[sdiy] [AH] Re: Minimoog non-linear keyboard response

[Pete Hartman]

Each of the oscillators has its own expo converter.  You might compare the
output of oscillator summers to see that each of them is responding the
same.  If not, then the problem is between the common point of the CVs and
the summer for Osc 1.  If they are the same, then your problem is likely in
the expo converter or possibly the core for Osc 1....

One also wonders if a fault in the high end compensation could possibly
lead to this kind of response.

When you say " slightly out of tune" between and above the high F ...  are
we talking sharp between and flat above, or vice versa?  Or is it flat or
sharp in both places?  Does the octave range of Osc 1 have any impact on
how good/bad the tuning is?  Does an external CV work any better than the
keyboard CV?

Thanks

Pete


On Sun, May 1, 2016 at 5:00 PM, Eric Frampton <eric at ericframpton.com> wrote:

> A follow-up to this:
>
> For starters, I’m on the old oscillator board.
>
> The floating component mod for the key/envelope board has not been done on
> this unit.
>
> It’s only oscillator 1. #2 and #3 seem to be tracking fine.
>
> Furthermore, if I try tuning Osc 1 so the low and high F’s are in tune,
> not only are the notes in between them slightly out (especially in the 3rd
> octave), but the half an octave above the top F is out of tune as well.
>
> I’ve put in a matched resistor set on the incoming CV lines, so that’s
> probably not it.
>
> I’m seeing similar 20mV changes in octaves when measuring pin 6 of the
> summing 741’s on all three oscillators. Not identical, of course, but not
> wildly different on any one oscillator.
>
> e
>
> > On Jul 23, 2015, at 12:15 AM, Eric Frampton <eric at ericframpton.com>
> wrote:
> >
> > …or “unevenly exponential” might be a better way to put it.
> >
> > If I can get the A’s on each end of the keyboard to scale up perfectly
> for each oscillator, and I can get the octaves on all three oscillators to
> switch evenly, what would cause the internal intervals between those A’s to
> not be linear? In other words, if A1 and A4 are correct, A2 might be a
> little sharp and A3 might be a little flat, but it’s enough of a deviation
> that when switching octaves, the oscillators don’t stay in tune with each
> other.
> >
> > The same situation happens whatever octave it’s in - the keyboard itself
> seems to be triggering non-linearly (non-exponentially) internally, no
> matter where the tuning on a given oscillator is set. If the ends are
> right, the middle is wrong.
> >
> > I’ve got a buffer board installed, so I don’t think it’s that.
> >
> > Thoughts? Time for new resistors on the actual pratt-read assembly? Or
> was this what the field service mod on the keyboard PCB was supposed to
> solve?
> >
> > e
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20160501/53f1d4ad/attachment.htm>