Doepfer

Hi Andy

Am 21.02.2019 um 12:09 schrieb man.of.mystery@ntlworld.com [Doepfer_a100]:
>   It tends to go from a little bit too quiet to way too loud, rather 
> suddenly, 
That is how exponential VCAs do work. There is nothing to adjust, that 
would change that behaviour.

Florian

Hello Andy,

over the years 3 different version of the A-131 were manufactured. For
details please refer to our website:

www.doepfer.com > English > PRODUCTS > A-100 > Module Overview > A-131

Only for version 2 (with CA3080) it was possible to trim the exponential
scale a bit by means of P7 (2k5 trimming potentiometer at top left corner of
the pc board).

As already mentioned by Florian for exponential VCAs the amplification scale
differs clearly from linear VCAs.

Best wishes
Dieter Doepfer

> -----Ursprüngliche Nachricht-----
> Von: Doepfer_a100@yahoogroups.com [mailto:Doepfer_a100@yahoogroups.com]
> Gesendet: Donnerstag, 21. Februar 2019 12:09
> An: Doepfer_a100@yahoogroups.com
> Betreff: [Doepfer_a100] Doepfer A-131 exponential VCA - calibration
> procedure
>
>
> Is there a calibration procedure document, for the Doepfer A-131
> exponential VCA?
>
> I bought one used, a while ago, and I've never been sure it's set
> up as it should be.  It tends to go from a little bit too quiet
> to way too loud, rather suddenly, making me fear for my eardrums,
> on headphones - which makes me wonder if the previous owner
> fiddled with it, then gave up on it.
>
> - Andy

Hello

Am 21.02.2019 um 13:07 schrieb man.of.mystery@ntlworld.com [Doepfer_a100]:
>  I get a very gradual increase, then a very sudden one.

Do you get this too, if you apply an external linear voltage? Or does it 
happen only with the "gain" potentiometer?

Florian

-- 
http://www.florian-anwander.de

Version 3 of the A-131 is based on the THAT2150 and the exponential scale is
predefined by the THAT2150. The module is a plain standard application of
the THAT2150 and there is nothing special.
The trimming potentiometer "GAIN OFFSET/P6" is used to define the max.
amplification (of amplification offset). In the factory it's adjusted so
that with Gain control fully CW (and without external CV) the
"amplification" is about 1.
Another trimming potentiometer "OUTPUT OFFSET/P7" is used to minimize the CV
feedthrough of the VCA.
The control scale of the module is defined by the control scale of the
THAT2150. I can send you the THAT2150 data sheet and schematics if you tell
me your email address.

I suppose that your "problem" has to do with the exponential scale. As a
rule it's always a problem with exponential VCAs because an exponential VCA
does theoretically never reach zero amplification (but in reality the signal
disappears eventually in the noise floor). On has to find a compromise
between the exponential scale and the minimum amplification (i.e. the
amplification with manual GAIN control fully CCW):

If the exponential scale is too "flat" the amplification with GAIN fully CCW
is not small enough (i.e. there is still some feedthrough @Gain=0).

If the amplification with manual GAIN control fully CCW is very small
(e.g. -80dB or 1/10000 with Gain=0) so that no audio feedthrough is
noticeable it takes a while until the audio level becomes audible because at
center position (5) of the gain knob it's still about -40dB (or 1/100), and
even at 75% of the rotating angle (about pos. 7-8) it's still -20dB (or
1/10). In the range 8...10 then the level increases suddenly from 1/10
(-20dB) to 1 (0dB).

Hope this helps to understand the exponential VCA
Dieter Doepfer

> -----Ursprüngliche Nachricht-----
> Von: Doepfer_a100@yahoogroups.com [mailto:Doepfer_a100@yahoogroups.com]
> Gesendet: Donnerstag, 21. Februar 2019 13:08
> An: Doepfer_a100@yahoogroups.com
> Betreff: [Doepfer_a100] Re: Doepfer A-131 exponential VCA - calibration
> procedure
>
>
> I have a degree in maths, and I studied electronic engineering
> for a while.  I also own more synthesizers than I probably should
> - so I understand the concepts involved.
> What I mean is, the response curve of this one, for gain vs CV,
> is far from desirable.  I get a very gradual increase, then a
> very sudden one.  I haven't attempted to graph it.
> I have seen that website page before, and read the manual, but I
> haven't seen anything official about adjusting or calibrating
> one.  I have read that there are preset pots to adjust, but only
> vague details.
> It seems the one I have is version 3, and there are only two
> presets / trim pots, labelled Gain Offset, and Output Offset.  Is
> there anything to be gained by adjusting those?
> I own multimeters, basic oscilloscopes, soldering irons, and a
> large heap of resistors, capacitors and various other components.
>  I am quite prepared to modify it, if necessary.  Is there a
> schematic available?
>
> - Andy
>

Hello Andy,

the gain knob and the CV inputs behave in the same way. The voltages coming
from the Gain control and the CV inputs are just added and the sum is used
to control the exponential VCA based on the THAT2150.
In other words: if you move the gain knob uniformly from fully CCW to fully
CW it's the same as if you apply a linearly increasing CV from 0...+5V (with
appropriate setting of the CV attenuator).

I'll send the documents to your email address.

Best wishes
Dieter Doepfer

> -----Ursprüngliche Nachricht-----
> Von: Doepfer_a100@yahoogroups.com [mailto:Doepfer_a100@yahoogroups.com]
> Gesendet: Donnerstag, 21. Februar 2019 14:26
> An: Doepfer_a100@yahoogroups.com
> Betreff: Re: AW: [Doepfer_a100] Re: Doepfer A-131 exponential VCA -
> calibration procedure
>
>
> Thanks, Dieter.  I'll give that some thought.
>
> I've sent you an email, to hardware@doepfer.de
>
> To Florian:  To be clear; it's the response of gain vs CV inputs
> that I'm talking about, not the gain knob.  If I set it for a
> non-deafening level, at maximum CV input, there is very little
> response at lower CV levels.
>
> - Andy

Am 21.02.2019 um 14:25 schrieb man.of.mystery@ntlworld.com [Doepfer_a100]:
> To Florian:  To be clear; it's the response of gain vs CV inputs that 
> I'm talking about, not the gain knob.  If I set it for a non-deafening 
> level, at maximum CV input, there is very little response at lower CV 
> levels.

I still think you underestimate difference in the received loudness 
changes between linear and exponential control.

lets take this graph:
http://www.florian-anwander.de/exponential_control.jpg

Lets assume you use no external CV and control the VCA only by the 
"Gain" knob.
The X-axis of the graph represents the amount of the gain knob.
The Y-axis of the graph represents the amplification of the VCA.

As you can see at a knob position of "5" will get only 10% of the 
possible amplification. Thet other way around: to achieve 50% of 
amplification you have to turn up the gain knob to position "9" out of "10".


It does not matter whether you are using the gain knob or an external 
control voltage, as the gain knob simply provides internally a control 
voltage.

Florian

-- 
http://www.florian-anwander.de

Hi Andy

I am not a native english speaker as you may recognize, but I really 
have difficulties to understand your descriptions.

You wrote: " it's the response of gain vs CV inputs that I'm talking 
about". And that is what the graph displays: the response of 
gain(=y-axis) versus the controlling voltage (=x-axis).

I will try to find time to make a comparing video (displaying the 
outputlevel of the same signal source through a  linear VCA Left and a 
exponential VCA right on a stereo peak meter).


Florian

Hello


Am 21.02.2019 um 17:06 schrieb man.of.mystery@ntlworld.com [Doepfer_a100]:
> Okay, Florian.  What are the units of your graph?
>
> Hearing is logarithmic, so try redrawing that graph with the y axis in dB.
>
> What do you see?

There is no unit necessary. The unit might be "count of goldfish in a pool".

The important thing is the difference to a linear control. 
Mathematicians understand exponential control like in
https://upload.wikimedia.org/wikipedia/commons/thumb/3/38/Exp_e.svg/2000px-Exp_e.svg.png
where the common point of the linear function and the exponential 
function is at X=0.
(as in 
https://upload.wikimedia.org/wikipedia/commons/thumb/3/38/Exp_e.svg/2000px-Exp_e.svg.png)

But in the synth world the common point of linear and exponential 
control is the Xmax. (and in aproximation Xmin).
http://www.florian-anwander.de/exponential_control.jpg (I added a linear 
control line).


Florian

-- 
http://www.florian-anwander.de