[sdiy] 4069 / 4049 VCA

[Scott Gravenhorst]

media at mail1.nai.net wrote:
>
>>>The 4049 vca originated from Popular Electronics some time (ages) ago.
>>>The connection uses only one transistor of each inverter pair...
>>
>>Tomg's cookbook Hex VCA connects Vdd and Vss to ground.
>
>Is the recipe online??

I don't believe it is anymore.  I wouldn't want to post
it without his permission.  Perhaps contacting him privately
or on the EFM list?

>>So it is attempting to use both transistors.  I don't know if it
>>really makes that much difference given the transistor asymmetry.
>>
>>>All 4049 are UB... there is no buffered version.
>>
>>Ummm yes there is.  I have a National data sheet that says so.
>>There may not have been a B version at the time of that article,
>>but there certainly is now.
>
>Wow!!  You guys really are talking about the CMOS chip.  When I saw "4049"
>I thought that's what you meant until I got to the word "VCA".
>
>Afaik, National has handed over their CD4000 series to Fairchild, who lists
>the 4049 and 4069 with the suffix "UB".

That's good, because the unbuffered ones are the ones that work
in linear mode.  4069, I'm told, has symmetrical transistor
pairs which is more useful for linear mode.  And then there's
4007 which also has interesting properties.

>In the Harris book I have, both have the suffix UB.  The 4049 is is an
>inverting "hex buffer/converter", and the 4050 is non-inverting.  The
>CD4069UB is simply described as a "hex inverter".
>
>The latest Digikey catalog lists both a TI CD4069UBE and a CD4069UBM as
>"NEW!".  I don't know if that means they're new to Texas Instruments or new
>to Digikey, or what the difference is.  If I had to guess, I'd say the "M"
>is for military spec.
>
>>Perhaps the idea of an _all_ 4069 synth is not quite practical.
>>I have strayed from that anyway in my building since I have used
>>4007 and 4027 as well as standard opamps (the linear V-I converter).
>>At this point, my project is a "Mostly CMOS" synth, with the idea
>>that many of the 4xxx parts are still common and inexpensive.
>
>d00D!!  You're building a CMOS synth??

I sure am trying.  An _almost_ all CMOS synth.  I had to use 
opamps for the V-I converter, but other than that and regulators
plus a few scattered bipolar transistors, it's all CMOS IC.
The VCO is a René Schmitz design to which I've added a triangle
waveshaper and a triangle suboctave, both fully CMOS.  I discovered
that a 4069 gate operated in linear mode does not hard clip when
overdriven, rather it soft clips turning the triangle into something
more like a sine wave.

>Hmmmm....maybe that isn't as insane as it sounds.  You can get triangle and
>square waves out of a 4046, a sub octave with a 4027 flip-flop, some sort

My triangle suboctave is made of a 4027, 4069 gates in linear mode
and a 4066.

>of "ring mod" from a 4051/52/53 multiplexer, 

I'm thinking of goofing with a 4069 for a ring mod.
I haven't started anything on it yet.

And let's not forget the digital version of a ringmod
for square waves: 4070 - XOR gates.

>and I guess there are a number
>of ways making a VCO by making the control voltage the supply voltage, but
>I have no idea how you can make an exponential converter, 

Well, I use linear synths, so I'll let someone who knows
about expo...  I won't say it's impossible.

>VCA, or VCF, or

VCA should be some morphodite or other of a ring modulator.
I will also be messing with 4069 based VCF designs.
I don't know if they will work, work well or be 
practical yet.

>make an ADSR that doesn't rely on RC networks for timing each stage.  

The Magnus design, which is so far untested still uses RC, but 
driven by 4069s for voltage controlled resistors, amplification
and Schmitt triggering

I also have a working triangle LFO made from 1/2 of a 
4069.  (dual LFO!)

>That
>certainly doesn't mean that it can't be done, it's just that I have no idea
>how the insides of these chip work.  For all I know, they could contain
>tiny colonies of superfast ants.

Don't they?

>Anyway, if you have any ideas, I'd love to hear them as most 4xxx chips are
>inexpensive, and I already have a bunch of them lying around from various
>sequencer clock projects.  Well, that's if the ants haven't died.

If you've kept them frozen, they wake up when thawed...  (c:

The availability and price of these chips was one of the 
attractors for me.  I think it was Magnus who started the
ball rolling, then René chimed in with what I think is a
pretty nice saw VCO.

Be warned, however, as I was reminded by Happy Harry,
CMOS doesn't translate into low power when the gates are
used in linear mode.  I'm finding that my V+ regulator is
getting warm.  It has a tiny heat sink and poor ventilation,
but there is definately some current draw.  It's a 7808
T-220 package running the VCO, LFO, and both waveshapers.
There are quite a few linearly operated gates involved.
I think with a decent heat sink and ventilation, it should
handle at least triple what it is now.  But I think batteries
are pretty much out of the question.  Well, maybe a car 
battery...

=========================================================  
- Government: The other religion.
- The media's credibility should always be questioned.
- Lambs who lie down with lions are lunch.

-- Scott Gravenhorst | LegoManiac / Lego Trains / RIS 1.5
-- Linux Rex         | RedWebMail by RedStarWare
-- FatMan: home1.GTE.NET/res0658s/TekLab_FatMan_WebSite/index.html