Wow, that's very useful info! Thanks Johannes!
The control voltage must be crucial for the expo conversion to work
properly then. If any of the amplifier circuits would be broken
(IC17, 18 or 19), could that cause this behaviour do you think, or
would that just make everything fail "from start"?
Well, speaking of which... The synth no longer tunes up correctly
after 20 minutes. Now the tuning and the note intervals are screwed
up period. So whatever needed "warming up" a few days ago seems to
have failed permanently now.
I just went over the soldering points around the components you
pointed out and resoldered most of them to be sure. I still get the
same problem, though, so I guess it's time to start replacing
components. I'll definitely try testing with cooling spray. Good
suggestion!
BTW, it seems I can replace all components in the expo conversion
circuitry easily except Q12, Q3 and PC-1. Can't seem to find info on
these anywhere. You wouldn't happen to know of components compatible
to any of these three, would you?
Peter
--- In
PolySix@yahoogroups.com, Johannes Hausensteiner
<jo.synth@c...> wrote:
> The hold capacitors are C56 in unit0..5. But I do not think that it
> is the hold capacitors very likey. If it were, not all six voices
> would be affected equally. No, it must be something common. If it
> is not the wires to/from KLM-396, then the "expo converter" is my
> no.1 candidate. KLM-396 seems to be a part of it. The KBD works
> logarithmically; that is going up a certain number of keys (e.g.
> 12 for one octave) multiplies the frequency of the generated note
> by a certain factor (times 2 in case of an octave). So the KBD CV
> logarithmic by its nature, whereas the oscillators work linear;
> that is, the generated frequency changes linearly with the control
> voltage (e.g 1V gives 1000Hz, 2V give 2000Hz, 3V give 3000Hz,
etc.).
> These two schemes are not compatible to each other but it is
possible
> (and necessary) to convert from one to the other. When converting
> from logarithmic to linear you have to do the opposite of logarithm
> which is exponential function. It happens that diodes and
transistors
> have exponential transfer curves, so they can be used for this
> purpose. But the exponential function is extrem sensitive to small
> variations; a small variation of the input causes a big change at
> the output. Furthermore the human ear is very sensitive to changes
> in pitch. This means that there must be taken special measures that
> the whole thing does not run out of tune.
> The expo converter in the Polysix is made of IC17, 18, and 19, Q12
> (the expo component), PC-1 (kind of an opto-coupling device), Q3,
> and surrounding components. In case of "old production" units
> KLM-396 belong to this circuitry, too. Polysix has 6 voices but
> the analog (de)multiplexers for the CV have 8 inputs/outputs. The
> additional 2 inputs/outputs are used for some kind of calibrating
> the exp. amplifier and thus stabilizing tuning of the instrument.
> I am not 100% sure how this works as the CPU never measures any
> value and therefore cannot adjust to changing values. From what I
> can see from the code there is always the value 0 ouput to "voices"
> 6 and 7.
>
> According to what you tell, it may be a thermal problem. So I would
> carefully check all solder points and try to heat up / cool down
> the PCB in the area where these components are located. With
cooling
> spray you may identify a faulty component.
>
> Johannes
>
>
> Peter Andersson wrote:
> > Yes, it's an old production unit alright. I've checked the wires
> > between the 366 and the 396. It seems ok, the solder joints too.
> >
> > Can you tell me which major components are actually involved in
the
> > lin/log conversion/tuning? I've been told to "check" the key
voltage
> > selector chip, the DAC and the S/H capacitors as well as
something
> > called an expo converter. (I don't understand half of this stuff
to
> > be honest - I know you're not surprised. I'm learning, though...
> > quickly!). These are some of the key components of the 366 as
far as
> > I understand. Do they handle the conversion/tuning you're
refering
> > to? I thought I'd start out by replacing the S/H capacitors. The
> > cheapest and easiest operation to start with, I thought.
However, I
> > haven't even been able to locate them. It's nearly impossibe to
read
> > the numbers on the 366 schematics I got off the Old Crow web
site.
> > (Man what a fantastic resource his site is, btw!).
> >
> > Again, thanks for helping!
> >
> > --- In PolySix@yahoogroups.com, Johannes Hausensteiner
> > <jo.synth@c...> wrote:
> >
> >> From what you tell I would say it is a problem on the KLM-366,
in
> >>the log/lin conversion/tuning section.
> >>Or maybe you have one of the rare "old production" units. Is
there
> >>a small add-on board "KLM-396" to the KLM-366?
> >>It could be a lose wire (if "old production") and/or a bad solder
> >>joint.
> >>I can see that you are located in Sweden; maybe you ask Ricard
> >>( http://home.swipnet.se/ricard2/p6index.html ) for technical
help.
> >>
> >>
> >>Peter Andersson wrote:
> >>
> >>
> >>>Actually, tones are quite stable even during the first 20
> >
> > minutes, but
> >
> >>>the synth is not in tune generally. Also, an octave is a quite
> >
> > stable
> >
> >>>major 7 interval for the first 20 minutes, until suddenly, the
> >
> > synth
> >
> >>>goes into tune and then the octave is correct as well. I'll
check
> >>>tonight if a single generated tone changes a lot during the
> >
> > first 10
> >
> >>>sec.
> >>>
> >>>All six voices are affected. If I play a chord when it suddenly
> >
> > slides
> >
> >>>out of pitch, all notes follow. It actually sounds like I'm
> >
> > turning
> >
> >>>the pitch wheel. (I disconnected the pitch wheel, btw, just to
> >
> > be sure
> >
> >>>it wasn't playing tricks on me).
> >>>
> >>>It is not related to the MG setting and I've played around with
> >
> > all
> >
> >>>three - VCA, VCF and VCO. I'm quite sure it's not affected by
the
> >>>waveform setting either (whether I use a saw or a pulse) but
I'll
> >>>double check that tonight. The PWM setting does not make any
> >>>difference.
> >>>
> >>>Does this give you any clues? Thanks for helping!
> >>>
> >>>
> >>>--- In PolySix@yahoogroups.com, Johannes Hausensteiner
> >
> > <jo.synth@c...>
> >
> >>>wrote:
> >>>
> >>>
> >>>>It is not normal that it takes 20 miuntes to tune up. Usually
it
> >>>>takes about 10sec to tune and then it is quite stable.
> >>>>Check the following: hold a key down when switching power on.
The
> >>>>generated note should change quite a lot until it stabilizes
> >
> > after
> >
> >>>>a few secs.
> >>>>Are all six voice affected? Is the effect related to the MG
> >
> > setting
> >
> >>>>(VCA/VCF/VCO)? Is it related to PWM setting?
> >>>>
> >>>>
> >>>>Peter Andersson wrote:
> >>>>
> >>>>
> >>>>
> >>>>>Hi all,
> >>>>>
> >>>>>I'm the proud new owner of a Polysix. I love it! Can't take my
> >>>
> >>>hands
> >>>
> >>>
> >>>>>off those knobs - the filters are just great! But I'm having
a
> >>>>>problem with it and I'm hoping someone can tell me what it's
> >>>
> >>>about
> >>>
> >>>
> >>>>>and maybe even what can be done...
> >>>>>
> >>>>>After I've played the synth for a while, notes start to go
out
> >
> > of
> >
> >>>>>pitch every now and then. It happens randomly every 10-30 sec
or
> >>>
> >>>so.
> >>>
> >>>
> >>>>>The notes go off pitch with a note or so and always jump or
> >
> > slide
> >
> >>>>>back to the original pitch again within half a second.
> >>>>>
> >>>>>At first I thought it was some filter setting of mine that
> >
> > caused
> >
> >>>>>it, but as it turns out I get this no matter how the Poly6 is
> >>>>>programmed (or if I use a preset or a fresh "manual" sound).
(By
> >>>
> >>>the
> >>>
> >>>
> >>>>>way, it takes about 20 minutes for the unit to get in perfect
> >>>
> >>>tune
> >>>
> >>>
> >>>>>after a cold start. Don't know if that's normal or if it's
> >>>
> >>>related
> >>>
> >>>
> >>>>>to this problem in any way).
> >>>>>
> >>>>>I'd be truly grateful for any help!
> >>>>>
> >>>>>Peter
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>>>PolySix "Digiest" Page: http://www.acc.umu.se/~amber/Poly6
> >>>>>
> >>>>>
> >>>>>Yahoo! Groups Links
> >>>>>
> >>>>>To visit your group on the web, go to:
> >>>>>http://groups.yahoo.com/group/PolySix/
> >>>>>
> >>>>>To unsubscribe from this group, send an email to:
> >>>>>PolySix-unsubscribe@yahoogroups.com
> >>>>>
> >>>>>Your use of Yahoo! Groups is subject to:
> >>>>>http://docs.yahoo.com/info/terms/
> >>>>>
> >>>>>
> >>>>>
> >>>>>
> >>>
> >>>
> >>>
> >>>PolySix "Digiest" Page: http://www.acc.umu.se/~amber/Poly6
> >>>
> >>>Yahoo! Groups Links
> >>>
> >>>To visit your group on the web, go to:
> >>> http://groups.yahoo.com/group/PolySix/
> >>>
> >>>To unsubscribe from this group, send an email to:
> >>> PolySix-unsubscribe@yahoogroups.com
> >>>
> >>>Your use of Yahoo! Groups is subject to:
> >>> http://docs.yahoo.com/info/terms/
> >>>
> >>>
> >>>
> >>>
> >
> >
> >
> > PolySix "Digiest" Page: http://www.acc.umu.se/~amber/Poly6
> >
> > Yahoo! Groups Links
> >
> > To visit your group on the web, go to:
> > http://groups.yahoo.com/group/PolySix/
> >
> > To unsubscribe from this group, send an email to:
> > PolySix-unsubscribe@yahoogroups.com
> >
> > Your use of Yahoo! Groups is subject to:
> > http://docs.yahoo.com/info/terms/
> >
> >
> >
> >