Semiconductor parametric testing characterizes electrical properties under defined conditions. Start with the device structure and the decision the result must support. A useful wafer-level test plan connects identity, applied stimulus, measurement settings, calibration and status flags to an explicit acceptance rule.
Choose a measurement for the parameter
A current-voltage (I-V) test observes current and voltage under a specified stimulus sequence. Capacitance-voltage (C-V) and RF characterization address different behavior with additional method requirements. Keysight's parametric-test handbook describes these measurement families and the role of fast or pulsed methods.
Write down the device type, terminals, wafer site, temperature and parameter definition. If extracting a value from a curve, preserve the extraction algorithm and fitting interval. Confirm the approved device limits before defining bias, sweep direction, dwell time or compliance.
Choose a representative control structure and a way to detect setup problems. Keep wafer identity linked to the site map and raw data. A valid measurement assigned to the wrong site produces an invalid manufacturing conclusion.
Make the stimulus and measurement path explicit
On a narrow screen, scroll the table horizontally. Keyboard users can focus it and use the arrow keys.
| Item | Record | Interpretation |
|---|---|---|
| Stimulus | Bias sequence, polarity, timing and actual output | Defines the condition experienced by the device |
| Compliance | Limit setting and whether it was reached | Identifies when the requested source condition may be constrained |
| Measurement | Range, integration time, delay and status | Supports comparison of recorded values |
| Connections | Probe, force/sense, switching and guard/shield configuration | Preserves the relevant electrical path |
The Keysight E5270B quick reference separately identifies source settings, compliance, ranges and wait times. For the instrument actually used, retain the equivalent configuration and status fields. Treat a compliance event as information to review, rather than accepting the requested setpoint as the achieved condition.
Read calibration and uncertainty specifications for the selected range and conditions. Display resolution is the smallest indicated increment; the result's usable accuracy also depends on the instrument and test setup. Retain self-test and calibration evidence as separate parts of the record.
Investigate low-current setup effects
Keithley's Low Level Measurements Handbook explains leakage, settling, environmental influences and the distinct roles of guarding and shielding. Guarding reduces certain leakage paths by controlling potential; shielding addresses interference. Apply the instrument's connection scheme, with appropriate protective grounding and rated fixtures.
Check clean insulation, stable cables and the specified environmental conditions. Use an appropriate fixture/control measurement to investigate offsets while preserving the device test configuration. A background observation supports diagnosis; correcting DUT data requires a validated correction method and uncertainty treatment.
Switching can introduce transients. The Keithley switching guide explains why measurements need sufficient settling after a relay change. Determine the delay experimentally for the relevant setup within its approved test limits.
Use time dependence to investigate a result
Fictional leakage observation: under one approved fixed bias, measured current is 32 pA after 0.1 s, 14 pA after 1 s and 12 pA after 5 s. The combined device and setup response is time dependent.
Repeat the observation and compare the control, switching history and environmental record. The three readings alone neither establish the final settled value nor identify the device as the source of the change. Select a delay and stability criterion using the applicable test procedure and required uncertainty.
Preserve out-of-range, compliance and invalid-measurement flags with the values. Test the analysis software with such records so they cannot quietly become ordinary passing numbers.
Keep qualification tied to the complete method
For C-V or RF work, define frequency, amplitude, calibration plane, fixtures and any de-embedding separately. A validated DC leakage setup establishes its own scope; extending the test family requires corresponding validation.
Review the affected evidence after changes to probes, switching, software, ranges or device structures. Keep the raw data and procedure revision with the accepted result, and use the timing measurement guide when the decision specifically concerns edges, delays or jitter.
Use the working record
Download the editable parametric test plan. Save a copy, complete the actual requirements and evidence fields, and review open issues with the responsible owner. The plain-text record includes instructions, units where needed, the fictional example, and source references.