SV: SV: [sdiy] A book about analogue synthesizer circuits?

[Richard Wentk]

At 10:54 04/04/2003 +0200, Thomas Kahn wrote:

>My suggestion is that you do it the other way around: top-down. Select a
>number of _complete_ projects with schematics, card-layouts, parts lists
>and building instructions. After that you start explaining the different
>solutions that are a part of the circuit the reader has just built (or
>at least read about). After that you go even deeper describing how the
>properties of the different components are utilized and how they work
>together. After that you dig into the theory with formulas etc.

I'm not sure this would work. The main problems are:

1. All the kits would have to be electrically compatible in terms of 
voltage ranges, PSU requirements, and so on. So really the book would have 
to offer a standard modular design. This sounds like a good thing but I 
think it's a bad one because:

i. We already have that with the ASM1.
ii. I think it's more interesting to use the book as a focus for a 
community of new designers, rather than those who simply want an 
off-the-shelf modular and are prepared to do a bit of soldering to get it.

2. Kits are always problematic because you have to supply at least a PCB 
and a bill of parts, and preferably a complete set of components. The bill 
of parts issue will cause problems depending on the countries where people 
live. E.g. as someone based in the UK, a Mouser parts list is a lot less 
useful to me than one from Farnell and RS or even Maplin.

Now, you could say that instead of being a rigid book it's more of an open 
source collaborative project, and people can contribute things like a 
local-equivalent BOP. I think that's probably a more useful approach than 
something that's more like a paper book.

3. I think it's *much* more interesting to split the project up into 
functional headings, and then allow any kind of information to be uploaded 
to each one, with possibly some editorial tidying and organising if things 
get out of hand. So again it's more like an open-source publishing project 
rather than a book. (And also more like the original Electronotes.)

So I'd imagine something like:

The basics
         Information about components, soldering, and so on
         Details of (say) cap types and what they're best for
         Thermal control
Sound generators
         Assorted VCO types and circuits
         Noise sources, digital and otherwise
         Physical modelling and other digital approaches
Filters
         Diode, tranny, and other VCF designs
         Tone controls and fixed filter banks
         Para and graphic EQ circuits
Modulation sources
         EGs, step sequencers, LFO designs
Modifiers
         VCAs, waveshapers, etc
Effects
         Delay lines, reverbs...
Other
         Ideas that don't fit anywhere else, e.g. vocoders
DSP techniques
         Doing it digitally, both h/ware and s/ware
         Converters
Hardware
         Case options, panel design options, pros and cons of different 
connectors
Resources
         Links to other information, including app notes
         Commercial links, so if you sell kits or hardware, you can be 
listed here
Sound programming
         Fun patch ideas for modulars and digital systems like Csound
Retro
         Retrospective info about specific existing designs - e.g. 
discussion of your fave Moog, Oberheim and Roland circuits
         CEM/SSM chip notes

Each section would be a mix of theory and practice, perhaps like Rene 
Schmitz's fine articles about tempcos. So beginners could find enough start 
to build, and the experts can also trade design ideas. And unlike many 
books, I think the emphasis should be on practical hands-on engineering 
smarts, so that any theory is actually useful and not academic.

Making each heading open to any kind of contribution would make the whole 
thing an evolving resource. And if you attach a message board to it you can 
also create a place where beginners can ask questions - which is not 
something that's usually possible with a fixed book.

And of course if people want to upload entire kit designs, they'll be 
welcome to do that.

Appointing a couple of moderators to look over articles before they're 
uploaded should keep the signal to noise ratio high and also nix any 
possible trouble from spammers.

Ideally the whole thing would eventually turn into the definitive one-stop 
modular resource.

So - bottom line would be:

Mix theory and practice
Make the project on-going, open-ended and open-access
Keep it electronic. I suppose there might be some mileage in doing a 
best-of paper collection, but organising something like that is a lot more 
hassle than keeping it on the web.
Inclide MP3s, PDFs and text. (Which again is hard to do in a book.)

Richard