voltage measurements
(
thermocouples
(p. 331),
current loops
(p. 344),
resistance
(p. 332),
and
strain
(p. 343))
is
located in sections for those measurements.
Single-Ended or Differential?
Deciding whether a differential or single-ended measurement is appropriate is
usually, by far, the most important consideration when addressing voltage
measurement quality. The decision requires trade-offs of accuracy and precision,
noise cancelation, measurement speed, available measurement hardware, and
fiscal constraints.
In broad terms, analog voltage is best measured differentially because these
measurements include noise reduction features, listed below, that are not included
in single-ended measurements.
•
Passive Noise Rejection
o
No voltage reference offset
o
Common-mode noise rejection, which filters capacitively coupled
noise
•
Active Noise Rejection
o
Input reversal
—
Review
Input and Excitation Reversal
(p. 326)
for details
—
Increases by twice the input reversal signal integration time
Reasons for using single-ended measurements, however, include:
•
Not enough differential terminals available. Differential measurements
use twice as many
H
/
L
terminals as do single-ended measurements.
•
Rapid sampling is required. Single-ended measurement time is about half
that of differential measurement time.
•
Sensor is not designed for differential measurements. Many Campbell
Scientific sensors are not designed for differential measurement, but the
draw backs of a single-ended measurement are usually mitigated by large
programmed excitation and/or sensor output voltages.
Sensors with a high signal-to-noise ratio, such as a relative-humidity sensor with a
full-scale output of 0 to 1000 mV, can normally be measured as single-ended
without a significant reduction in accuracy or precision.
Sensors with a low signal-to-noise ratio, such as thermocouples, should normally
be measured differentially. However, if the measurement to be made does not
Содержание CR800 Series
Страница 2: ......
Страница 4: ......
Страница 6: ......
Страница 32: ......
Страница 34: ......
Страница 54: ......
Страница 92: ......
Страница 190: ...Section 7 Installation 190 FIGURE 40 Running Average Frequency Response FIGURE 41 Running Average Signal Attenuation ...
Страница 310: ......
Страница 446: ...Section 8 Operation 446 8 11 2 Data Display FIGURE 100 CR1000KD Displaying Data ...
Страница 448: ...Section 8 Operation 448 FIGURE 102 CR1000KD Real Time Custom ...
Страница 449: ...Section 8 Operation 449 8 11 2 3 Final Storage Data FIGURE 103 CR1000KD Final Storage Data ...
Страница 450: ...Section 8 Operation 450 8 11 3 Run Stop Program FIGURE 104 CR1000KD Run Stop Program ...
Страница 452: ...Section 8 Operation 452 FIGURE 106 CR1000KD File Edit ...
Страница 456: ......
Страница 462: ......
Страница 523: ...Section 11 Glossary 523 FIGURE 116 Relationships of Accuracy Precision and Resolution ...
Страница 524: ......
Страница 526: ......
Страница 556: ......
Страница 558: ......
Страница 560: ......
Страница 597: ......