High-Value Printed Resistors: Leakage and Guarding in Measurement

Decide which current belongs to the device

Draw the resistor, pads, substrate surface, supports, cable shields and instrument input as a circuit. Mark the two terminals across which the reported resistance applies. A surface path between those terminals may legitimately be part of an assembly acceptance test even when it obscures the intrinsic film resistance. A guard that diverts this current changes the measured quantity; document that change rather than treating every higher reading as automatically more correct.

For development, it is useful to retain two quantities: the guarded material-oriented result and the terminal-to-terminal assembly result under its specified environment. Their difference helps locate insulation contributions. Comparing a guarded incoming coupon with an unguarded finished assembly without preserving this distinction can incorrectly attribute a packaging or contamination problem to a change in resistor composition.

Convert the error allowance into a leakage requirement

In the simplest case, leakage behaves as a resistance in parallel with the intended resistor. Conductances then add. If the intended resistance is one gigohm and the unwanted path is 100 gigohms, the indicated value is approximately 990.1 megohms, about 0.99 percent low. The fixture resistance sounding much larger than the device resistance does not make it negligible when the acceptance band is narrow.

At ten volts, a one-gigohm resistor carries ten nanoamperes. A 0.1 percent current allowance corresponds to ten picoamperes. This is a hypothetical budget calculation, not an instrument or substrate rating. Use it to determine the required blank-fixture stability and current resolution. A nominally small zero correction is inadequate if its drift consumes most of that allowance during the measurement sequence.

Give the guard and shield different jobs

A guard is driven near the potential of the sensitive node so that little voltage remains across selected insulation leakage paths. An electrostatic shield intercepts interference and follows the instrument's prescribed reference or protective connection. These functions are not interchangeable. Carry the driven guard toward the fixture using the instrument-compatible cable and terminal configuration.

Use the manufacturer's connection diagram for the exact measurement mode. A terminal labeled guard may behave differently between source-current, source-voltage and electrometer configurations. Confirm the guard's voltage range and loading restrictions. Keep any exposed or accessible structure within the fixture's electrical design requirements; a driven guard is not necessarily at ground potential. Record the connection arrangement with the data so another laboratory can reproduce the boundary.

Separate charging current from steady conduction

Immediately after applying voltage, the observed current may include charging and dielectric relaxation as well as resistor current. Choosing a short fixed delay can make measurements depend on cable length, substrate support and the preceding voltage history. Capture a current-versus-time sequence during development. Look for a stable region relative to the required uncertainty, rather than selecting a delay solely because the instrument display has stopped changing its leading digits.

A simple resistance-capacitance model gives a time constant equal to resistance multiplied by capacitance. One gigohm with 100 picofarads gives 0.1 second, but a real insulator can relax through several time scales. The single time constant therefore establishes only a starting estimate. Define the energizing, reading, discharge and rest intervals together, and use the same sequence for blanks and populated fixtures.

Use controlled substitutions to locate the extra path

An open fixture establishes what the cables and supports draw without the specimen, but it does not reproduce the electrical field created by the installed substrate. Use a clean nonconductive blank of relevant geometry when appropriate, and repeat measurements with individual supports or connectors substituted. Keep the voltage, guard arrangement and observation time identical so that the differences remain interpretable.

Check the voltage without confusing it with time

A resistor may show voltage-dependent behavior, while insulation leakage can also change with applied voltage. Measure a defined ascending and descending voltage sequence with the same settling criterion at each step. Report current and voltage as well as calculated resistance. A plot of current against voltage exposes behavior that a table of rounded resistance values can hide.

Keep the maximum test voltage within the specimen and fixture's established limits. Increase voltage only when the expected signal improvement is worth the additional electric field and power. If apparent resistance decreases at higher voltage, first compare the blank and guarded conditions. Reversing polarity can help identify offsets or asymmetry, but it does not automatically cancel surface conduction or erase the previous electrical history.

Make humidity and surface handling reproducible

Humidity should be recorded where the specimen is actually conditioned and measured. Transferring a dry specimen into humid room air starts another time-dependent experiment. State whether the reading is taken inside the conditioning enclosure, immediately after transfer or after an agreed equilibration period. These conditions can produce different valid answers to different application questions.

Control gloves, cleaning chemistry, drying, storage containers and the time since handling. Avoid adopting a cleaning treatment that leaves residues or attacks the printed material, termination or coating. When cleaning changes resistance, retain both electrical measurements and surface observations before deciding that the original part was defective. A production specification needs an accepted surface condition, not simply an instruction to clean until the desired number appears.

Report enough information to reproduce the result

The measurement record should identify applied voltage, measured current, sample temperature, humidity, timing, guarding and fixture blank results. Include the direction of any correction and its uncertainty. When a correction is comparable to the specimen signal or varies unpredictably, improve the setup before using the corrected value for a narrow acceptance decision.

For a resistor drawing review, connect the electrical requirement to the physical leakage routes. Identify nearby conductors, substrate edges, protective coating and the intended installation environment. This allows the resistance requirement to be evaluated alongside insulation and spacing. The useful outcome is a repeatable measurement that represents the installed function, with any intrinsic-film characterization clearly distinguished by its stated terminal boundary.