Problem solved!

It turns out that the filter section simply needed to be re-calibrated. I followed (and corrected along the way) the service manual procedure, and I discovered there was a small DC offset present at the filter output, and also that the resonance VCA and level were off.

There are 2 corrections to the service manual and they are as follows:

All on page 21

Item 6.2 at some point it reads: "...If oscillation level is off, adjust VR3 to obtain 20V p.p at both sides of KLM-355 R10 (with CUTOFF at 10 and RESO at 10)"

It should read instead:"...If oscillation level is off, adjust VR3 to obtain 20mv p.p at one side of KlM-355 R10 (with CUTOFF at 10 and RESO at 10)


Item 7 at some point it reads:"...Then adjust KLM-355 VR4 to obtain 60mv p.p across both sides of KLM-355 R44."

It should read instead:"...Then adjust KLM-355 VR4 to obtain 60mv p.p on one side of KLM-355 R43."

These two corrections correspond to the ckeck points "E" and "F" on the KLM-355 board. And following these corrections, I was able to obtain exactly the specified required measurements and the resonance artifacts disappeared completely!

Oh, and the graph in figure 2 is wrong. The drawing on the left is okay, displaying a 1-1.5v p.p total waveform amplitude. But the one on the right should have read the same with values of 1.5-2.0v p.p. for the TOTAL amplitude of the waveform including the resonance!

The idea being that the filter is supposed to keep a constant amplitude of the signal regardless of the resonance level.

Anyway, thanks Florian for the help!

Alain.

P.S. Of course these corrections apply to both versions (old production and new production) of the schematics, even though the checkpoints "E" and "F" are only identified on the new production schematics.

--- In korg_mono-poly@yahoogroups.com , "Alain" <hubertalain@...> wrote:
>
> Hi everyone! Lately I've noticed that when I "play" the filter in self-resonance with all VCOs off, it doesn't produce as pure a sine wave as before. I can hear what seems to be faint harmonics at different frequencies. And when I look at a spectrum analyzer, I can in fact see some "other" frequencies at very low levels.
>
> When I sweep the cutoff, they bounce all over the place. A little like aliasing in a way. Don't worry I've eliminated everything else by take the output straight from the synth. So nothing else can add artifacts.
>
> Could it be due to failing capacitors in the filter section Especially those directly connected to the SSM2044
> Or could it be something else. I've already replaced the SSM2044 with a more recent one and they both exhibit the same behavior.
>
> It's a very minor annoyance since it's inaudible when filtering a signal from the VCOs or white noise.
>
>
> Thanks for any suggestions you might have on this. And I'd be curious to know if anyone else has these artifacts on their MP4.
>
> Alain.
>