The circuit described below was used for the U6 model, the U12 circuit was the same, utilising the equivalent voltage sourcing and measuring terminals.
Voltage was sourced from the DAC0 digital to analogue converter output terminal and applied to the series circuit comprised of the DUT, a shunt resistor and a ground terminal. The shunt resistor was a known value resistor which was used in order to measure the current in the circuit, as the LabJack had no in built capacity for such measurement. This was done by measuring the voltage across it using the analogue input AIN1, with respect to ground. From this voltage, ohms law was then used to calculate current. The voltage across the DUT was similarly measured using AIN0 and AIN1.
Voltage was dropped across each component as described by Kirchhoff series circuit law. It was established in the first term that drawing the maximum possible voltage across the device being tested required making the shunt resistance as low as possible. However too great a variation between the two resistances resulted in a decrease in voltage resolution. Conversely increasing the shunt resistance to equivalent or greater value to that of the device under test increased voltage resolution but at the cost of a decreased maximum voltage that could be dropped over the DUT.
Additionally a problem which principally occurred using the U12 model was that, due to the lower input impedance, lowering the total series resistance of the circuit too far (this problem was first encountered when using a 1Ω shunt and measuring an 11Ω DUT) results in a lowering of the possible current resolution of the setup below the size of the increments being passed through the circuit.
These effects could be compensated for by...
... middle of paper ...
... the manufacturer) resistor, utilising a 10KΩ shunt resistor taken at room temperature.
Figure 8- 1.8KΩ resistor I(V) Characteristic utilising a 10KΩ shunt resistor taken at room temperature.
The gradient of this plot gives; via ohms law, an average resistance of 1.75KΩ, well within the manufacturers tolerance, and within 3% of the 1802±0.005Ω as measured using a professional multimeter. It is worth noting however, that whilst the average value of resistance was a good match to measured values, initial data points during the sweep give variance from this average of up to 40% and so a lengthy sweep is required for this characteristic in even linear resistors. The voltage dropped over the device under test was largely determined by the relative resistances in the circuit, and so the functions outlined in Sections 3.1 and 3.2 are an integral part of this calculation.