Discussion about the Korg PolySix synthesizer

(This is crossposted to both vintagesynthrepair and polysix groups).

I'm currently trying to calibrate a Trident (Mk I, not Mk II). The overall
shape is good and everything seems to work somehow.

Problem is that I cannot get the antilog amp to calibrate by following the
directions from the service manual and I am not completely convinced that
the service manual is correct.

The procedure starts as follows:

1) Measure antolog output with a digital multimeter, connecting the GND
side to -11V and the positive side to the exposed lead of R57 (47 ohms,
connected to IC 4558 no.2 pin).

...and this is where the fun starts. As far as I can see there is no -11V
supply to be found on the whole board; and I find it rather unusual to
create a voltage divider first to do any measurements. In addition R57 is
not connected to pin 2 of the 4558 at all, it is connected to pin 6 (and
the other side is connected to the input of the demultiplexer which I
assume is the actual test point in question).
On the Trident Mk II at this very point (the leg of the 47 ohms resistor
that is connected to the demultiplexer input) is TP 4 located so I'm using
this as the test point here as well.

So let's continue:

2) Play C3 eight times and adjust total tune to obtain 6.00mV (+/-0.1mV)
antilog output. The total tune knob should be within +/-1 scale marking of
center.


"total tune" obviously means the master tune knob on the front panel.
However I guess the 6.00mV figure is wrong and should actually read 600mV.
Interestingly the directions are not very helpful if you cannot get to this
value at all...

3) Set VCO scale to 4'.

...for whatever reason because it doesn't affect the antilog amp at all...

1. Play C6 eight times and adjust VR-6 to obtain 4.800V (+/-0.003V).
2. Play C3 eight times and adjust VR-7 to obtain 600.0mV (+/-0.3mV).
3. Play C1 eight times and adjust VR-5 to obtain 150.00mV (+/-1mV).
4. Repeat steps 1 through 3 as many times as necessary to meet tolerances.

So I'm rigging a voltage divider for the -11V and try to get as close as
possible to the figures for steps 1 to 4. But it doesn't work. I'm getting
close but there are always a few mV missing for the second and third step.

However: When I got the Trident it was more or less in tune. With a bit of
fiddling I can get the antilog amp calibrated by ear so that the
oscillators track quite well over the whole keyboard at this point (the
oscillators need to be calibrated as well, but the antilog calibration
naturally should come first).

So now I mess around with my voltage divider so that the measurement for
the second step now reads 600mV proper. It turns out the other measurements
are in the ballpark now as well. However the voltage the GND connector of
the voltmeter is connected now is about 11.15 volts instead of the 11 volts
required.

So here are the questions:

1. Am I doing something wrong?
2. Is the service manual to be trusted here at all?
3. Does this behaviour hint at a possible problem in the antilog amp or is
it merely an issue of "how to calibrate"?
4. Does anybody have the directions for the Trident Mk II, and are those
the same?

Thanks and best regards,

Malte