Channel EQ Cuts Cause Level Increase..?
May 2nd, 2021, 08:44
Neil Gilmartin writes:
Quote:
I have a channel strip with a Bitcrusher and Channel EQ on it. The Channel EQ is only cutting the very low and the very high frequencies. When I place the Channel EQ before the Bitcrusher, there’s no clipping. When I place the Channel EQ after the Bitcrusher, there is clipping. If I bypass the Channel EQ altogether, there's no clipping.
Why is the addition of a Channel EQ making any difference to the heat of the signal? This confuses me all the more because I'm only attenuating frequencies, not boosting any. I would have thought that attenuation would result in an overall gain reduction, not increase.
Thanks very much,
Neil
: End Quote
Hi Neil -
It's one of the more counter-intuitive aspects of DSP (digital signal processing), but sometimes cutting or filtering with an EQ can result in a slight increase in signal level. Just like all analog EQs, regular (minimum-phase) digital EQs (like Channel EQ) introduce phase shift. As a result, when EQ/filtering is applied it can change the cancellations & reinforcements inherent with any phase-shifted audio, and -- depending on the audio signal and the notes played -- the change to the waveform can result in the overall (instantaneous peak) level of the signal actually increasing somewhat, especially if the level of the signal was already close to the clipping point (it may not be audible as a level increase, but it's a real effect).
If you were to swap out the (minimum-phase) Channel EQ for the Linear Phase EQ with the same settings you shouldn't see the same effect -- I tried it and an saw increase of around 2 dB on the (peak) meter with the Channel EQ with highpass & lowpass filters engaged, but no increase with the Linear Phase EQ..