[sdiy] Sequencer tuning update. Good & bad news.

[Rykhaard D.A.M.I.A.N.]

Firstly, the good news. :)

All 4 of my sequencer steps, with 12 position rotary switches, are happilly
pumping out the notes 'C to B'.

Now - the somewhat bad news.  :(

Not every note within those 12, on each of the rotary switches, is perfectly
tuned to it's respective note. :(  The notes at each end of the resistor
string around each rotary switch, ARE in tune though.

Each of the notes are at least, within Cents, if not exactly on.  So the 4
steps are usable.  I'm concerned about the tuning problems though.

The first solution that I tried, on the first rotary switch, was to reselect
the 1% resistors, manually, to 0.1%.  On the first rotary, I found
121.1ohms to be there mostly, so I hand selected 9 more
121.1 ohm resistors.

The result?  3 or 4 of the notes within the 12 notes around that rotary
switch, are STILL out of tune, by a few cents.  ????  I'm really confused
about this.  It couldn't require picking the resistors to 0.01 or
0.001%could it??  Maybe ...... my wonderings here:

It were suggested that I work with a voltage reference, instead of the
original 'constant current source' that I was going to go with.  The
reasoning, made good sense.  You need the CCS, if you're working with a
keyboard.  The CCS takes care of things, if you are holding down more than 1
key.
I don't have this trouble, as I can only select 1 note at a time, on a
rotary switch. :)

Looking at Barry Klein's modified Electronotes keyboard interface, where a
CCS is used - I noticed that the very last resistor in the resistor string,
is a Grounded pot, that can be used for the overall TUNING of the keyboard.
:O
I can't do this, with my Voltage Reference though.  (An LM329 giving
6.9Vinto a Voltage Follower, to buffer it.)
When I change the value of the last resistor in my resistor string, on each
step, I have to readjust the 1V/octave trimmer again.  If the resistor value
is too large, then my 1V/octave trimmer, can't be adjusted for enough, for
1V/octave. :( :(
That - rules out adding 'tuning' to each step; or adding another resistor
value to subtract 1 octave, from the step's tuning. :(  I would then have to
reset the trimmer resistor for 1V/octave - which I probably could not do -
as it wouldn't have the range. :(

For those in the know ..... ;)  ....... (knowing far more than I do, theory
wise) my exact resistor string for each of the 4 steps is:

- LM329 6.9V reference into an
- LF353 Voltage follower into the
- input of a CD4051 1:8 multiplexer with each step into a
- 5K6 resistor into a
- 1K 20 turn trimmer into the
- 12 position rotary switch, with 12 x 121 ohm 1% resistors around it,
finally into a
- 1K resistor

The control voltage from each step, is taken from the point between the
trimmer and the rotary switch.  (Changed from the previous string setup.)

Can anyone tell me, what I may be doing wrong??  :(  I'm going to try
setting up a Constant Current Source on my breadboard, and feeding that into
the multiplexer, to see how things change, with it.

I've figured logically - that I'm just doing a 'small keyboard' with 12
semitones.  There shouldn't be anything differen't.  Should there?

Final note - I was also told that my resistor value, compared to my 121 ohm
resistors on my rotary switch, should be at least 10 times that value.

I can't do that though. :(  With any total resistor value above the
7.6Kthat my 1st resistor; trimmer and last resistor total to, I'm not
able to
tune the entire string, to 1V/octave. :(  (The total value of all of the 121
ohm resistors around the rotary switch is 11 x 121 or 1331 ohms.  10 times
that - is 13.31k. )

Again - thank you greatly, for any potential help with this - and many
thanks to everyone, who has helped over the past week, or so. :)

The sequencer IS running now, at a usable tuning. :)  But I'd like to get
it, as perfectly tuned, as possible. :-/

--
Take care,
Warmth and Peace,
Ryk

http://deathlehem.bravehost.com/damian.html - D.A.M.I.A.N.;s webpage - Jan.
0406 update
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/20060107/68e86cd3/attachment.htm>