Author |
Message |
iorobyy
Joined: Sep 26, 2005 Posts: 65 Location: Italy
|
Posted: Sun Nov 29, 2009 6:34 am Post subject:
|
|
|
other views
Description: |
|
Filesize: |
856.52 KB |
Viewed: |
519 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
898.43 KB |
Viewed: |
508 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
Description: |
|
Filesize: |
867.09 KB |
Viewed: |
484 Time(s) |
This image has been reduced to fit the page. Click on it to enlarge. |
|
|
|
Back to top
|
|
|
mars
Joined: Apr 20, 2007 Posts: 8 Location: berlin
|
Posted: Tue Jan 05, 2010 2:44 am Post subject:
|
|
|
I'm excited to find other people are doing stuff with the CEM3396. I've got two of them and am trying to figure out how I can make a working circuit with these and an ATmega micro controller.
My first aim is to create a minimal, working prototype circuit with a single CEM.
@iorobyy: your circuit looks great, but even with the comments I struggle to understand it. Sorry. So I have a bunch of questions for you:
1) What are the counters and op-amp for?
2) How come you don't control the filter CV input with a simple 0 - 5v pot?
3) And how do you drive the timer input (I suspect the answer is in the first question).
4) The output goes to a capacitor, and then where? Can you drive a line out connections directly with it?
sorry for the many questions, any answers at all would be much appreciated!
best regards,
/m |
|
Back to top
|
|
|
iorobyy
Joined: Sep 26, 2005 Posts: 65 Location: Italy
|
Posted: Tue Jan 05, 2010 10:58 am Post subject:
|
|
|
Hi!
I have to say to you at first that the pcb you see is a prototype
for my experiments.
I never did a update with the new informations i've found or modifications.
My post is intended as a start point for who want to experiment with this chip using analog solutions.
The driver is based on a 4024 cmos divider...i used this solution for to have various octave from the same driving clock.
The circuitery is very simple , the 4024 and a little transistor driver.
You can found more information in various sites as ken stone - cat girl synth , or the same circuit i used in the neil johnson :
If you want to drive the mithic cem3396 using a micro maybe you'll dont need this transistor , only apply the clock on 4024- pin1.
The vcf controls section is experimental too , the opamp is an inverter for the analog eg...the control voltages in the chip are inverted , differentely than other vcfs or filter chips.
However if you wanto to drive the chips digitally you dont need opamp or other....maybe some cmos inverters.
More...in the output i dont need a buffer/amplifier cause i use this pcb in
a modular system synth that have all the necessary to process all audio signals.
However a simple transistor or phone opamp amplifi-buffer could be used
without problems if you need ....why not!
However a good suggestion is to spend some time to read the datasheet cause the cem3396 is born for digital and maybe the circuitery you need is very very simple...more simple than my pcb.
Uhmmm...you never seen the midibox cem3396 project? ...maybe could be another good start point for your experiments. |
|
Back to top
|
|
|
boothnavy
Joined: Sep 03, 2010 Posts: 2 Location: Boston
|
Posted: Thu Mar 31, 2011 6:21 am Post subject:
|
|
|
Any progress on this project? I am sure people would buy a PCB for this project if it were available. Seems you can still get your hands on a 3396 for not too much... |
|
Back to top
|
|
|
oldmanfury
Joined: Oct 20, 2006 Posts: 57 Location: CA
|
|
Back to top
|
|
|
jojjelito
Joined: Jun 27, 2007 Posts: 50 Location: Stockholm, Sweden
|
Posted: Mon Apr 04, 2011 4:35 pm Post subject:
|
|
|
oldmanfury wrote: | I have a pair of 3396's left from a failed repair job on a M1000. I sold the ailing synth for more than I paid for it, and bought a working white-face one. Lovely synth.
I hope people consider making wave-shaper/filter modules from this chip, rather than going the microprocessor route which in the end will at best create a "Matrix-1". Using a VCO, you'll have all of the slight tuning issues of a VCO and silence the silly critics of the M6/M1000 saying it's a DCO synth.
Anyway - I didn't see anyone posting these schematics. They are overkill for a VC synth, but might have some value.
-gerald |
Interesting that you should say this since the UK-made Cheetah MS6 was running the timer chips off some RC net instead of the crystals used by Oberheim. So, someone has made a less DCO-ish use of the 3396 chip in the past.
/JT |
|
Back to top
|
|
|
Digitalresonance
Joined: Apr 25, 2015 Posts: 4 Location: Uk(london)
|
Posted: Thu Apr 30, 2015 11:53 pm Post subject:
|
|
|
Hi..
Anyone built this? I have a couple of these chips and am keen to have a play with them. Is it possible to drive it with an arduino?
I have been trying to find a simple diy synth project made with this chip but for all the millions of chips I can't find a single project to follow or build..
I thought it would be quite straight forward using these chips as most inputs are just 5v pots.. But I'm struggling with a few things..
Anyone out there willing to lend a hand?
Thanks in advance. |
|
Back to top
|
|
|
jackdamery
Joined: Apr 26, 2010 Posts: 75 Location: UK
|
Posted: Thu May 07, 2015 5:58 am Post subject:
|
|
|
Hey Digital resonance. I also got a widebody 3396 with a view to doing something with it. Want to collaborate on this project? You can email me on {my username} @ hotmail.co.uk |
|
Back to top
|
|
|
Ricko
Joined: Dec 25, 2007 Posts: 251 Location: Sydney, Australia
Audio files: 27
|
Posted: Sun May 17, 2015 7:59 pm Post subject:
cem3396 driven by synth |
|
|
By the way, i think that there is another approach to using the 3396. That is to slave the oscillators to external vcos.
I think you could take two existing vcos, tap/buffer/scale their cv summers to provide the waveshape kcv component, and use the sync outs for the reset pulses.
in this case the 3396 would be just used for waveshaping/filtering/vca not as independent Oscillators. But no mucking around with digitalia.
Rick
(Why tap the cv Summer? So that vibrato and pitch bend & octave etc get added. If you just use the midi kcv + 8ve without vibrato and pitchbend, the waveshape wont change with the wobble entirely. Actually, that may not be unpleasant.) |
|
Back to top
|
|
|
cloudberry
Joined: Jun 30, 2009 Posts: 51 Location: uk
|
Posted: Wed Dec 14, 2016 1:20 pm Post subject:
|
|
|
I've been looking at doing something whith the 3396 also. It seems to me that one can do without all the timing capacitor gubbins at pin 4, since all it needs is a sawtooth wave applied there. The amplitude of the saw determines the waveshape.
The waveshape CV in the datasheet is simply a way of compensating for the varying amplitudes at different frequencies of the timing capacitor.
So one can forgo all the microcontroller stuff; have a traditional CV summer feeding a sawtooth core feeding the 3396, no? |
|
Back to top
|
|
|
Ricko
Joined: Dec 25, 2007 Posts: 251 Location: Sydney, Australia
Audio files: 27
|
Posted: Thu Dec 15, 2016 8:08 am Post subject:
|
|
|
cloudberry wrote: |
So one can forgo all the microcontroller stuff; have a traditional CV summer feeding a sawtooth core feeding the 3396, no? |
Cool! (But it is at the expense of not having the gapped waveshapes, yes?)
-Rick |
|
Back to top
|
|
|
cloudberry
Joined: Jun 30, 2009 Posts: 51 Location: uk
|
Posted: Thu Dec 15, 2016 10:00 am Post subject:
|
|
|
? |
|
Back to top
|
|
|
Ricko
Joined: Dec 25, 2007 Posts: 251 Location: Sydney, Australia
Audio files: 27
|
Posted: Thu Dec 15, 2016 6:16 pm Post subject:
|
|
|
I meant the triangle/clipped waves. But looking at the data sheet again, I think I was wrong: if pins 2 and 10 are not being used for height compensation, they can still be used for the triangle/clipping waveshaping, yes? That is great!
Rick |
|
Back to top
|
|
|
cloudberry
Joined: Jun 30, 2009 Posts: 51 Location: uk
|
Posted: Fri Dec 16, 2016 12:12 am Post subject:
|
|
|
The waveshapes are selected simply by varying the amplitude of the sawtooth presented at pin 4 (datasheet page 6, para 4)
If one does it this way, I'm not sure what pins 2 and 10 need. |
|
Back to top
|
|
|
Ricko
Joined: Dec 25, 2007 Posts: 251 Location: Sydney, Australia
Audio files: 27
|
Posted: Fri Dec 16, 2016 2:33 am Post subject:
|
|
|
Could you use those pins to have voltage control over the amplitude/clipping, rather than varying the incoming sawtooth with some extraneous components? If it is built-in, why duplicate that? (Genuinely don't know, not being snarky!)
Rick |
|
Back to top
|
|
|
cloudberry
Joined: Jun 30, 2009 Posts: 51 Location: uk
|
Posted: Fri Dec 16, 2016 3:49 am Post subject:
|
|
|
Good point!
Honestly don't know. Breadboarding and further thought.... |
|
Back to top
|
|
|
cloudberry
Joined: Jun 30, 2009 Posts: 51 Location: uk
|
Posted: Wed Dec 28, 2016 3:31 pm Post subject:
|
|
|
Doubtful, as the waveshape cv at pin 2 is used to control a current source, which is what is charging the capacitor Cta. No Cta, no point in varying the voltage at pin 2. If someone's got any better ideas I'm all ears!
It seems at the moment, the best thing to do is to use pin 4 as if it were for an external input. I'm not sure if a sink for the current source is necessary, as pin 2 would probably be grounded. |
|
Back to top
|
|
|
stynx
Joined: Aug 20, 2017 Posts: 1 Location: germany
|
Posted: Sun Aug 20, 2017 12:15 pm Post subject:
Using the CEM3396 in an old computer (Apple II) |
|
|
I have implemented a first working prototype of a CEM3396 based synthesizer. It is based on the ALF MC16 music card that uses a 8253 Timer for frequency generation and 3 separate 8bit DACs for envelope.
I modified the layout of the card to allow the connection of 3 CEM3396 chips to this setup without loosing the compatibility to the original ALF MC16 card. This enables me to use the existing player software. The CVs for anything other than log. gain (envelope) are currently manually controlled.
I have to note that the capacitor for the frequency input (5nF in the schematic of the CEM3396 datasheet) are MUCH to big. I could only use 1nF-1.5nF without loosing the lower frequencies. BUT this reduces the possible waveform modifications and narrows some CV.
I will design a small PCB to be used as a card in the Apple II and publish the schematics as well. I think I might be ready at the end of the year...
Here are some samples of the music
https://soundcloud.com/vintage-micro-music/sets/mc96-apple-ii-analog |
|
Back to top
|
|
|
|