Doepfer

Hi everyone. I'm a relative newcomer to the A-100 scene - bought mine
last September and its getting bigger with alarming rapidity! I was
going to post this on the Doepfer website forum, but (at last) found
the exact address of this one by accident, and decided it would be
better here. It relates to the compatibility of the A111/115/117, as
discussed in some old messages in the other forum ('A117 with A111',
'Most hated...'), so apologies in advance if I'm going over old ground
(I've not had chance to spend that much time looking back at old stuff
here...)

In both the 117 and the 115 the input signal is merely switched
through a transistor in order to produce a 'squared up' pulse which is
then fed into the appropriate chip (4006 shift register in 117, 4024
binary counter in 115). The 'rise time' of the leading edge of each
pulse depends on the 'quality' of the input signal, and is quite
critical to the correct operation of the module. The recommended rise
time for the 4000 series family chips is 4 microsec at 10V supply down
to 1 at 15V -  we have 12 volt supplies so our target is somewhere in
between. On my 117 if I feed a low frequency sine wave in it is quite
possible not to get any output at all (i.e. similar to problem
reported by 'jim' I guess, with A111, 11/28/01) - investigation shows
that the shift register chip just doesn't run at all (it's designed to
run at frequencies up to about 30MHz, so we're just toying with it
really!). Slowly cranking the frequency of the sine wave up has the
effect of decreasing the rise time, and when it gets to about 3 or 4
microsec, lo and behold, it will burst back into life! (To be fair it
must be said it is obviously much happier with a square wave as
input!) The effect on the 115 seems similar - the binary counter seems
to 'miss a beat' now and again, which you can hear as clearly 'a not
properly divided' signal. The square wave coming off the CEM3340 chip
in my A111 is pretty awful -  square is hardly the correct word, it is
well rounded, and becomes more so with increasing frequency, so it
doesn't really drive the 117 particularly well either.

Two solutions: there are 2 modules that can help get around this - the
idea with both is the same: stick the signal through some nice CMOS
logic that'll give a lovely 'sharp' pulse out. The A150 is probably
the simplest - square wave into CV input;+5V, -5V into the two inputs
(I use a 129/3 as a simple CV source, plus an inverter - a 176 or 174
would do just as well);then patch output into the 117 (or 115). This
certainly gives the 'playable noise' effect with the 117 using the 111
on my system. Alternatively the divided output from the A163 probably
comes off a chip (I've not checked in detail), but that is nice and
square - just set it to divide by 1.

Overall I'm having great fun with my A100 - learning lots about sound,
music (I'm not a musician) and electronics - I love it! I'm expecting
to get a copy of Chowning's paper on Frequency Modulation soon, and
then I intend to really start exploring the linear FM input on the
A111 (and it will be interesting to see what limitations analogue
places on it, rather than doing it digitally!).

I also have a question of sorts - do people find their questions
eventually get answered through this medium? I only ask as looking
back at some old posts here I noticed a query about the A-150/A-151
inability to switch clock pulses. The answer I think is in the manual
- they will only cope with signals less than 8 volts (shown in a
diagram somewhere - this is from memory!), and generally clock pulses
go from ground to supply, i.e. 12 volts, and this will upset the
circuitry somewhat, and it allows some of the signal through.
Depending on where you're trying to send them, simply attenuating the
signal (e.g. A-138 linear mixer) might do the trick.

hello and welcome to the list!
thank you for your valuable info,.you seem to know a
lot about the a100!as for the gate signals in a100,are
you sure that the gate signals are from 0volts to
12volts??i thought that they were from 0volts to
+5volts...anyway if you are right,then your theory
explains the problem with the a150/151 and clock
signals...
synthfreak2000
[doepfer_a100]group moderator

--- stinchcombe_t <tstinchcombe@qinetiq.com> wrote:
> Hi everyone. I'm a relative newcomer to the A-100
> scene - bought mine
> last September and its getting bigger with alarming
> rapidity! I was
> going to post this on the Doepfer website forum, but
> (at last) found
> the exact address of this one by accident, and
> decided it would be
> better here. It relates to the compatibility of the
> A111/115/117, as
> discussed in some old messages in the other forum
> ('A117 with A111',
> 'Most hated...'), so apologies in advance if I'm
> going over old ground
> (I've not had chance to spend that much time looking
> back at old stuff
> here...)
> 
> In both the 117 and the 115 the input signal is
> merely switched
> through a transistor in order to produce a 'squared
> up' pulse which is
> then fed into the appropriate chip (4006 shift
> register in 117, 4024
> binary counter in 115). The 'rise time' of the
> leading edge of each
> pulse depends on the 'quality' of the input signal,
> and is quite
> critical to the correct operation of the module. The
> recommended rise
> time for the 4000 series family chips is 4 microsec
> at 10V supply down
> to 1 at 15V -  we have 12 volt supplies so our
> target is somewhere in
> between. On my 117 if I feed a low frequency sine
> wave in it is quite
> possible not to get any output at all (i.e. similar
> to problem
> reported by 'jim' I guess, with A111, 11/28/01) -
> investigation shows
> that the shift register chip just doesn't run at all
> (it's designed to
> run at frequencies up to about 30MHz, so we're just
> toying with it
> really!). Slowly cranking the frequency of the sine
> wave up has the
> effect of decreasing the rise time, and when it gets
> to about 3 or 4
> microsec, lo and behold, it will burst back into
> life! (To be fair it
> must be said it is obviously much happier with a
> square wave as
> input!) The effect on the 115 seems similar - the
> binary counter seems
> to 'miss a beat' now and again, which you can hear
> as clearly 'a not
> properly divided' signal. The square wave coming off
> the CEM3340 chip
> in my A111 is pretty awful -  square is hardly the
> correct word, it is
> well rounded, and becomes more so with increasing
> frequency, so it
> doesn't really drive the 117 particularly well
> either.
> 
> Two solutions: there are 2 modules that can help get
> around this - the
> idea with both is the same: stick the signal through
> some nice CMOS
> logic that'll give a lovely 'sharp' pulse out. The
> A150 is probably
> the simplest - square wave into CV input;+5V, -5V
> into the two inputs
> (I use a 129/3 as a simple CV source, plus an
> inverter - a 176 or 174
> would do just as well);then patch output into the
> 117 (or 115). This
> certainly gives the 'playable noise' effect with the
> 117 using the 111
> on my system. Alternatively the divided output from
> the A163 probably
> comes off a chip (I've not checked in detail), but
> that is nice and
> square - just set it to divide by 1.
> 
> Overall I'm having great fun with my A100 - learning
> lots about sound,
> music (I'm not a musician) and electronics - I love
> it! I'm expecting
> to get a copy of Chowning's paper on Frequency
> Modulation soon, and
> then I intend to really start exploring the linear
> FM input on the
> A111 (and it will be interesting to see what
> limitations analogue
> places on it, rather than doing it digitally!).
> 
> I also have a question of sorts - do people find
> their questions
> eventually get answered through this medium? I only
> ask as looking
> back at some old posts here I noticed a query about
> the A-150/A-151
> inability to switch clock pulses. The answer I think
> is in the manual
> - they will only cope with signals less than 8 volts
> (shown in a
> diagram somewhere - this is from memory!), and
> generally clock pulses
> go from ground to supply, i.e. 12 volts, and this
> will upset the
> circuitry somewhat, and it allows some of the signal
> through.
> Depending on where you're trying to send them,
> simply attenuating the
> signal (e.g. A-138 linear mixer) might do the trick.
> 
> 


=====
synthfreak(parallel worlds)
athens-greece
[Doepfer_a100] group owner-moderator

__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com

> as for the gate signals in a100,are
> you sure that the gate signals are from 0volts to
> 12volts??i thought that they were from 0volts to
> +5volts...anyway if you are right,then your theory
> explains the problem with the a150/151 and clock
> signals...

Yes, I'm sure, I checked them last night. Of the modules I have the
clock pulses from the 160/161 are among the largest output by any
module, and are 0 - 12V. The digital noise from the 117 is also 0 -
12V; the gate signals from the 119 and 191 are also quite big, at 0 -
9V. I don't have a 151, but my 150, even when off, will basically
output a 0 - 8V pulse with a 0 - 12V pulse as input - the size of this
pulse decreases as the size of the input pulse decreases, until the
switch behaves properly when the input is down to 8V. I don't fully
understand the mechanism by which this happens yet, but I suspect its
the 4053 chip rather than the circuitry. Its behaviour is dependent on
the direction through the switch and also whether the unused I/Ox
socket is grounded or not. I suspect the curcuitry in the 151 is
similar, and thus probably suffers similar problems.

> thank you for your valuable info,.you seem to know a
> lot about the a100!

I don't know about this, but I'm having loads of fun working out what
makes everything tick!

[The views expressed above are entirely those of the writer and do not
represent the views, policy or understanding of any other person or
official body.]