CEM3310 in Oberheim OB-8

[WeAreAs1 at aol.com]

Hello list,

I was looking at the service manual for the Oberheim OB-8 and I noticed
something that I had not noticed until now.  The OB-8 uses Curtis CEM3310's
for its envelope generators.  In early versions of the OB-8 (software version
A-3 or less), the output of each filter envelope (Ux11) is routed through a
CEM3360 VCA for filter envelope depth control.  This 3360 is controlled by the
signal named "VCFM".  In later versions (software A-4 or later), this 3360 is
bypassed and the unattenuated envelope output is directly connected to the
filter CV summing bus.  Additionally, the filter envelope amount control
signal "VCFM" is re-routed to PIN 3 of the each CEM3310.  These are the only
changes to the hardware.  The service manual states "The 3360 VCA is bypassed
and the 3310 in conjuction with the software modulates the amplitude of the
envelope.  A-4 or later software must be installed.  This improves temperature
stability and tracking of the filter envelope modulation."  These changes are
described in ECO405, dated 12/12/82.

After looking again at the CEM3310 datasheet, I find this to be a very unusual
way of connecting it.  In fact, I have never seen PIN 3 even connected on any
other application of the 3310.  The data sheet refers to PIN 3 as a "Peak
Threshold Voltage Pin".  The internal circuit diagram shows this pin as a
connection to the minus input of comparator.  The plus input of the comparator
is internally connected to PIN 2 (the Envelope output).  

Although the data sheet makes no specific mention of it, this OB-8 circuitry
indicates that it is possible to set a peak envelope amplitude by introducing
a voltage into PIN 3.  The data sheet does mention that "a Sustain control
voltage greater than the Peak Threshold Voltage will result in a 'jump' to the
sustain level".  They show that a positive feedback loop (with an opamp
buffer) from the Threshold pin to the Sustain CV pin will "prevent the Sustain
level from rising above the envelope peak", eliminating the aforemention
undesirable 'jump'.

Interestingly, the OB-8 does not incorporate this feedback loop at all.  I am
guessing that the A-4 software version somehow compensates by scaling the
Sustain CV so that it never goes above the "VCFM" CV that is now being input
at PIN 3.

Not having access to an OB-8, I was hoping that some list member who has an
OB-8 with software version A-4 or higher (maybe Juergen H.?) might take a few
moments to measure the voltages found at PIN 3 (the Peak Threshold) and PIN 9
(the Sustain CV) of the CEM3310 at various settings of the Filter Sustain
Level and Filter Envelope Modulation Amount.  Is anyone game?

This would be useful information for anyone who wants to use the 3310 in new
designs.  (yes, some of us have a few of the chips..)

BTW, the CEM3310 data sheet that can be downloaded from Paul Schreiber's site
is not complete, and does not contain any information about using PIN 3.  The
complete 4-page data sheet can be found in the Prophet V and Pro-One, and
Memorymoog service manuals.

Best regards,
Michael Bacich