Doepfer

Hi Roel,
First I went back and checked your original post - I note you bought 
your system about 2 years ago. As far as I'm aware, the new version 
of the A-131 came out in the autumn of 2001 - thus I suspect yours 
*is* one of the originals. This would use the PA382 chip , the 
exponential version of the PA381, which is linear and used in the A-
130. New A-131's use a different chip - a CA3080. Thus most of what I 
said before doesn't apply, and the fact that you have confirmed the 
outputs sound roughly the same at full gain ties in with what Florian 
said.

The VCAs don't really 'amplify' - they give you voltage controlled 
gain, which at its *maximum* is about 'times 1'. This should explain 
why your patch sounds louder directly to the 138, rather than through 
the 130/131: generally the output of a 130/(old)131 is less than the 
input!

As for the difference between lin and exp, this is not so easy to 
explain in words! For the linear VCA, the gain you get is directly 
proportional to the gain knob setting (or the CV in) - equal 
increments in the gain setting change the gain by the same 
increments. for the exp VCA, starting at gain = 0, large changes of 
knob setting give small changes in gain, then gradual you work up to 
where smaller changes of knob setting give bigger changes of gain. If 
you can think of an 'x squared' curve you'll get the idea (it's not 
quite right, but good enough to give the idea). The way the circuitry 
works out between the two, it appears the exp curve is virtually 
always 'below' the linear one, and comes up to meet it near the max 
gain setting. This is not the case for the newer A131, where the exp 
curve finally overtakes the linear one, hence why it is louder! (I'll 
check the PCB of my A-130 to see how the values changed for the old A-
131, and see if I can do some calcs to back all this up...)

After thought: if you have the manual check out pikkies on page 3 of 
the A130/131 entry!!!!!

The difference between the 138's is just the potentiometers, being 
either 'linear' or 'logarithmic' (which is just the inverse of 
exponential - again if you have the manual, in the figures, 'mirror' 
the curve line to be the other side of the straight one). Basically 
you get a different 'feel' between how much the knob turns for the 
effect on the signal. Because of the way the ear hears, things 
following the a log or exp relationship should sound more natural 
than a linear one.

As for more VCAs, some people find the A132 dual VCA to be very 
handy - check out some of the past discussions on this!

Tim

--- In Doepfer_a100@yahoogroups.com, "Roel Steverink" 
<r.steverink@w...> wrote:

> Hi Ethan, Tim, Florian and Bakis,
> 
> Thanks for all this great info!
> I have been experimenting with the options you gave me. 
> I noticed some strange things.
> First I patched the A-110 into the in of the A-131 (Exp.)
> and turning up the gain, nothing was heard until I 
> reached position 6. I repeated the same experiment with
> the A-130 (Lin.) I also began to turn up the gain and
> immediately at position 2 I began to hear the continious
> drone.
> So there's the big difference which must be due to fact
> that one is the exp. and the other the linear VCA.
> Now, I'm curious at which position will they be equal in
> volume. Can it be before they reach full gain?
> Nop, your right Florian both are equal in volume only 
> when they reached full gain (=10).
> So I happily conclude that there's nothing wrong with
> my VCA's!
> Before I forget the both in's were at position 8. And the 
> general out also on position 8.
> But when I turn both in's (from the two VCA's) to 10,
> I immediately hear that the linear gives more power!
> 
> I also have to say that I directly plugged the A-100 out to
> my mixer in this case. 
> 
> In the next example I patched 3 waveforms (pulse, triangle and
> sinus) into the A-138 mixer, the out of the mixer goes to 
> the audio in of the A-121 Multimode filter. Filter low out
> first into audio in of A-131 and the A-130 (settings of the
> VCA's are the same as above). There's barely any change.
> Same picture the linear gives a slight more power then the
> exponential. 
> 
> Now, when I plug the low out of the A-121 directly in the
> input of the A-138 mixer (lin) the volume is cranked up
> several levels. How come? Using this mixer means a more beefed
> up sound then using one of the VCA's???
> I believe this picture will be the same if one uses the
> A-138 mixer (exp)! Or am I wrong here?
> 
> Can anyone tell me what's the difference between 
> the A-138 exp and lin. mixer? 
> The same goes for both the exp and linear VCA's.
> I have to expand with more VCA's, cause 2 are to little.
> But which one should I take?
> Any ideas!
> 
> Thanks,
> 
> Roel.

Hi Ethan, Tim, Florian and Bakis,

Thanks for all this great info!
I have been experimenting with the options you gave me. 
I noticed some strange things.
First I patched the A-110 into the in of the A-131 (Exp.)
and turning up the gain, nothing was heard until I 
reached position 6. I repeated the same experiment with
the A-130 (Lin.) I also began to turn up the gain and
immediately at position 2 I began to hear the continious
drone.
So there's the big difference which must be due to fact
that one is the exp. and the other the linear VCA.
Now, I'm curious at which position will they be equal in
volume. Can it be before they reach full gain?
Nop, your right Florian both are equal in volume only 
when they reached full gain (=10).
So I happily conclude that there's nothing wrong with
my VCA's!
Before I forget the both in's were at position 8. And the 
general out also on position 8.
But when I turn both in's (from the two VCA's) to 10,
I immediately hear that the linear gives more power!

I also have to say that I directly plugged the A-100 out to
my mixer in this case. 

In the next example I patched 3 waveforms (pulse, triangle and
sinus) into the A-138 mixer, the out of the mixer goes to 
the audio in of the A-121 Multimode filter. Filter low out
first into audio in of A-131 and the A-130 (settings of the
VCA's are the same as above). There's barely any change.
Same picture the linear gives a slight more power then the
exponential. 

Now, when I plug the low out of the A-121 directly in the
input of the A-138 mixer (lin) the volume is cranked up
several levels. How come? Using this mixer means a more beefed
up sound then using one of the VCA's???
I believe this picture will be the same if one uses the
A-138 mixer (exp)! Or am I wrong here?

Can anyone tell me what's the difference between 
the A-138 exp and lin. mixer? 
The same goes for both the exp and linear VCA's.
I have to expand with more VCA's, cause 2 are to little.
But which one should I take?
Any ideas!

Thanks,

Roel.

Just my little pearls of knowledge????

As far as I understand it the practical application of lin/exp mixers 
this is that an exponential curve gives lot of control over the lower 
range as opposed to the linear curve which gives more control over 
the overall balance. For example, a mix balance of 1:2 1:3 is much 
easier to achieve using a linear mixer, while a subtle balance of 
1:1.001 would be done with an exponential one. 

Also the linear VCA is primarily useable in the middle of the signal 
path (for instance a VCA before a VCF) where linear control is 
easier. For our ears an exponential volume curve is more natural 
because of the way they work. Therefore an exponential VCA is usually 
used at the end of the signal path. 

Regards,
Patrick

--- In Doepfer_a100@yahoogroups.com, "Roel Steverink" 
<r.steverink@w...> wrote:

> Hi Ethan, Tim, Florian and Bakis,
> 
> Thanks for all this great info!
> I have been experimenting with the options you gave me. 
> I noticed some strange things.
> First I patched the A-110 into the in of the A-131 (Exp.)
> and turning up the gain, nothing was heard until I 
> reached position 6. I repeated the same experiment with
> the A-130 (Lin.) I also began to turn up the gain and
> immediately at position 2 I began to hear the continious
> drone.
> So there's the big difference which must be due to fact
> that one is the exp. and the other the linear VCA.
> Now, I'm curious at which position will they be equal in
> volume. Can it be before they reach full gain?
> Nop, your right Florian both are equal in volume only 
> when they reached full gain (=10).
> So I happily conclude that there's nothing wrong with
> my VCA's!
> Before I forget the both in's were at position 8. And the 
> general out also on position 8.
> But when I turn both in's (from the two VCA's) to 10,
> I immediately hear that the linear gives more power!
> 
> I also have to say that I directly plugged the A-100 out to
> my mixer in this case. 
> 
> In the next example I patched 3 waveforms (pulse, triangle and
> sinus) into the A-138 mixer, the out of the mixer goes to 
> the audio in of the A-121 Multimode filter. Filter low out
> first into audio in of A-131 and the A-130 (settings of the
> VCA's are the same as above). There's barely any change.
> Same picture the linear gives a slight more power then the
> exponential. 
> 
> Now, when I plug the low out of the A-121 directly in the
> input of the A-138 mixer (lin) the volume is cranked up
> several levels. How come? Using this mixer means a more beefed
> up sound then using one of the VCA's???
> I believe this picture will be the same if one uses the
> A-138 mixer (exp)! Or am I wrong here?
> 
> Can anyone tell me what's the difference between 
> the A-138 exp and lin. mixer? 
> The same goes for both the exp and linear VCA's.
> I have to expand with more VCA's, cause 2 are to little.
> But which one should I take?
> Any ideas!
> 
> Thanks,
> 
> Roel.

Hmm, I don't think I agree. The exponential curve causes the lower 
range to have a higher resolution, i.e. turning the knob has less 
effect, but allows for a lot more fine grained control (hence the 
1/1.001). As opposed to the higher range, where a subtle change in 
the knob causes a big difference and it is therefore a lot more 
difficult to control small changes in the high range. 

But I don't want to start a war on semantics here ;-) I think we both 
have the correct picture. It's just how we define control.

Greetz,
Patrick

--- In Doepfer_a100@yahoogroups.com, Florian Anwander 
<Florian.Anwander@c...> wrote:
> Hi Patrick
> 
> > As far as I understand it the practical application of lin/exp 
mixers
> > this is that an exponential curve gives lot of control over the 
lower
> > range 
> nearly correct. Exponential gives more control over the upper 
range. And it
> is not only "more control". In fact our ear senses "loudness" 
not "level".
> The loudness sensing is not linear but quite similar to an 
exponential
> characteristic. So a the position of a logarithmic fader will 
represent the

> loudness sensing quite correct. The same is valid for the VCA.
> 
> Florian

Hi Roel 

> Nop, your right Florian both are equal in volume only
> when they reached full gain (=10).
In my articles about the analogue synthesis I always explained, that the
VCA module should not be called amplifier but a level decreaser, because
the working range of the module is between "output is equal to input",
"output is less than input" and "no output at all". So there is never a
point, where the signal is amplified in the original meaning of the word
(="output is more than input").


> Before I forget the both in's were at position 8. And the
> general out also on position 8.
The basic level of the input does not matter. What matters should only the
ratio between input and output.

> But when I turn both in's (from the two VCA's) to 10,
> I immediately hear that the linear gives more power!
So this may be a different setting of the audiomixer, which is included in
the A130 or A131. Perhaps this is simply caused by the deviation of the
resistors in the mixer section. If it is a wished design, I would think it
more senseful, to have the A131 with more signallevel than the A130.



> Now, when I plug the low out of the A-121 directly in the
> input of the A-138 mixer (lin) the volume is cranked up
> several levels. How come? 
As far as I know there could be also two revisions of the mixers; the older
ones did not amplify and the newer ones had a smaller feedback resistor at
the summing amp, which causes some level increase, if the input is turned
up to 10.

I remember that some of the betatesters had the wish, that the mixer should
amplify. But this should be verified by Dieter.

Florian

Hi Patrick

> As far as I understand it the practical application of lin/exp mixers
> this is that an exponential curve gives lot of control over the lower
> range 
nearly correct. Exponential gives more control over the upper range. And it
is not only "more control". In fact our ear senses "loudness" not "level".
The loudness sensing is not linear but quite similar to an exponential
characteristic. So a the position of a logarithmic fader will represent the
loudness sensing quite correct. The same is valid for the VCA.

Florian

1/10/2003 1:01:27 AM, Florian Anwander 
<Florian.Anwander@consol.de> wrote:


>In my articles about the analogue synthesis I always 
explained, that the
>VCA module should not be called amplifier but a level 
decreaser, because
>the working range of the module is between "output is equal 
to input",
>"output is less than input" and "no output at all".

We could still call them VCAs, except in this sense VCA doesn't 
stand for voltage controlled amplifier, it's voltage controlled 
attenuator!

;0)

-Brandon