EHX Steel Leather? What's it doing?

[Rodgre]

I was just speaking with a friend who uses a Steel Leather in conjunction with a few other pedals for a really specific sound and it got me wondering what the Steel Leather actually does. I couldn't find a schematic so I thought I would ask the family here.

It's termed a "bass expander" and that just sounds like buzzwords to me, unless it is literally an expander circuit (based on something similar to a polite noise gate). I feel the same way about the Analogizer pedal, which doesn't really explain what it technically does, but that's neither here nor there.

Does anyone have an idea what the Steel Leather is doing?

Roger

[teemuk]

According to manual it indeed is an expander, like in the literal sense.

But the catch is that the dynamic range expansion is applied only to rather narrow band of frequencis; according to manual its tuned to effect frequencies that make up the "attack" of picking the strings, so it's like tuned to effect only the very upper harmonics, not the whole spectrum of a bass.

The "Response" knob is a sensitivity control to dial in at which amplitude the expander effect is toggled. At higher settings quiet notes are not expanded and at lower settings the expander is pretty much toggled all the time.

"Effect" knob blends between dry (non-processed) and wet (expanded) signals.

[Rodgre]

Thanks Teemuk! That makes a lot of sense. It sounds like a pretty forward-thinking design on EHX's part.

So one could do something similar, either with blending in an expander circuit on a split of the input signal that is either heavily high-passed or else to do something like the Aphex exciter trick and distort and create higher harmonics that weren't there to begin with, high-pass that and then apply the expander to it, am I right?

Roger

[Mark Hammer]

Boy oh boy, does this ever look like a job for a dual ground-leg gain stage, with one leg using an LDR to set gain for high-end.

Let's use the Hollis Flatline as a jumping-off point.  Here's Madbean's lovely drawing of his take on it.

The "Sus" pot sets the gain of that op-amp stage from a minimum gain of 3x, when the pot is at max resistance, to a max gain of 23x when the Sus control is set to zero ohms.  The rectifier circuit, built around IC1_B takes IC1_A's output, at whatever the gain is, to light up the LED in the vactrol, reducing the resistance of the LDR, thereby reducing IC1_A's gain.  So far so good.

But now let's imagine that we move the LDR half of that vactrol to be more or less in parallel with the Sus pot, BUT we stick a 47nf cap between pin 2 of IC1_B and that LDR.  When the LED is dark, this new /047/LDR path will have little audible impact on the sound, since the LDR's resting resistance is likely higher than the 220k feedback resistor.  Gain of that stage will continue to be set by the Sus pot.

However, as the gain is increased and the rectifier starts to drive the LED in the vactrol, the LDR's resistance will decrease, and the gain will increase for whatever can pass via that path.  In other words, that frequency content will have more gain applied to it than to the rest of the spectrum.  If we were to lush the LDR's resistance down to roughly 2k, that would provide a gain of roughly 111x for content above 1690hz.  Meanwhile, the gain for the rest of the spectrum might be only 10-15x.

Naturally, one would wish to have only sensible boosts occur, so as to NOT be restricted to only a tiny fraction of the Sus pot's range.  So, for instance, the LDR itself might be in parallel with a fixed resistor (e.g., 470k) with a series resistor between the cap and other components, to provide both a minimum and maximum resistance range for that leg.  What you'd want to aim for is something that gets you a reasonable increase in "bite" (stuff over 1.5khz), without taxing the op-amp and getting distortion.

If our cap remains at .047, and we stick a 1k5 resistor in series with it, pushing the LDR to absolute zero resistance (highly unlikely to happen, especially with a max op-amp gain of 23x), we would end up with a gain of 147x for content above roughly 2.3khz.  But since the LDR is unlikely to descend much below 500R, our max is likely to be around the 111x for content above 1690hz.  So let's make the series resistor 2k2, such that the LDR+resistor rarely go below a combined resistance of 2k7.  Pushing the LDR down to a minimum resistance of 500R, this would get us a maximum gain of 82x for content above roughly 1250hz.  Much better.

And so on.  The gist is that the overall gain of that op-amp stage can be used to increment the gain of a range of upper frequencies, via use of a second ground leg incorporating an LDR.  The rest of the Flatline/Afterlife circuit remains the same.

[ElectricDruid]

This dual ground leg gain stage stuff is all the rage around here at the moment, isn't it?!? 

I like your idea though. Hollis' flatline is a typically stripped-down design, doing quite a lot with not a lot. John Hollis is a master of that. And it *is* easy to see how to tweak it into an expander instead of a compressor, *and* easy enough to see how you'd make that frequency-dependent so that it only expanded treble to give a bit more "slap" or "bite" to a note.

[Mark Hammer]

If there's one thing I'm good at, it's taking ONE idea and beating it to death with a stick.