Request: Slow Gear Circuit Analysis

[Paul Marossy]

Has anyone successfully breadboarded one of these? Of all the schematics I've seen, there has always been doubt cast on it or the diagram is incomplete, or with questionable component values. Would be interested to see if there's a verified vero out there somewhere.

I don't know if there exists a vero layout anywhere or not, but I built one about 12 years ago from a PCB layout on a website I don't think exists anymore which used this same schematic. Works just fine...

[PRR]

> why not just us Op-Amp Precision Rectifier?

WHY?

This is not a precision job. We do not need a ruler with perfect 1/64" divisions over the whole length. We only need to know "this high?" If under-level, do nothing. If over-level, turn-down.

The full-wave precision rectifier always needs more stuff. We have to take the reading and *compare* it to the "this high" voltage. We need a one-way valve because we typically use a capacitor for fast-attack / slow-decay. We often need a high-ish input impedance to avoid loading.

In days when opamps cost more than transistors, this 3-transistor affair was very effective, cheap, and small. (It is used in the Orange Squeezer etc.)

[R.G.]

It's about a two hour layout, start to finish, and much of the circuit is a lift from the Dynacomp.

I would be hard pressed not to redesign it as I went along. One dual opamp as two buffers,  another as an envelope gain (as in the current stage) and another to buffer/filter the envelope filter a bit better.

The finicky part is probably tied up in that JFET. As in all JFET circuits, the JFET is the self-centered prima dona that prevents the thing from working. Case in point - the Trem Face. I recently restored two Trem Faces for a friend. One of them had a dead BF244B JFET, and I replaced it with a 2N5485, which has similar Vgsoff, Idss, and other specs. The TF has a trimmer pot to adjust the off-bias for the JFET so the LFO can wobble it around in the hot zone for variable resistance. I fixed the other by replacing the trim pot, but when I started to debug the bad-JFET one, I could not tune it in to be a tremolo. It took me a while to tumble to what was happening - the sweet spot of Vgs was about one nano-turn on the trimmer pot. Miss that one spot, no effect.

In this light, using drain to gate feedback would be a welcome thing, not only for the tolerance of bigger signals and less distortion, but also for making the JFET less $%@^#%$@
sensitive about the sweet spot of Vgs for variable resistance.

[Transmogrifox]

I've always wondered about the function of R19/C12. why do we want spikes driving the transistor?

At this point in the circuit it isn't only spikes driving the transistor.  This network still performs approximately like a high-pass filter until the op amp is cranked so high that everything is clipping -- then it's a high-pass with nonlinear RC time constant.  In this case it probably looks like a series of spikes.  But then, the spikes are filtered by the following capacitors so you don't actually get much for spikes in the gain function.

The attack of a plucked or picked string has a lot of high frequency information in it.  By making that stuff stronger the attack time is faster.

I don't have a SlowGear nor have I built this circuit, but my guess is the effect of C12 is a helper, and not a defining feature of the circuit.

[R.G.]

Probably the best way to understand it is to pop it into a simulator and play with the cap values and such.

[Digital Larry]

D-S-P!
D-S-P!

[duck_arse]

I had the slow gear on the BB some many years ago, worked ok. then it transmuted into a tremolo, I think enveloped. ended up sounding better tremoloing for mine.

[Transmogrifox]

Probably the best way to understand it is to pop it into a simulator and play with the cap values and such.

LTSpice has the ability to input and output .wav files so you can simulate with a collection of real signals if you so desire.

[R.G.]

D-S-P!
D-S-P!

Yeah. People have been chanting that at me for at least twenty years.

My response has been to sing:

I-I-I  know
It's only an-a-log
But I like it...

[Digital Larry]

D-S-P!
D-S-P!

Yeah. People have been chanting that at me for at least twenty years.

My response has been to sing:

I-I-I  know
It's only an-a-log
But I like it...

Ha ha, no problems.  It did make me realize how much "art" there is in analog design, to get things to do stuff that they're not necessarily good at.  I reflected... did I know this trick?  Had I ever known it?  And I think the honest answer was... "I don't think so".

[Transmogrifox]

It did make me realize how much "art" there is in analog design

I have become a much better analog engineer attempting to emulate analog circuits with DSP.  Analog circuits do all the math for us but with a microprocessor you have to define it explicitly.

Simple math and simple programming does not always sound as smooth as something that intentionally creates non-ideal sine oscillator LFO's distorted with typical analog circuit exponentials and parabolic relationships, and filters with a small amount of distortion.

For example, I never really much liked the sound of the straight-forward phaser implementation modulated with a pure sine wave.

DSP designs can be informed by the key principles in good sounding classic analog designs, but doesn't have to stop there.  Once you know what is going to sound good, then the DSP implementation allows precise control.

That said, there is a lot of "art" in DSP design also.