Ideas wanted for clever sweep generator for VCF

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

StephenGiles

Well-known member
Joined
Jun 4, 2004
Messages
792
Location
South Croydon, England
This is what I have now:
envgen.gif


With a guitar connected to the preamp (not shown), when a note is played the signal at the base of the transistor turns it on, and its emitter goes high to the voltage set by the Filter Start Frequency pot. The voltage then decays at a rate set by the Sweep Time pot to a voltage set by the Filter Stop Frequency pot. This sweep generator is quite clever because if the Stop Frequency voltage is set higher than the Start Frequency Voltage you get a reverse sweep at the VCF. Sounds familiar? Well yes, Electroharmonix had this down to a fine art 25 years ago. Now I would like to make the Start Frequency voltage be a measure of the strength if the note played, but keep the sweep time as is. Problem - how does one transfer the Peak value of the input signal on to the + input of the Filter Start Frequency buffer IC without interacting with the Sweep time.

What I am ultimately after is a ripple free sweep, starting at a point controlled by the strength of note played. A normal envelope follower cannot provide ripple free operation from a guitar output.
Stephen
 
> I would like to make the Start Frequency voltage be a measure of the strength if the note played, but keep the sweep time as is.

Obviously: Feed the peak detector to pin 10 of the 324.

Tell me why that won't work, so I can understand the problem better.

For low-ripple envelope following: use two or more all-pass filters to give 0, 60, 120, 180, 240, 300 degree shifted signals. Peak-detect all of those, and you get much lower ripple (same as a 3-phase rectifier can give pretty-clean DC without any filter at all).
 
There is nothing obvious about this, otherwise I wouldn't be asking . This is the whole point - if I do that, all we have is an envelope follower, following the input level - result is bad ripple. Shall I put it more simply, what I want is the attack / decay sweep generator to have it's initial attack point determined by the strength of the input signal, but it's decay totally independent of the decay of the input signal. Perhaps some sort of sample and hold could be used to pinch off the instantaneous input peak using a 4016 gate - could be worth trying. I don't really want to start building dozens of all pass filters (unless I'm making a Barberpole Phaser)
Stephen
 
> all we have is an envelope follower, following the input level - result is bad ripple.

If that schematic is accurate: the 1uFd cap will only switch-on the Start Voltage for a split-second at the beginning of the note. After that the sweep is set only by the 2X10uFd caps and the pot.

Or: while it is sweeping, it is clearly ignoring pin 10 LM324.
 
So what you are saying PRR is that the transistor would be acting as a kind of gate as is and should pass the more or less instantaneous peak voltage fed into pin 10 of the 324 and nothing else until the next brief trigger signal hits the base of the transistor - if you see what I mean. I'll have to breadboard this and test it. Thanks.
Stephen
 
> you are saying PRR is that the transistor would be acting as a kind of gate

What it really is, is Mickey Mouse. Dumping a cap into a Base without any resistor is Goofy. And uncalculable and a poor bet in mass production.

But if it isn't a gate, then what is it?
 
Hey PRR - you're up late if your time difference is 5 hours behind London! I do see what you mean in your comment about the gate, The original schematic from which I borrowed the trigger and sweep generator comes from an Electroharmonix toy called a Crashpad, which I believe was primarily aimed at drummers to provide them with an electronic noise when hit with a drumstick - there would have been a small crystal microphone fixed to a pad glued to the top of the case. I'm at work at the moment, but have with me a similar circuit used in the EH Microsynth, a guitar driven processor, which has the same Start/Stop Frequency type of sweep generator, which indeed has a resistor connected between the attack detector and the transistor base. I'll have some time to experiment on breadboard over the next couple of evenings while my wife is out teaching Spanish so I'll see what turns up.
Stephen
 
Back
Top