UAD 1176 Sidechain Filter feature confusion and limited info online

[skipper01]

Hi, can anybody advise regarding the UAD 1176 sidechain feature that was introduced fairly recently as an update.
The one that ‘lifts’ on the Release knob.
I discovered it on the blue strip version when compressing a vocal.
I’ve checked the manual and there’s limited info there and online.
It appears to be an on or off feature but no info is given on the centre frequency to which the slope is working on the side chain?
I was also confused as with it engaged (release knob out) it appears in my example to be compressing more than with it not engaged as standard. I expected the opposite in theory.
To my understanding without it engaged (release knob in as standard) it would (material dependent) more likely be compressing/triggered more as it’s working full frequency spectrum including bass?
With it engaged (Release knob out) it is not going to be triggered by low frequency content as a side chain filter?
Does that make sense?
Does anybody know the frequency it operates on or if I’m confused.
It just seems to be sounding better engaged on my vocal but maybe I’m in a mess with my understanding and control of what’s going on???

[Wonks]

UA don’t mention the frequency, but say it’s a gentle 3dB/octave high-pass filter. So it probably starts between 200Hz-300Hz if it’s to have any noticeable roll-off at bass frequencies e.g. if its 3dB down at 100Hz, if will be 6dB down at 50 Hz and 9dB down at 25Hz. So, relatively gentle stuff, but enough for the bass to have less effect on the compressor operation.

[skipper01]

Yes those sort of figures make sense so thanks for those thoughts.
It’s just surprising the difference it makes to the gain reduction when applied.
I appreciate that it’s obviously material dependent but it appears to be very influencing on the gain reduction far far more than I’d expect.
It appears necessary to completely alter the input level in use with the side chain off or on in my vocal processing application.
Ho hum

[Urthlupe]

Hey Skipper - I’m wondering whether you’re simply hearing the effect of the compressor more directly on signal content above the sidechain filter corner frequency…..? It’s often not the easiest thing in the world to hear how and understand why compression is behaving.

Maybe instate your own HPF in the sidechain of an 1176 and play around with the corner frequency and slope to see if you can replicate the outcome you’re hearing (although of course those two factors alone may not be the whole story - HPF’s are flexible beasties).

Loopy

[Wonks]

3dB/octave filters are rare beasts. You use a low pass one to convert white noise to pink noise, but there isn't much call for them outside of that situation. So finding a high-pass 3dB/octave filter to simulate the UA sidechain bass filter could be quite hard. You could get a general idea of the effect of a filter on the compression with a standard 6dB/octave filter, but it won't be exactly the same.

Are you high-pass filtering the signal to cut out any subsonic artefacts before compressing?

[skipper01]

Hi, thanks for these further thoughts/suggestions.
Its a pretty dynamic vocal im trying to dynamically control with a UAD 1176 first (for transient control) then LA2A following.
Firstly, yes i actually employed a 40hz high pass on the vocal at source (on analogue channel strip). I actually even knocked off (low shelf) around 3db going in from 160hz too just cause it sounded 'better' to me going in.
I guess given its transient heavy material perhaps this is a reason for the results when employing the sidechain on the plugin?

Also, sorry if im being dumb but what do you mean by me employing my own LPF on the side chain as i dont see or think there is a side-chain option (other than the 3db per octave one built in is there?)
Assuming its being used as an insert on a track...

Thanks again

[Wonks]

No, I think we assumed too much there. There's no external sidechain input, so forget that with the 1176, though you could try it with a different compressor that has one, though obviously it won't behave quite like the 1176 will.

The 1176 is more like a limiter than a standard compressor, with pretty fast attack times - 20 to 800 microseconds (0.02 to 0.8 ms), so very easy to trigger with very fast transients.

As is normal on a compressor, the release times are much slower; between 80ms and 1100ms.