Hi Dieter,

Without doubt adding a separate PCB would be the easiest way by far
of modifying the module. I am more than happy for you to do this if
you think it is worthwhile, as I hardly think I can make any 'claims'
on the circuit, as all I've done is copy it from several sources
(Chamberlin's book and Barry Klein's book, both probably sourced from
Electronotes, where it appears at least half a dozen times)!

Unfortunately your assertion that it needs 'no power' is not entirely
true - I mentioned at the bottom of the page that I had omitted two
trimpots, and one of these *is* across the rails. I didn't include
them originally as I didn't fully understand the implications of the
component value selections: however I have just updated the page and
now included them. Not fully understanding the circuit as yet, it is
not clear to me what spread of values will work, but simulations show
that altering them could easily affect the range of adjustment
available. For the circuit to work at all the triangle must swing by
more than the JFET pinch-off voltage, and the circuit needs to be
adjusted to just the right spot, hence the 'shape trim'. At the
moment I'm using a 2N3819: with Vp approx -8V, I've had to slighlty
amplify the triangle to get the circuit to work. However with a
BF245B, it should run directly from the triangle available in the A-
110 (I hope to get a few of these soon, to find out!).

I have the original paper on order from my local library, and when I
get it in a few weeks, hopefully it will explain how the circuit
works, and in particular whether the 150ohm resistors will need
tailoring for the Idss of a BF245B and the slightly lower tri swing
of the A-110 (the other references to the circuit seem to be based on
a +/-10V triangle).

Such a PCB would make modifying the module pretty easy: remove R42,
R52, C9, D1, D2; input to the PCB from one end of where R42 was,
output back to 'O4' to the other; straps for the power and ground
rails are the biggest headache! Setting the 'symmetry' trimpot (which
nullifies the effect of any residual DC in the triangle) can probably
be done well enough by eye using a scope; getting the 'roundness'
right - the 'shape' pot - seems a little trickier (but possibly do-
able by ear for those able to distinguish presence/absence of
harmonics).

Incidentally I also wondered whether the component values for the
existing set-up could be better 'optimised': after a handful of
simulations, and then discovering the other circuit, I was convinced
that that path wasn't worth pursuing further!

Regards,
Tim

--- In Doepfer_a100@yahoogroups.com , <hardware@...> wrote:
>
> Tim,
>
> excellent work. Sounds very interesting - especially as the circuit
> obviously does not need a power supply. Maybe we could built a
small pc
> board that can be added to the A-110 (provided that we get the
permission
> from you). Only three wires are needed (tri input, sine output,
GND).
> Another solution would be a modification of the small output board
of the
> A-110 (adding the additional parts to this board).
>
> Best wishes
> Dieter Doepfer