Doepfer

Hi,

I was just playing around with the a-135.

I basically want to be able to make seperate outputs for each channel
after the cv attenuator circuitry. I am having trouble figuring out
exactly how the circuit works - I was thinking that the final stage
before each channel's signal get summed and then led through the
makeup gain ic and then to the master out - is on the pa 381 ic. So my
plan was to just tap the seperate outputs from the two pa381 ic's and
then lead each channel into it's own seperate makeup gain opamp,,,, 

So I tried this but wasn't getting any signal.... I'm sure if I could
see the schematics I could make better sense of everything,,,, anyhow
I am hoping somoene can offer some insight on how to actually do this.

It would be great to have the seperate outs,,, would be like have an
extra 4 vca's available!

Thank You!

To do this looks relatively straightforward. The PA381 is a current 
output device: the currents output from pins 1 & 7 of both chips all 
come together at one end of J1, and are summed by one op amp in the 
TL082 - pin 6 is the summing junction, pin 7 the o/p, pin 5 grounded. 
So you would have to replicate this op amp set-up - a 47k feedback 
resistor, 22p stability cap (C13) and 1k o/p resistor - for each 
section. The second op amp in the TL082 is unused, but it would 
probably be easier to use a TL084 on a separate piece of 
stripboard/whatever.

It's getting late in the UK, but I can maybe supply more help 
tomorrow if needed!

Tim

--- In Doepfer_a100@yahoogroups.com, "danv_83" <danv_83@y...> wrote:
>
> Hi,
> 
> I was just playing around with the a-135.
> 
> I basically want to be able to make seperate outputs for each 
channel
> after the cv attenuator circuitry. I am having trouble figuring out
> exactly how the circuit works - I was thinking that the final stage
> before each channel's signal get summed and then led through the
> makeup gain ic and then to the master out - is on the pa 381 ic. So 
my
> plan was to just tap the seperate outputs from the two pa381 ic's 
and
> then lead each channel into it's own seperate makeup gain opamp,,,, 
> 
> So I tried this but wasn't getting any signal.... I'm sure if I 
could
> see the schematics I could make better sense of everything,,,, 
anyhow
> I am hoping somoene can offer some insight on how to actually do 
this.

> 
> It would be great to have the seperate outs,,, would be like have an
> extra 4 vca's available!
> 
> Thank You!
>

ok, i was trying to use the lf444 quad op amp,,,, and i couldn't get a
single thing to happen other than some crummy low end noise.

perhaps i'm not biasing something correctly,,, i am new to electronics
theory and am only just now getting into more hands on things.

i didn't implement any caps inline with the lf444, i was simply
following the basic circuit listed on the pa381 data sheet.....

yes, more help is greatly appreciated, especially if you could show me
how to do this using the lf444 chip instead.

thank you very very much!


--- In Doepfer_a100@yahoogroups.com, "Tim Stinchcombe" <timothy@t...>
wrote:

>
> To do this looks relatively straightforward. The PA381 is a current 
> output device: the currents output from pins 1 & 7 of both chips all 
> come together at one end of J1, and are summed by one op amp in the 
> TL082 - pin 6 is the summing junction, pin 7 the o/p, pin 5 grounded. 
> So you would have to replicate this op amp set-up - a 47k feedback 
> resistor, 22p stability cap (C13) and 1k o/p resistor - for each 
> section. The second op amp in the TL082 is unused, but it would 
> probably be easier to use a TL084 on a separate piece of 
> stripboard/whatever.
> 
> It's getting late in the UK, but I can maybe supply more help 
> tomorrow if needed!
> 
> Tim
> 
> --- In Doepfer_a100@yahoogroups.com, "danv_83" <danv_83@y...> wrote:
> >
> > Hi,
> > 
> > I was just playing around with the a-135.
> > 
> > I basically want to be able to make seperate outputs for each 
> channel
> > after the cv attenuator circuitry. I am having trouble figuring out
> > exactly how the circuit works - I was thinking that the final stage
> > before each channel's signal get summed and then led through the
> > makeup gain ic and then to the master out - is on the pa 381 ic. So 
> my
> > plan was to just tap the seperate outputs from the two pa381 ic's 
> and
> > then lead each channel into it's own seperate makeup gain opamp,,,, 
> > 
> > So I tried this but wasn't getting any signal.... I'm sure if I 
> could
> > see the schematics I could make better sense of everything,,,, 
> anyhow
> > I am hoping somoene can offer some insight on how to actually do 
> this.
> > 
> > It would be great to have the seperate outs,,, would be like have an
> > extra 4 vca's available!
> > 
> > Thank You!
> >
>

The CEM3381/PA381 VCAs have current outputs (not voltage outputs as usual).
The currents of the four CEM3381/PA381 are summed up in a current-to-voltage
converter to obtain a common voltage output. This is why you do not obtain
an signal at the CEM3381/PA381 (current) outputs. One would require a
separate current-to-voltage converter for each section of the CEM3381/PA381
VCAs followed by a (voltage) summing unit. But this would require a lot of
modifications of the A-135 module.

Best wishes
Dieter Doepfer

> I basically want to be able to make seperate outputs for each channel
> after the cv attenuator circuitry. I am having trouble figuring out
> exactly how the circuit works - I was thinking that the final stage
> before each channel's signal get summed and then led through the
> makeup gain ic and then to the master out - is on the pa 381 ic. So my
> plan was to just tap the seperate outputs from the two pa381 ic's and
> then lead each channel into it's own seperate makeup gain opamp,,,,
>
> So I tried this but wasn't getting any signal.... I'm sure if I could
> see the schematics I could make better sense of everything,,,, anyhow
> I am hoping somoene can offer some insight on how to actually do this.
>
> It would be great to have the seperate outs,,, would be like have an
> extra 4 vca's available!
>
> Thank You!
>
>
>
>
>
>
>
> Yahoo! Groups Links
>
>
>
>
>
>

Dieter has already mentioned the key fact which I omitted last night
(I wasn't sure just what your experience level was then), in that the 
TL082 converts the currents from the PA381s to the voltage which is 
then output. It is this current conversion that would require
separate op amps for each channel, and getting at the individual 
currents is the hard part. I can think of 3 ways to do this, easiest 
first:

1) Cut the tracks from both sets of pins 1 and 7 at some convenient 
point, and then add wires from the pins to the new op amps

2) (Really depends on your soldering skills): remove the PA381s
from their sockets, desolder the sockets, then bend pins 1 and 7 on
the sockets out so that they don't go through the PCB, and to which
new wires can be attached, then re-solder sockets into PCB
(advantages are that no tracks need to be cut, but it will be quite 
hard to make a tidy job of it!)

3) Remove the PA381s and bend out pins 1 and 7 on _them_ and
re-insert, again so that wires can be attached to the IC pins (given 
the cost and difficulty of getting replacement PA381s I personally 
would certainly *not* do this, but it is an option!)

If you a this point you are still interested, come back to me (perhaps 
off-list?)

> ok, i was trying to use the lf444 quad op amp,,,, and i couldn't get a
> single thing to happen other than some crummy low end noise.
> 
> perhaps i'm not biasing something correctly,,,

The op amp summing the currents holds the summing point at ground 
potential, so the voltage at the output from all the PA381 amps is 0V, 
which is why you got nothing. (This is in fact a necessary condition 
for running the PA381 - it uses the voltage at pin 7 as a reference 
voltage.)

> i didn't implement any caps inline with the lf444, i was simply
> following the basic circuit listed on the pa381 data sheet.....

Which is basically all the TL082 is/does, apart from the feedback 
resistor value (and the cap).

> yes, more help is greatly appreciated, especially if you could show me
> how to do this using the lf444 chip instead.

I can't see any reason why this wouldn't work as well as, say, a
TL084 - but since it is not a simple matter of picking off some
voltages, it all comes down to whether you want to go ahead at all!

Tim