24/7 emergencies 814 777 6921812 230 6079WhatsApp 811 786 0768
Electrónica MonterreyElectrónica MonterreyLaboratorios Creativeit Talk to an engineer
Menu

How do you reduce noise in a low-level signal measurement?

To reduce noise in a low-level signal measurement, you work on three fronts: keeping noise from getting in, lowering the measurement bandwidth, and averaging whatever gets through anyway.

Three fronts

A small signal is vulnerable because the noise coupled into it is comparable to the signal itself. The strategy is to couple in less noise, measure with less bandwidth, and extract the signal from the remaining noise by averaging.

Keeping noise out

  • Shielding and twisting: a shielded cable reduces electric coupling; a twisted pair reduces magnetic coupling, because the induced noise cancels between twists.
  • Single-point grounding: multiple ground connections form loops through which currents flow and add to the signal.
  • Distance from noise sources: keep cables away from power lines, motors, switching power supplies and digital equipment.
  • Low source impedance: a high-impedance source picks up more noise and generates more thermal noise.
  • Clean connections: oxidized contacts and dissimilar metals generate thermal voltages that show up as drift.

Measuring with less bandwidth

Random noise grows with the square root of the bandwidth. A low-pass filter or a longer integration time reduces noise, at the cost of a slower response. This is acceptable for slowly changing signals, but not for fast ones.

Integrating and averaging

High-resolution multimeters integrate the signal over a whole number of power-line cycles, which rejects that frequency and its harmonics. Increasing that time improves resolution. Averaging many readings reduces random noise with the square root of the number of samples, but it does not eliminate systematic errors such as offset or drift.

What to check first

  1. With the input shorted, check the instrument’s own noise.
  2. Connect the source through the final wiring and compare to see how much the cable adds.
  3. Switch off nearby equipment one piece at a time to find the source.
  4. Observe whether the noise changes with the power-line frequency.

How to interpret the result

If the noise stays the same when the source is disconnected, the problem lies in the instrument or the wiring; if it disappears, it comes from the source. Periodic noise points to external interference, random noise to thermal or electronic noise.

We do real tests, on video! Watch them on our YouTube channel →

Do you have a problem nobody has been able to measure?

Talk directly to the engineer in charge, with no middlemen.

Talk to an engineer