Using the Guard Terminals
- Updated2023-09-15
- 3 minute(s) read
Using the Guard Terminals
Guarding is a technique used to remove the effects of leakage currents and parasitic capacitances between HI and LO.
Guard terminals are driven by a unity gain buffer that follows the voltage of the Output HI terminal. In a typical test system where guarding is utilized, the Guard conductor is a shield surrounding the Output HI and Sense HI conductors. By making this connection, there is effectively a 0 V drop between Output HI and Guard, or Sense HI and Guard, so no leakage current flows from the Output HI or Sense HI conductor to any surrounding conductors. Some leakage current might still flow from the Guard output to Output LO. However, because the current is being supplied by a unity gain buffer instead of Output HI, the current does not affect the output or measurement of the SMU.
Cable insulation impedance is typically high, but it can have a significant effect when measuring small currents from high-impedance loads.
- When guarding is not used, the cable insulation impedance is in parallel with the load and causes the current measured at the device (IMeasured) to be the sum of the load current (ILoad) and the leakage current (IL), hence IMeasured = ILoad + IL.
- When guarding is used, the cable insulation impedance is still present, but because the voltage between Output HI and Guard, or Sense HI and Guard is 0 V, no leakage current flows between them. The capacitance between Output HI and Guard, or Sense HI and Guard also doesn't have to charge. Some leakage current (IG) flows from the Guard to Output LO, but this leakage does not affect the measurement because the Guard's power is sourced from its own unity-gain buffer. The result is the device accurately measures the current through the load, hence IMeasured = ILoad.
In the following figures, the external shield of the cable could be connected to chassis ground, and the figures assume that Output LO is connected to chassis ground.