Indicator verification instructions
The following two practical methods can be used to detect the accuracy of 0.5% FS, and the detection accuracy is higher than 0.00625% FS.
1、Input and output signal equivalent product testing:
The relays and the semaphore switch in the figure are common components (the foot switch can be used), and there is no special requirement for contact resistance and withstand voltage. The accuracy, drift, linearity and other indicators of the meter can be 0.5% FS, but the resolution should be 1μA (4-bit 1/2 portable or benchtop).
This is a relative error measurement method; the instrument with the input and output signals equivalent (such as input and output signals are 4-20 mA), and the size of the two signals is detected by a single measurement meter. In a short period of time (such as 10 seconds), the meter will not change, and the relative errors of the two signals measured during that time will not change. The measurement accuracy is independent of the size of the input signal and is independent of the absolute value of the measurement table, only related to the resolution of the measurement table. The resolution of the measurement table reaches 1μA, and the absolute error of the measured meter is very close to 1μA.
Relative error: The measurement error of 1 μA relative to 4-20 mA signal is 0.00625% FS, which is 80 times higher than the accuracy of the measurement table itself (0.5%).
Absolute error: Depending on the resolution of the meter and the linearity error, if the linearity error of the meter is 0.5% FS, set the input and output difference to ∆I, and the linear error of the difference is ∆I/20mA*0.5%. The sum of this value and the resolution of the meter is the absolute error. For example, the input-output difference is ±10μA, and the absolute error of the measurement itself is 1μA+10uA/20mA*0.5%=1.0000025μA, which is regarded as 1μA, that is, the absolute error is measured as the resolution of the measurement meter.
The ratio of automatic control "equivalent" instruments is higher than 80%, so the relative value measurement method is widely used. Metering problems for most "equivalent" meters can be solved on site with low-cost meters.
2、Input and output signals are not equivalent to product detection:
The universal safety barrier (isolator) uses four input terminals to switch signal inputs such as power distribution, current, thermocouple, thermal resistance, millivolt, and slip line resistance. The system commonly used input signals are included. During on-site construction, the signal configuration can be performed in the state where the power supply, input and output terminals are suspended (not energized), and the instrument and signal generator are not required. The accuracy after power-on is better than 0.03% FS.
Verification is necessary, but if all of the general barrier (isolator) input signals are detected, the workload is huge and even an unfinished task. Therefore, the actual inspection is mainly based on the process requirements, and the verification is spot-checking.
The general-purpose safety barrier (isolator) contains 4~20mA signal and current input signals, which are equivalent to the output signal and can be detected by relative error detection method. The internal reference and measurement circuit used by the distributor signal and current input signal are the same component as the other non-equivalent signals. If the output value corresponding to the distribution signal or current input signal is qualified, from the verification accuracy, After the other input signals are configured, more tests are performed, but they can be left unchecked.
In addition to the necessary metrological verification (concentrated on pre-factory calibration), the relative error detection method avoids the use of high-precision detection instruments and signal generators to detect the safety barrier (isolator), and does not require the inspection table. Periodic measurement calibration (relative value measurement does not have transmission error). At the same time, the calculation of the error is simplified, and the measurement can be performed at any point within the range. The difference between the input and output differences is 0.01% FS per 1.6 μA.