[sdiy] Aaron Crams cem3394 interface question...

Antti Huovilainen ajhuovil at cc.hut.fi
Tue Jun 8 15:28:13 CEST 2010

On Tue, 8 Jun 2010, Harry Bissell wrote:

> I don't think its so easy... the 3394 was intended to be used under CPU 
> control, with charging current and a sync pulse provided to make the 
> "VCO" work. Its really a "DCO" and while this circuit might operate, it 
> will not be temperature (or otherwise) stable. The CPU in most cases 
> also takes care of auto-tuning the circuit.

Yes and no. 3394 has true VCO unlike 3396 (which is DCO). However, you do 
need a tempco resistor, but the same issue is with CEM3374 dual VCO.

If you were to use an MCU with, say, 12 bit dac to control it, you could 
get autotune, while still allowing analog modulation (with the caveat that 
modulation wouldn't be temperature compensated, but who cares if LFO depth 
is 5% off?). Afterall, you're likely using MIDI to control the synth.

An interesting alternative is using digital potentiometer to change the CV 
gain under MCU control and perform tuning at startup / every N minutes.

Btw, you _really_ should use cheap voltage references (TL431, LM336 or 
even a zener) instead of tapping the rails for CV pot reference.


"No boom today. Boom tomorrow. There's always a boom tomorrow"
   -- Lt. Cmdr. Ivanova

More information about the Synth-diy mailing list