What can Go Wrong...: detecting noise floor issues using a field calibrator

Noise floor issues are very difficult to detect in the field. Fortunately most SLM owners already have a useful tool for measuring part of their SLM's noise floor.

I had a customer who had two monitoring locations near each other, and noticed the L90 of one location was consistently higher than the other.

They brought the two systems (NTi's XL2 SLM used with non-NTi microphone) back to their office. Background noise in their office was in the high 30 dB(A) so it was difficult to test. However, they could see an issue in the 4 kHz and above 1/3-octave bands.

Higher 1/3-octave bands are most sensitive to electrical noise, which typically resembles white noise. The higher the 1/3-octave filter, the broader its bandwidth, the higher the level when measuring white noise. Like stairs climbing up to the right.

(Plot of B&K Type 2250 1/3-octave self-generated noise from BE-1712 manual)

Most offices and outdoor measurement sites are much louder then the SLM's noise floor.

All wise SLM owners have a field calibrator. Calibrators are poor at blocking low frequencies (most everything is poor at attenuating low frequencies) but they do a decent job, 30-40 dB attenuation, at mid to high frequencies. Below is a plot of a CAL200 calibrator's attenuation as a function of frequency. (Of course the calibrator is off when using it for noise floor testing).

The consultant was fortunate he had two SLMs to compare. Take a quick measurement in a quiet space with the microphone mounted in the calibrator turned off.

That saved measurement is your SLM's noise floor baseline to help you detect noise floor problems in the future.