Design Article
Tips and tricks for optimizing low current and high resistance measurements, part 2
Mary Anne Tupta, Keithley Instruments Inc.
8/10/2012 4:10 PM EDT
Editor’s note: Click here to review part one of this feature.
One of the first steps in setting up a system for making very low current measurements is to determine the offset and leakage current of the entire measurement system. This identifies the noise floor limit of the entire system and sets a starting point for troubleshooting potential problems and making improvements to the system, if necessary. Two types of offsets must be taken into account: internal and external offsets. Internal offsets include the input bias current of the ammeter; external offsets include the offsets due to the connecting cables, adaptors, test fixtures, and probes.
First, quantify the input bias current of the ammeter, which is determined by measuring the open circuit current on the lowest range. Place a metal cap (this is not a short but is used as a noise shield) on the input of the ammeter. Select the lowest current range and measure the current once the readings are settled (see figure 4). Compare this current to the instrument’s specifications.

Once you’ve measured the input bias current, verify the offset of the rest of the system. Repeat the offset current measurement by adding pieces one at a time to the test circuit and repeating the current measurement. The final measurement will be performed with all the cables, connectors, and test fixture or probe station in the test circuit, but with no device connection. This process will help determine any trouble spots, such as a shorted cable in the measurement circuit.
Sources of measurement error
Low current and high resistance measurements are subject to errors from a variety of sources, including leakage current and settling time.
The settling time of the measurement circuit (i.e., the time the measurement takes to stabilize after the current or voltage is applied or changed) is particularly important. The measurement range used will affect settling time; usually, the lower the current range, the longer the settling time. The shunt capacitance, which is due to the cables and test fixtures, and the source resistance also alter the settling time—the higher the shunt capacitance and source resistance, the longer the settling time will be.
Electrostatic interference occurs when an electrically charged object approaches the circuit under test. At low impedance levels, the effects of the interference aren’t noticeable because the charge dissipates rapidly. High resistance materials don’t allow the charge to decay quickly, however, which may result in unstable, noisy measurements.
Typically, electrostatic interference is an issue when making current measurements of less than 1 nA resistance measurements greater than 1 GΩ (see figure 5). To minimize problems, the circuit being measured should be enclosed in an electrostatic shield. The shield can be a simple metal box or a meshed screen that encloses the circuit. Connect the shield to the test circuit common, which is the LO terminal of the ammeter. Also, keep all charged objects and conductors away from sensitive areas of the test circuit.

Leakage current presents another error source. Leakage current is an error current that flows through the insulation of the test circuit when voltage is applied. It typically becomes an issue when the impedance of the DUT is comparable to that of the insulators in the test circuit. The following tips help reduce leakage currents in the test circuit:
One of the first steps in setting up a system for making very low current measurements is to determine the offset and leakage current of the entire measurement system. This identifies the noise floor limit of the entire system and sets a starting point for troubleshooting potential problems and making improvements to the system, if necessary. Two types of offsets must be taken into account: internal and external offsets. Internal offsets include the input bias current of the ammeter; external offsets include the offsets due to the connecting cables, adaptors, test fixtures, and probes.
First, quantify the input bias current of the ammeter, which is determined by measuring the open circuit current on the lowest range. Place a metal cap (this is not a short but is used as a noise shield) on the input of the ammeter. Select the lowest current range and measure the current once the readings are settled (see figure 4). Compare this current to the instrument’s specifications.

Click image to enlarge
Figure 4. Input bias current of an SMU as a function of time.
Once you’ve measured the input bias current, verify the offset of the rest of the system. Repeat the offset current measurement by adding pieces one at a time to the test circuit and repeating the current measurement. The final measurement will be performed with all the cables, connectors, and test fixture or probe station in the test circuit, but with no device connection. This process will help determine any trouble spots, such as a shorted cable in the measurement circuit.
Sources of measurement error
Low current and high resistance measurements are subject to errors from a variety of sources, including leakage current and settling time.
The settling time of the measurement circuit (i.e., the time the measurement takes to stabilize after the current or voltage is applied or changed) is particularly important. The measurement range used will affect settling time; usually, the lower the current range, the longer the settling time. The shunt capacitance, which is due to the cables and test fixtures, and the source resistance also alter the settling time—the higher the shunt capacitance and source resistance, the longer the settling time will be.
Electrostatic interference occurs when an electrically charged object approaches the circuit under test. At low impedance levels, the effects of the interference aren’t noticeable because the charge dissipates rapidly. High resistance materials don’t allow the charge to decay quickly, however, which may result in unstable, noisy measurements.
Typically, electrostatic interference is an issue when making current measurements of less than 1 nA resistance measurements greater than 1 GΩ (see figure 5). To minimize problems, the circuit being measured should be enclosed in an electrostatic shield. The shield can be a simple metal box or a meshed screen that encloses the circuit. Connect the shield to the test circuit common, which is the LO terminal of the ammeter. Also, keep all charged objects and conductors away from sensitive areas of the test circuit.

Click image to enlarge
Figure 5. Unshielded vs. shielded current measurements of a 100 GΩ resistor show the dramatic increase of noise for unshielded systems.
Leakage current presents another error source. Leakage current is an error current that flows through the insulation of the test circuit when voltage is applied. It typically becomes an issue when the impedance of the DUT is comparable to that of the insulators in the test circuit. The following tips help reduce leakage currents in the test circuit:
- Use good quality insulators in the test circuit, for example, Teflon and polyethylene.
- Minimize humidity in the test environment.
- Always use guarding. A guard is a low impedance point in the circuitry that’s at nearly the same potential as the high impedance lead being guarded. Guarding may be necessary to prevent leakage current due to the cabling or test fixture. Using triaxial cables will enable guarding to minimize leakage current and achieve a faster settling time. Guarding reduces the effects of the cable capacitance, which decreases the response time of the circuit.
Navigate to related information


I_B_GREEN
8/21/2012 1:43 PM EDT
Lots of good nitty gritty details and advise...On Mark
Sign in to Reply