ASM1 mods/Electronics project

[Magnus Danielson]

From: "Andy Main" <mainas at aston.ac.uk>
Subject: ASM1 mods/Electronics project
Date: Sat, 2 Oct 1999 10:04:49 +0100

> Hello everyone,
> 
> I'm new to the list and was wondering if anyone can help me with my ideas.

You are wellcome!

Speak out! (and by magic you did)

> I'm planning to build a synth for my final year electronics degree project.
> My first idea was to model an analogue synth using a dsp and stuff, but then
> I remembered that I hate programming, and unless its nice and simple it never
> works.  This is why I'm an analogue freak (why is there never the jobs for
> analogue guys?).

(Comment on your mailer: Make it automatically break lines at 72 charaters, the
above appeared as a single line for me and probably others.)

Ah, so you went with analogue after all, then you are certainly wellcome ;)

> Anyway my latest thought is to build an ASM1 (great looking synth from the
> design).  However this is not enough, so I'm thinking of the following mods:

The great thing about the ASM-1 is that it is fairly open in it's design.
Only in a few places does it need to be modded to be totally CV controlable.

> 1.    Full midi-CV implentation (of my own design).
>
> 2.    A virtual patching system - like the nord modular in a way but using
>       ADC's to pass a digital values of the output of one module to the DAC
>       at the input of another module.  Thus keeping it mainly analogue.
>       Note this system can be bypassed for true performance using bananas
>       (not the yellow edible ones!).

I have been thinking about a similar thing, a virtual patching system allrigth,
but with these nice 16x8 fully programable CMOS switch blocks that you can buy
(from Harris Semiconductor among others). They are basically a matrix of the
size 16 x 8 with a CMOS switch sitting in evety junction. This makes 128
switches. There is a digital control interface with a 7 bit address and a 1 bit
data (and some control signals) so that you can individually set or reset
every switch. With this you have the full freedom.

Now, the resistance of a switch is fairly high, so I planed to have input and
output buffers in order to isolate the environment. These chips are fairly
cheap. Buffering will add to cost thougth.

Actually, (feeling a bit nasty here) one could maybe even have the matrixes
to do 1:1 mixing. However, the nonlinearity of the switches may not be to
advantage here... or maybe it will, it's a synth after all, not a "pro-hifi
system".

I don't mind digital control, but why go over digital system with analog
signals?

> 3.    An option to switch any of the ports to linear frequency CV for use
>       with other modules.  Note this won't be true analogue but a digital
>       conversion.

The VCO has both linear and exponential CV. The VCF could be patched to have
linear CV input. This way you would just select input.

> 4.    Patch storage for the virtual patches, and controller data.

Good idea, it comes natural with a virtual patching system, so offcoarse have
I got it too...

> Questions:
> 
> Is this going to work - or will there be a problem when connecting mutiple
> patches into a module (loading won't be the same as before)?

You could probably make it work well. If loading is a problem, just trow in
buffers.

> What sort of ratio do most linear CV's work at?
> 
> If you don't like the above suggestions (I've been kind of dreaming what
> would be good to have), can you suggest anything better?

I think you got some feedback allready.

In any case, don't forget to put your work up on the Internet, so that I
can link it in from the ASM-1 homepage ;)

> Thanks for any help.
> 
> Andy Main
> ---------------
> 
> PS - Can anyone give me a job/recommend a company in analogue (preferebly
> musical) design in the UK when I graduate next year?

There are many companies in the UK that I can think of, SSM, Calrec, BBS,
Turbosound, EMS...

Isn't it time for EMS to bloom again?

Cheers,
Magnus