By using a guarded scanner and two sources to form a guarded resistance bridge, measurements from 100KW to 10GW can be made with excellent accuracy. This system provides a simple yet effective way to activate both the high and low guard circuits. Adjusting the source outputs to set the high impedance side of the bridge circuit to zero volts reduces errors caused by meter circuit loading. The system described here is similar to one to the one used by NIST for measuring resistors in the GW range.

This circuit is a wheatstone bridge where two legs of the bridge are voltage sources. A DVM measures the voltage across the bridge and a low thermal scanner is used to switch the resistors in the test.

The standard low thermal scanner has leakages of about 1012W. In a normal arrangement the leakage currents would cause errors of about 1 ppm at the 1 MW and 100 ppm at 100 MW. By using a fully guarded scanner, leakages can be significantly reduced.

Because the sources have low impedances, the high guards can be connected directly to the source outputs. The low guard can be connected directly to ground because the sources are always adjusted so that the DMV reads zero.

Keeping both sides of the bridge at zero volts reduces leakage errors. This is done by using two voltage sources for two arms of the bridge as shown in the diagram below. Voltage source #1 is adjusted so that the DVM always reads zero, which sets the center point of the two resistors being compared to zero volts. The tare standard is always in the circuit, and the low thermal scanner is used to switch the standard and test resistors into the circuit one at a time.

The High-Ohm Resistance method is now included in the Data Proof OhmRef software. This provides a convenient means to set up an make high resistance measurements. OhmRef will allow up to 8 resistors to be compared at a time.

A non-guarded scanner can be used to compare resistors to 10MW with an uncertainty of about 1ppm. With a guarded scanner good results can be obtained comparing 1GW resistors with an uncertainty of about 10 ppm.

Different values of resistors can be compared over a wide range with the uncertainty is primarily dependant upon the scaling accuracy of the voltage source used.

Image12.gif (9101 bytes)

Page 11

Home Page Contents (Page1) Previous Section Next Section