Robot Pitchshifter (Holtek HT8950)

Started by sundgist, July 08, 2011, 07:51:15 PM

Previous topic - Next topic

sundgist

There are a few threads relating to the DH Pitchshifter using the Holtek HT8950 ic.

http://www.diystompboxes.com/smfforum/index.php?topic=88829.0
http://www.diystompboxes.com/smfforum/index.php?topic=6314.0
http://www.diystompboxes.com/smfforum/index.php?topic=87793.0

I've had one of these on a breadboard for a while and keep going back to it as an experimental/learning project.
I've mentioned in one of the above threads that I've added to and made changes to Dean Hazelwanter's stock schem at GGG.

Anyway.... I got a PM from a member here asking for a schem of my modded circuit. In short, I don't have one.
I've sketched out a snapshot schem of the current state of my breadboard. There is nothing original here, all the stages are taken either from datasheet app notes, Deans schem and bits of other various projects.
All I've really done is added NE571 compander stages either side of the 8950. I found after building the original version that if the input to the chip is too high there is too much distortion and too low causes gating and cutting off of note tails.

The first 571 stage is set up as a auto gain control to squash the input level. The output here is split, some goes to be processed by the 8950, some goes via a pot to the output.
8950 processed output goes to the second 571 stage for expansion and joins the clean signal via a pot. The clean and processed sounds can therefore be blended, useful for certain modes.
The 571 stages started out as a basic compressor/ expander to cut down some of the noise, but I kept tinkering and moving stages around until it's in the state it is now. The clean pot after the first 571 stage should really be at the before the compression but I didn't want to move too much around at once on the breadboard. Besides why not turn down the processed pot and have a useful compressor. the first opamp stage is not really neccessary, just adjust the 1.5k and 39k resistors at 8950's internal opamp.
Some values used were just what I had to hand. There are no doubt some unnecessary components and better ways to do things and I'm sure I'll get around to tidying things up at some point.

Please note, I am using the HT8950 "A" version of the ic. Exactly the same as the HT8950 but with a different pinout and mode switching method. Here is the datasheet for both types. http://dl.dropbox.com/u/15632336/lektronix/StompFX/Robot%20Pitch/HT8950.pdf

As long as I've drawn everything accurately, it will work as it is. But go ahead and change things up and add some more stuff.

I'll update any changes/ additions and try and get some soundclips up.
I've played some nice patterns using the robot mode with the oscillator speed between quarter and midway with some added delay. There are some useful sounds with the octave up and unison with different blend levels and oscillator speeds. A lot of the pitch intervals are really discordant and best used without any clean level. Even so the tracking is not perfect but does make some interesting noise.

I don't think I need to add, this is not what you would class as a high fidelity effect.