Sunday, March 31, 2013

Simple MFOS LFO & Acid Etched Front panel

After all the trouble I was having with the 8038 Chips, I decided to build a new LFO module based on a different design.  Early on, I tried building the MFOS super simple dual LFO but I couldn't get it to work.  I decided to try it again, now that I'm better with electronics and it worked fine.  In retrospect, the mistake I made the first time-was most likely my interpretation of the schematic. Notice the inverting and non-inverting inputs on IC1-D and IC1-A are flipped, so if you are not looking carefully, you may wire them incorrectly.

I always breadboard my projects first and especially when they might need to be modified because they supply voltages are different.  This schematic worked well but I changed a few of the values based on what kind of components I had around.  I changed out R7 with a 20k resistor to bring the square wave output up to +/-10v, so I can use it as a clock signal.  I changed R90 to a 2k4 resistor because I didn't have any 3k resistors.  I also changed out R8 with 1K resistor because when the frequency knob was at it's farthest left point, the wave would stall out, I don't know if this had something to do with using +/-15v vs +/-12v.

Sunday, March 24, 2013

"Gate keeper" MorphLag Portamento and envelope generator

UPDATE!!! 8/19/13  I just finished a newer version of this module.  I changed some component values and ended up with a better version of this design.  I also added in a gate activated on/off switch.  it will be posted soon. 

I was super excited when I found the Schematic for this simple but versatile portamento circuit.  I Had an idea similar to this, a little while back when I was building the Ian Fritz AD/AR generator.  I noticed then, even the Fritz AD/AR was similar to a simple lag circuit, and it put the idea in my head for a simple portamento with seperate controls for up(attack) and down(decay) slide.  Nerds will notice right away that this can also be used as a simple envelope generator. 
The original circuit can be found here.  Below is the version I printed and made some modifications to.  Also, I clearly spilled coffee on it.  Since the schematic already used three op amps,  I decided to go ahead and use a forth for the indicator LED, which is probably unnecessary, but turned out really cool.  I thought it would be cool to have one led correspond to the attack and another correspond to the decay.  I tried a differentiator op amp set up, comparing the input voltage to the output voltage.  That schematic is in the bottom right hand corner. 

I breaded it up and everything worked great except for, when the shape knob was at its most ccw point. It was doing this weird self oscillation thing.  I rectified that by adding a 240Ω resistor.

I also added a bypass switch.  

Thursday, March 21, 2013

4017 "baby sequencer" 10 Step Sequencer

UPDATE 6/1/13!!! After playing with my sequencer for a little bit,  I've noticed some stuff I'd like to change about it.  A redesign in the near future which will include a switchable gate almost identical to the Gate Sequencer I built around the 4017 in this post.  I also noticed that you need to convert the incoming clock signal to a very very short square wave,  I am probably going to use a "gate to trigger" circuit similar to ken stones, which I think is basically a differentiator.  Otherwise when the square wave (clock signal) falls from High to Low, the output voltage of the 4017 changes just enough to make the notes flat.  expect a newer design soon!!
This is my first attempt at a sequencer.  It's built around the 4017 Decade counter chip.  It's my first foray into using CMOS's chips (unless you count the CMOS 555 in the ADSR).  The design I came up with is an amalgam of the Baby Sequencer and various sequencers on MFOS.  I was trying to get away with as little circuitry as possible.  This version of the sequencer doesn't include a gate output.  So you can sequence control voltage only, you can't skip steps.  My plan was to build a seperate gate sequencer, which I thought would stretch the 10 steps farther.

I did a mock up of my design using little trim pots, just to test the basic concept,  it seemed unnecessary to do this for all 10 channels. 

So,  it's really straight forward design and I never even made a schematic for it.  I went straight from breadboard to PCB.  so, I don't have a schematic to show you,  but the PCB is incredibly easy to follow, so we'll work from there. Read the whole post before you build anything. 

Parts list
1x CD4017
1x TL072
10x 3904 NPN trannies (or equivalent)

10x 100k Pots
1x 50k Pot

2x 1/4" jacks
1x Rotary switch (10 or 12 position)
2x momentary switch (you want ones with a throw, not just open and closed)

22x 100k resistor
10x 680Ω or 1k resistor
1x 10k
1x 100Ω
2x 10Ω

2x 22uf electrolytic cap

100K POTS: Lets start with how the pots should be wired.  I found this little picture online so I can refer to the pot pins without people getting confused.  The pot output from the board for each channel should be soldered to the CW of each of 100K pots.  The Center pin (labeled S) should be wired to one of the many returns above (the returns are all the points with 100k resistors, it doesn't matter which one, they all get mixed together).  the CCW pin should be soldered to ground.

50K POT: you need one 50K pot for the "Scale Knob".  Solder the CCW pin to one of the two scale pins on the PCB.  Solder the Center Pin to the other point.  Doesn't matter which one. 

LED: The LED point on each of the Channels should be soldered to the positive pin (longer pin) and the Negative side to GND.  I listed 680Ω resistors, but I ended up using 1K resistors with green LEDs.  

next up is the clock input and reset and all that junk.  this can get confusing.

CLOCK: Solder clock point from the board to the tip of a 1/4" jack, with Gnd to Sleeve.  I found later, that depending on what you use for a gate signal, you may want to solder a diode from sleeve to tip (positive end toward ground and negative end to tip),  this will stop all negative voltage from flowing into the clock.  So, if you use a square wave from an oscillator as a clock signal for example, it will eliminate the negative crests, which I noticed causing the CV voltage to change when the wave swung from positive to negative. 

RESET:  Solder a wire from the CW pin of each channel pot(except for channel 1) to a pin on a 12 or 10 position rotary switch (in order), any remaining points on the rotary switch should be soldered to ground.  Solder the common pin of the rotary switch to the Reset point on the PCB. 
OPTIONAL RESET BUTTON:  you can also add a push button if you'd like to be able to reset the sequencer with a button.  Disconnect the Common pin of the rotary switch from the Reset point on the PCB and Solder it to the Normally Closed pin of the momentary switch. Solder the common pin of a momentary switch to the reset point on the PCB.  Solder the +15V point from the PCB to the Normally Open point of the switch.  essentially, when you push the button, you disconnect the rotary switch from the Reset point on the PCB and Connect the Reset Point to +15v, causing the sequencer to reset. 

CLOCK ENABLE:  This allows you to pause the sequence.  Connect the Clock Enable point on the PCB to the common pin of a momentary or toggle switch.  Connect the Normally closed pin of the switch to Ground and the Normally open pin to +15v.  

In retrospect.  i think it would be nice to have a reset input.  that would be added in parallel with the reset button, but with a diode to keep voltage from flowing backwards through the rotary switch to the channels. 

let me know if you noticed any mistakes.  I am a novice. 

Here's a video of it in action with my yu synth wavefolder

and here's the PCB

Wednesday, March 6, 2013

8038 Triple LFO module

I would avoid building this module.  I ran into so many problems trying to find 8038 chips that I liked.  I decided instead to build a different BETTER LFO module.  that post is HERE.  If you have some 8038 chips laying around with the correct numbers on the, then, this could work for you.....

I recently built a pretty simple but very useful LFO circuit with the ICL8038 chip.  I decided to go ahead and build a triple LFO module in a double wide panel.  I took the time to make a nice PCB for the circuit with fat traces and solder pads.  When I finished and assembled my module, I was surprised to find that two of the three circuits didn't seem to be working correctly.  They had a limited range.  after checking all my traces, I realized that the two which were not working correctly were newer chips from a different batch/seller.  The one that was working was from the first module I built.  ultimately, I decided that the different chips did not work with my circuit.

this chip works with the circuit

and this chip doesn't


I tried ordering more chips with the correct numbers based on the picture of the chips on ebay and they sent different chips that didn't work.  I'm really annoyed with the 8038 chip situation. So, I'm going to revisit this LFO project with a different circuit,  one that doesn't utilize the 8038. 

So,  go figure.  Anyway,  here's the PCB.  I'm not going to spend a lot of time explaining this circuit because I did so in an earlier blog, HERE

So,  I've included the single LFO parts reference because it explains the how to wire the two potentiometers.

The triple LFO is on the right hand side.  In the center is my sequencer which will be included in the following blog.