1176 Sidechain Filters & Noise Problem

Help Support GroupDIY:

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
All:

So I've been working on squeezing active sidechain filters into my 1176-alikes.  I've attached schematics of the design I've been using.  The three filters are HPF, LPF and a "Mid" boost placed in series.  The HPF and LPF are both second-order Sallen-Key filters while the "mid" thing provides a boost centered at 1kHz intended to serve as a sort of "loudness" button, compressing only the midrange, leaving the bass and treble alone.  So far it's been fun when two are stereo linked.  It could also be retuned to provide a rudimentary de-essing function.

Op-amp is a TL074 operating on 30V with a 15V split.  There are 10k pull-down resistors on the negative side of the blocking caps to prevent popping, with a 1M pull-down at the first position of the HPF for the same reason.  Higher resistor so that other HPF settings, which will back-feed that first resistor/capacitor combo as another HPF, will see it further out-of-band.  Changing settings on the HPF doesn't pop, just engaging it.  The other ones are 10k because any filtering effects with the 22µF caps are still out-of-band and after experimentation, that value seems to minimize pops better than higher values.

The 15V "rail" is a voltage divider buffered by a TL072 elsewhere on the motherboard.  The filters, btw, are on a separate card that interfaces with the motherboard via a molex connector, so I can remove it easily for tinkering and adjustment.

All rotary switches are shorting types, and the "mid" boost is engaged via a toggle switch.

Now here's the problem:

While the "mid" boost is whisper quiet, engaging the HPF and LPF introduces a fair amount of high frequency hiss (the stereotypical hiss, around 8kHz or so) into the audio signal path, and I can't figure out why.  Any ideas based on this schematic?  Could it have anything to do with the 15V "rail" being unregulated?  How about additional buffering going into both the LPF and HPF?  This is driving me absolutely nuts.
 

dirtyhanfri

Well-known member
Joined
Jan 16, 2011
Messages
678
Location
Madrid - Spain
I'm not an expert, but I would regulate that 15V going to the IC.

Maybe I'm wrong, but IIRC I've read somewhere a well regulated supply helps avoiding HF noises, maybe that's the problem
 

L´Andratté

Well-known member
Joined
Jan 17, 2013
Messages
684
Location
Hamburg Germany
Hi!
In your drawing in the upper left corner +15V and Gnd are swapped I think?
I´d call it +15V/0V/-15V, where 0V is Gnd. I have done that a lot with buffered voltage divider  with 24VDC wallwarts and had never such problems...but who knows?

Are R1, R2, R9 etc. connected to 0V or -15V? Should be 0V, same as R76(?),R86(?can´t read).

 

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
dirtyhanfri said:
I'm not an expert, but I would regulate that 15V going to the IC.

Maybe I'm wrong, but IIRC I've read somewhere a well regulated supply helps avoiding HF noises, maybe that's the problem

Yeah, I think I'll try it for laughs.  I've got the parts lying around.  But that doesn't explain why only two of the three filters is noisy.
 

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
L´Andratté said:
Hi!
In your drawing in the upper left corner +15V and Gnd are swapped I think?
I´d call it +15V/0V/-15V, where 0V is Gnd. I have done that a lot with buffered voltage divider  with 24VDC wallwarts and had never such problems...but who knows?

Are R1, R2, R9 etc. connected to 0V or -15V? Should be 0V, same as R76(?),R86(?can´t read).

Power supply rails going onto filter card are correct (or, at least, as I intended), I've got decoupling caps for both 15V and 30V.  I wonder if creating a "local" 15V power supply (voltage divider + op-amp buffer) on the filter card would help.

And yes, R1, R2 and R9, etc. are connected to ground to prevent DC pops on the switches.
 

Latest posts

Top