Reduce noise in low-current measurements

Noise can cause errors when you make low-current measurements, especially at levels under 1 nA. Here are a few likely causes of noise in such measurements along with suggestions for ways to avoid them:

• Electrostatic coupling. Coupling occurs when an electrically charged object is brought near an uncharged object.
  Remedies: Be sure to shield the device or circuit. High-impedance materials don't let the charge decay quickly. Shielding the device or circuit under test greatly reduces the electrostatic interference. Connect the shield to circuit low.
  Also, avoid movement and vibration near the test area. Keep all charged objects (including people) and conductors away from sensitive areas of the test circuit.
• Low source impedance. When the impedance of a current source (Z_S) is low relative to an ammeter's input impedance (ZF), noise will cause measurement errors.
  Remedies: Try adding some series resistance when measuring current from a capacitive source to reduce noise (see figure). Alternatively, you can add a forward-biased diode in series with the ammeter's input. The diode acts like a variable resistance—low when the charging current is high and then increasing in value as the current decreases with time. For nanoamp measurements, the combined source impedance and added resistor or diode should be at least 1 MÙ.

  Adding a resistor or diode to a current source increases its impedance and reduces noise.

  
  
• Vibration. Vibration produces a triboelectric effect that can cause noise currents to flow.
  Remedies:You can use low-noise cables, which have graphite lubrication on the shield braid, to reduce the friction between insulators in the cable to reduce this effect. You should also secure the cables to avoid any unnecessary vibration.
• Offset current drift. Offset currents can drift over a period of time, usually because of temperature changes in the device and the meter.
  Remedy: Stabilize the temperature of the meter, the device, and the general environment of the measurement.

To learn additional basic measurement techniques, see Dale Cigoy's article "When high-resistance measurements are too low," Test & Measurement World, April 2005. www.tmworld.com/archives.