[sdiy] Yamaha patents

[jh.]

> I have read with interest about Korg releasing  the schematics of its
> patented filter, and  also I saw  the Moog's  patents collection of  Don
> Tillman.

Are you sure this is patented ? I thought not. If it would have been
patented,
it would not have been a secret for so long.

> Some time ago I was told by Mr. Kitazume of Yamaha (who also was so kind
to
> send me all the remeining spares for free) that they do not have anymore
> the schematics of the epoxy encapsulated modules (which later become the
IG
> chips on the CS80).
> Now I have this idea buzzing in my head: they  must had a patent on them
> and it is laying somewhere, probably in the US too.

I don't expect this to be patented either.
For a clue about the CS-80 VCF chip (and VCO chip, too), please see my
recently added CS-synthesizer page:
http://home.debitel.net/user/jhaible/jh_cs50.html
No transistor level schemos, but quite revealing block diagrams (more
detailed than in the service manual at least).
It's a pair of good old State Variable Filters. If you're asking me, the
secret of the CS-synths' sound is not in the filter circuits. A good deal
is in the filter topology and modulation routing scheme, however.

BTW, while I did not find a patent on the Yamaha filters themselves,
Yamaha have patented their keyboard scaling of the VCF tracking.
Which is a little more tricky than with V/Oct, if you think about it.
I don't have the patent number here, but I can look it up in a few days.

JH.


PS.: There was a discussion on the MOTM group recently about these
        Yamaha filters. The following is what I contributed to the
discussion.
        It's my current point of view - shortly after buying my second
        CS-synthesizer. Some of it might fit into this thread on synth-diy,
        too:

I've been thinking about the CS-60 filter quite some time now.
My *temporary* conclusion is that the *circuit* may be less special
than many people tend to think. What is it - a series connection of
two State Variable VCFs (as in the Oberheim SEM, or in a forthcoming
MOTM module I have designed earlier this year.)

I have recently bought two extra CS-80 voice cards, so I can make
more detailed tests / measurements, but I don't expect too many
surprises.

It's clear that most of the CS-synths' "warmth" comes from its VCOs.
(I just wanted to state that before I go on to filtering.)
Now what is IMO special about the CS-60 filters ?

(1) Limited Resonance Range
The front panel sliders were apparently designed to fit a range that was
intended to emulate acoustical instruments rather than harsh electronic
sounds. Emulate it with SEM-style filters ? Refrain from setting Q to
the extreme ...

(2) Limited parameters for the HPF / LPF pair. You can set the cutoff
frequencies manually to different values, but all the keyboard tracking,
After Touch and ENV modulation is the same for both. This is an
obvious limitation, but it has two pleasant side effects: The combination
of both VCFs will always have a certain "character", rather than providing
unlimited variety. And, more important: If the VCF combination does not
produce harsh sound at the initial setting, it won't during the whole
ENV cycle.
For comparison: With two separate, freely programmable VCFs, you can
have the first stage (resonant HPF) emphasising a certain harmonic, but
not distorting yet, and when the cutoff frequency of the second stage
(resonant LPF) crosses the the cutoff of the first one - bang, overload.
This can hardly happen on a CS-60. Which is better ? Both options
have their benefits. The Korg MS-20 is a prominent example for the
other option. (no SV filters here, but a resonant HPF / LPF pair, too)

(3) The 3rd VCF in the CS-60.
Yes, you're not dreaming. There's a 3rd VCF in the CS-synths. It is
Vactrol based (!) and it is called Expression Circuit. (;->)
The Expression pedal of the CS-60 does not simply work as a VCA.
It's a kind of "loudness correction" circuit which will reduce
the treble at lower volumes. I don't have an original pedal for
my CS-synths, neither for the C-50 nor the CS-60, so I can't really
tell how it behaves with these. But the pedal CV input is normalized
to a certain value, and this is *not* a "neutral" position for all
I know. In fact, for my taste the CS-60 sounds too "dark" as it
is (without the pedal). Yesterday I clipped out the capacitor in
question to get more high end. (This can easily be reduced by
outboard devices - I'm running the CS-60 thru a leslie emulator
at times, which cuts off everything above 5kHz.)

Bottom Line:
I really think you can emulate the CS-60 VCF with a pair
of these forthcoming MOTM SEM-style filters, by just
using the same CV (and same CV attenuation) for modulation
inputs, and by keeping the discipline to limit the resonance
range below a certain position.

As Peter Forrest so wisely stated in his A-Z book:

"The CS Polyphonics are one of these very few types of synth
that can rival the sound of an electric guitar ..."

And I'd like to add, it has limited parameters but immense
immediate control and "expression" from the player - and it's
that "freedom within borders" that's the secret of the CS.

Comments / discussion invited.

JH.