XS902 De-esser - First Prototype!

StephenGiles · Mar 2, 2008

This is one of the most fascinating threads I have ever followed, although I need to sit down for a couple of hours in peace to understand it. Which is the latest schematic and is it still likely to be improved.

Just a thought, could the high pass alone be used to clean up a hissy cassette recording?

drpat · Mar 2, 2008

Stephen,

The high pass is the part of the de-esser that reduces the gain of only the hi frequencies while it's de-essing. With it out, the gain reduction is broad band with the side chain still being triggered by the sibilence. Since hiss is a steady state problem, a notch filter, or even a multi-band expander would be more effective.

I'm not sure that they want to improve the design as much as they're working towards a solution to build the thing with readily available parts. In my opinion, there's not much you could do to improve this device. It does it's own thing extremely well.

Moby · Mar 2, 2008

Great thread guys :grin: Curtis, please keep on. BTW, when you reach some acceptable results, do you have an original 902 unit just to make side by side comparable audio clips. It would be great to hear how clone works. :wink:

JohnRoberts · Mar 2, 2008

[quote author="Curtis"]
Personally I've always found it extremely useful to have a de-esser on-hand for that extra "polishing", not just for it's obvious uses on overly "spitty" vocals, but also for things like de-emphasising a reverb send on a vocal, or cleaning up the bleed from hihats in the snare mike without resorting to heavy gating. The fact that you can also perform fancier moves on the 902 by just having it work on the HF signal (the mode switch), or extreme HF gain reductions makes it an interesting shaping tool to be playing with too.[/quote]

+!.. I recall back in the late '80s when stuck with a cheap reverb multi-efx that used too much HF pre-emphasis in some settings, by aggressively de-essing the send to it would allow me to hit it a lot harder without overload. Definitely a band aid, but a very useful band aid.

JR

JohnRoberts · Mar 2, 2008

[quote author="StephenGiles"]This is one of the most fascinating threads I have ever followed, although I need to sit down for a couple of hours in peace to understand it. Which is the latest schematic and is it still likely to be improved.

Just a thought, could the high pass alone be used to clean up a hissy cassette recording?[/quote]

You could rearrange the pieces to deliver a HF cut dynamically when you determine that the spectral signature is predominantly noise floor or just when HF is below some threshold.

You could make the HF attenuation follow a slope similar to a downward expander, so it doesn't sound choppy.

IMO this wouldn't be as effective as a dynamic sliding LPF as promoted by Nat Semi in that old DNR (?) chip, or the combination of sliding LPF with downward expansion which could be quite effective. The SENR (single ended NR) I did used two VCAs one for the variable LP and another for the downward expansion, but that's a little more complicated circuit.

Getting back to modifying this for SENR, the opamp comparator that defeats the de-essing needs to be retargeted to just stop the de-essing gain manipulation, while another opamp LOG computation is performed, with perhaps another threshold (and slope?) adjustment, determines when to start scraping off highs and an appropriate ratio. Note: In my experience even 1:2 is a pretty severe slope so ratio may want to be modest for best sounding result.

JR

Curtis · Mar 3, 2008

This is one of the most fascinating threads I have ever followed, although I need to sit down for a couple of hours in peace to understand it. Which is the latest schematic and is it still likely to be improved.

I haven't posted the full revised schematic yet, but you could probably get what I have so far by combining the original one on page 1 of this thread and the second revision I posted here

Great thread guys Curtis, please keep on. BTW, when you reach some acceptable results, do you have an original 902 unit just to make side by side comparable audio clips. It would be great to hear how clone works.

I haven't had access to one for years now, so all I'm going on is memory and comparisons against my favourite de-esser plugins. This is where Keith's offer for sweeps and measurements of his originals could come in handy to verify some of my results.

PCB's drilled and waiting for lacquer to dry. Stay tuned...