Hi,

I think the parameter dimensions of the VCF are not correct. I own two
JX-3Ps and I trimmed them both to identical values before I added the
upgrade to one of them:


1.) The cutoff frequency does not go low enough:
* DCO1 at 8" (DCO2 is off)
* VCF cutoff at 0
* no envelope modulation
* Keytracking at 1V/Oct
=>
At a standard JX-3P the DCO is suppressed completely
At a kiwitech JX-3P the DCO is still hearable (at the standard JX I have
to set the VCF to value 2 or 3 to have a similar result!)

This could be corrected using the trimmers on the voiceboard, but a
correct output value would be better.

2.) The keyboard tracking is wider than on the original JX-3P. This is
not necessarily bad, but one has to get used to it (and if used with an
PG200, then the "100%" marking at the PG200 is no longer valid.

3.) The envelope modulation does not modulate the cutoff high enough
* DCO1 at 8" (DCO2 is off)
* VCF cutoff at 0
* positive envelope modulation at maximum
* no keytracking
At a standard JX-3P the envelope modulates the cutoff frequency to the
maximum.
At a kiwitech JX-3P the envelope modulates the cutoff frequency to maybe
30% of the possible range. (I tried also an external keyboard which
sends velocity 127, but with no change.)

This must be changed.
The positive modulating envelope must open up the filter completely
without additional keyboard tracking of the filter. I assume the values
63 values must be multiplied by 2 before they are added to (or
subtracted from) the total value for the filter CV.

4.) Velocity to envelope amount behaves very weird. If the envelope
modulation is positive, then velocity inverts the envelope modulation
(as if it was a negative modulation). If the envelope modulation is
negative, then velocity has no effect.


Can some one else with two JX-3Ps verify this?


Florian