[sdiy] Jp-8 head scratcher

[cheater00 .]

> ----- Original Message -----
> From: cheater00 .
> To: Maarten Halmans
> Cc: synth-diy
> Sent: Wednesday, November 06, 2013 4:56 PM
> Subject: Re: [sdiy] Jp-8 head scratcher
>
>
> On 6 Nov 2013 16:50, "Maarten Halmans" <diy at artefacts.nl> wrote:
>>
>> Hi all,
>>
>> I have some serious strange thing going on in a Roland Jupiter-8 and I
>> have no more clues where to look. I tuned all 16 voices and they are very
>> nice in tune over all octaves. They are well tuned on start up (in test
>> mode) and when the synth is on for several hours. So far so good!
>>
>> The problem arrises when I use the auto tune. All voices stay tuned except
>> for 2 voices(on the same voice board). Oscillator 1 of both voices is tuned
>> to high. One voice is 12 cent off the other 6 cent. Oscillator 2 of both
>> voices is still in tune. So I swapped the voice card to make shure it is not
>> in de keyboard cv sample&holds or auto tune multiplexers and the problems
>> moves with the voice card. I also disconnected all the modulation sources
>> but all with no result. Swapped the CV summing amp and the transistor array,
>> which does not make sense but you have to try something. I am completly in
>> the dark now. Sounds familiar to someone or does someone have suggestions or
>> ideas the find the problem? They are very welcome!!
>>
>> Best Maarten
>>
>>
>
> Those voices are likely too quiet. Check the peak to peak voltage on the
> scope. The autotune requires a specific output voltage. You wouldn't hear
> the difference in voltage.
>
> D.


On Wed, Nov 6, 2013 at 5:29 PM, Maarten Halmans <diy at artefacts.nl> wrote:
> Hi!
> thanks for the reply! levels are the same and I don't see any strange things
> on the multiplexer input or the comparator output. I recorded the output of
> the comparator during the autotune sequence both with my scope and a
> soundcard, and I can see that tuning of the problem voices takes much more
> time. The 2 lower frequencies seem fine but it the highest frequency ( C8 )
> takes much more time.
>
> Best,
>
> Maarten

Is the HF of the signal attenuated when it gets to the ADC?

Maybe you have too much noise on the digital rail, distorting the ADC?

Right now I'm guessing.

Cheers,
D.