Doepfer

[Apologies to anyone who saw the first version of this - I spotted 
yet *another* mistake, but then found I could delete the post...
For those that know the movie "The Long Kiss Goodnight", I'd like to 
use "Oh darn it, I burnt the muffins!" euphemistically.]

Hi Ingo,

First I must correct two things I said in my previous post (I do so
hate putting incorrect information out there...):

> > As I mentioned before it is possibly that this change will move
the
> > whole resonance peak down to a lower pot setting - if this is
> > happening you should detect that the resonance *decreases* again
as
> > the pot is rotated further clockwise from some given point.

The last part of this, about the resonance decreasing again at higher
pot settings, appears to be complete and utter tosh! That's the
trouble with relying on simulation traces too much, as I have a
tendency to do: the simulation traces climb to a peak, and *do* then
decrease, but since I'm using a SPICE *AC* analysis it is
rather 'analytical', and requires the correct interpretation, which I
have clearly not found yet (I think it is probably to do with the
poles in the right-half plane or something...). In reality it is
clear from my simple breadboard S-K filter, that once it is
oscillating, even as you continue turning the knob past the 'optimum'
point to start oscillations, it *stays* oscillating. In terms of the
gain this is obvious - once you're past the minimum point for
oscillation, you do not go back below it, despite what my simulation
traces might look like! (Doesn't look like too many other people are
paying attention though, so I seem to have got away with it! - "Pay
attention, I'll be asking questions later!") Lesson learned!

My second correction has caused this confusion:

> Hello Tim, the wiper of my prototype is connected to pin 2 of the
> LM324 and an 100pF cap! Actually I thought the prototype would be
> technically the same as the A101-2 in production.

As long as the above means you were mistaken about the 47k going to 
the wiper as you said before (post 13172), I think they are the same, 
as this *was wrong*:

> the right-hand end - if your module *is* as the schematic I have, 
> this end should go to the 47k resistor, and pin 2 on the amplifier 
> chip (LM324), and a 100p cap.

Actually the schematic shows the right-hand pin going to the 47k 
resistor *and that is all*; the wiper goes to the 100p cap and LM324 
pin 2. The second suggestion was to connect these two points (I must 
have gotten ahead of myself), but it sounds like you may not need to 
bother with it anyway (easy enough to do/undo though, with a small 
bit of bare wire).

> To summerize the results: Now the self osc. can be dialed in and is
> absolutly not dominating the overall sound since the original signal
> is always in the foreground. Thus you can no longer overdrive the
> input signal and have a dominating self osc. since you have to dial
> down the input signal to alter the signal/self osc. ratio in favor
of
> the later. In my opinion absolutely no problem for I think the
module
> now much more versatile than before.

I have continued looking at various aspects throughout the day,
mostly improving my SPICE models of 'log' and 'reverse log' taper
pots. Even those these are now not so 'optimistic' as the ones I was
previously using, they do still point to the use of a 'C' law pot as
going in the right direction to remove the rather 'binary'
resonance/no resonance characteristic using the linear pot. I'm
pleased that having made this change it has made the module work
better for you!

> BTW usually, even though not always, a TL074 is prefered over an
> LM324, do you see any reason why the LM324 has been choosen for the
LPG?

I cannot see any glaringly obvious circuit reason for using a higher-
spec op amp, and I guess Doepfer themselves are aware that A-100
stuff probably doesn't appeal to those people who absolutely insist
on low-noise *everything*, so my best guess would be that it simply
comes down to economics - I suspect the LM324 is just cheaper than a
TL074. I might be wrong though, and so I am willing to be corrected
by Dieter!

Regards,
Tim (who really hopes he's got it right this time!)