Detecting open inputs

A useful option available to single-ended and differential measurements is the detection of open inputs due to a broken or disconnected sensor wire. This prevents otherwise undetectable measurement errors. Range codes appended with C enable open-input detection. For detailed information, see the CRBasic help (VoltSE() and VoltDiff() instructions, Range parameter)

The C option may not detect an open circuit in the following situations: 

• When the input is not a truly open circuit, such as might occur on a wet cut cable end, the open circuit may not be detected because the input capacitor discharges to a normal voltage through external leakage to ground within the settling time of the measurement. This problem is worse when a long settling time is selected, as more time is given for the input capacitors to discharge to a "normal" level.

• If the open circuit is at the end of a very long cable, the test pulse may not charge the cable (with its high capacitance) up to a voltage that generates NAN Not a number. A data word indicating a measurement or processing error. Voltage overrange, SDI-12 sensor error, and undefined mathematical results can produce NAN. or a distinct error voltage. The cable may even act as an aerial and inject noise which also might not read as an error voltage.

• The sensor may "object" to the test pulse being connected to its output, even for 100 µs. There is little or no risk of damage, but the sensor output may be caused to temporarily oscillate. Programming a longer settling time in the CRBasic measurement instruction to allow oscillations to decay before the ADC Analog to digital conversion. The process that translates analog voltage levels to digital values. may mitigate the problem.