[sdiy] saw core VCO tuning/scaling (Roland SH-2)

[Magnus Danielson]

Adam,

On 18/08/11 09:23, Adam Inglis wrote:
> I realise this is probably a timeworn query - my apologies to the veterans.
>
> I'm having some trouble with my SH-2 (dual VCO monosynth 1979):
> I've had it up on the bench a couple of times to calibrate VCO 1. VCO 2
> comes up sweet.
> I leave it on for an hour, follow the SM routine ("width", then
> "linearity" then footages) and think it sounds good, check it with a
> tuner, but after I put back in the rack it goes out (of tune but also
> scale width) after an hour - so I guess it is temperature related.
>
> The key CVs are spot-on, and the supply voltages seem solid. VCO 2
> remains fine. Measuring the voltages after the CV summing op-amp gives
> equal values between octaves. From that point, I'm not sure what to look
> for with my ancient scope and my Fluke - it's current-dependant, is it not?
> The expo converters are metal can uA726 - they both run quite warm to
> touch.
>
> I've read Ian Fritz's and Magnus Danielson's info regarding these type
> of Electronotes-based VCOs, but if anyone can point me to some
> troubleshooting articles, that would be most appreciated,

OK, by strange coincidence I have been discussing troubleshooting of 
uA726 based oscillators today, and then my name even pops up here, and 
it can't be recent I wrote much about the topic. Ah well.

The uA726 is unique in that it contains an oven. First thing to check is 
if the oven is running and if the oven setting resistor is properly 
hooked up. There can be other sources of this instability.

I haven't checked the SH-2 schematic, but if someone toss me the link to 
a good schematic I might help a little more.

Cheers,
Magnus