Unique summing situation – Confirm logic?

Feb 17, 2006 at 2:10pm

Unique summing situation – Confirm logic?

Hi,

I’m summing 4 signals. I know the typical logic would be to multiply each by .25 to keep the final level safe.

However, I’m using a series of crossovers to do “brickwall bandpassing” on the signals. e.g., one signal is 0hz to 600hz, the next 600hz to 1k, the next 1k to 6k, the last 6k to 22k.

So, when I sum these four signals, I don’t need to do any gain adjustment, right? I was working without any gain adjustment and I saw/heard no problems, but I’m trying to figure out if that’s by design or luck.

Thanks.

Adam

#24475
Feb 17, 2006 at 2:30pm

No you don’t have to adjust the gain like that. As long as the bands are not
overlapping there is no problem. It would also work with regular filters.
Cutoff frequencies are typically measured at -3 db. adding 2 of the same
signals together gives a 3db boost, so adding 2 signals, one lowcut at 500
and another highcut at 500 adds up to zero gain.

T_

#70831
Feb 17, 2006 at 2:33pm

filters without resonance that is….

T_

#70832
Feb 17, 2006 at 5:08pm

Quote: Adam Kendall wrote on Fri, 17 February 2006 07:10

> I’m summing 4 signals. I know the typical logic would be to multiply each by .25 to keep the final level safe.

or first mix them and then divide the sum /number of
channels.

#70833

You must be logged in to reply to this topic.