I’ve always wondered this and never really looked for an answer so if anyone knows why I’d be grateful for an explanation. I’ve noticed there are certain times where if I apply a low cut the output level meter goes up a decibel or two. Solely for example let’s say to a vocal low cut from 80hz on downward at -15db cut. Before the cut the output level for the channel is -6.0db, but after the cut it’s gone up to -4.5db. Typical thinking would be that there’s less audio level present since the low end was cut out, but yet the level has gone up. What causes this? Is it a phase issue or something? It also happens with every EQ, doesn’t matter if it’s Waves, Ultrafunk, the built in Cubase EQ’s, etc. It doesn’t happen often, but I’m rather curious to know why. Thanks in advance to anyone that can answer this, or even better provide a solution 
Rev.
Filter resonance maybe? Never noticed anything like this tbh, but then again I don’t closely watch the dB values in my mixer either, apart from the master of course.
Is this in a context of a mix or a solo instrument?
I noticed this a while back, and then soon thereafter came across the article below in SOS. My guess … it is because EQ’ng changes how the component frequencies/waveforms that make up the total sound add/subtract with each other: removing some frequencies can result in less destructive interference >> louder. As you pointed out, this changing of the additive/destructive interference pattern can also occur by the phase shift associated with some/most? EQs.
I don’t know why the math works out so that the peaks are more often raised than lowered. I’d guess (but haven’t tested) that in either case, the total energy RMS would go down when applying a filter cut, even if the peak happened to go up.
Love to hear more opinions on this!
Primarily solo instruments or vocal tracks since those are the only things I usually low shelve.
@Alexis - That is very interesting info. Thanks for the link, I will read it as soon as I get a chance - currently at work and it’s a busy busy day.
*EDIT - heck that was a short read, so I got to read it and it sure answers my question. Makes a LOT of sense that removing some frequencies might omit present phase cancellation, and without the cancellation the output would increase. Thanks again man!
Rev.
You’re welcome!
I do get the feeling there is a bit more to the story though … as the explanation doesn’t seem to answer why (when applying a high pass filter) there is apparently an increase in peak level more often than a decrease . Does this somehow suggest that bass frequencies tend to result in destructive interference patterns more often than additive/constructive ones? If so, why would that be?
Where are the acousticians of yesteryear?
From my personal experience I’ve seen it more often decrease the level, as one would usually expect, and the increases that I see are rarer which is why I found it so peculiar.
Rev.
Well, thank you for that! I only noticed it (higher peaks) once or twice when I started high pass filtering vox, but really didn’t check it out as much as I could have, as far as how often it happened. Comforting to know the usual is as “expected”.
I’ve only ever noticed an increase when using a HPF with a pronounced bump at the cutoff frequency.
But then as others I’m not always checking the levels visually in minutia!