Using 1K Tempcos with Electronotes designs

[Debby and Gene Stopp]

Hi DIY,

Gosh guys seems like I never contribute anymore. :(  The truth is that my
DIY stuff has come to a standstill, between babies and work I only get the
chance to lurk every other week or so. My last time-killer was a week-long
business trip to communist China (so life continues to be interesting but
just not synth-related).

Anyway I just wanted to mention that I laid out the ASM-1 so that the tempco
is located in a resistor divider after the CV summing amp (which has the
v/oct trimmer in it's feedback loop). I have seen some tempco's that do not
behave inside the feedback loop of the CV summer (an unexplained phenomena,
perhaps related to some kind of inductance causing the op-amp to oscillate).
I think it was Chris List who got them, and emailed me to say that they
didn't work, and then he sent me some and I tried them and sure enough,
totally non-linear behavior. After I found that they worked fine in a normal
divider after the op-amp's output, I stopped pursuing the issue to there it
rests. Other tempco users should perhaps be on the lookout for this -
probably many tempco's are wirewound so if it was an inductance-related
thing then it may resurface in other designs. Come to think of it, maybe it
was just a TL-082 related thing? Ah well, I'll spend no further time on the
matter.

Oh yeah my point is that when you use the tempco in the divider after the
op-amp, you can use either 1K or 2K types and just stuff the other
resistor(s) appropriately to maintain the 18.02mv at the base of the expo
converter tranny. I've done both with good results.

BTW I must mention that amongst the analog synths on the "MIDI/CV wall" in
my music setup, out of all the VCO's (including Moog 901's, ARP 2600,
Minimoog (latest VCO card), OB-8, Emu SSM chip demo board synth, Moog Taurus
II, an old ENS-74, a Sonic Six, and one ASM-1), the VCO's in the ASM-1 are
the ones that need the least frequent re-tuning. They are rock-steady over
extrememly long time intervals and power cycles. OK, I admit the Moog 901's
and the Sonic Six aren't pinnacles of accuracy, but the OB-8 (CEM3340) and
the Minimoog with the latest VCO card are pretty darn stable. But the ASM-1
VCO (from Electronotes, courtesy of Terry) seems to be perfect, at least
from a user's point of view.

- Gene


-----Original Message-----
From: Paul Schreiber <synth1 at airmail.net>
To: 'synth diy general post' <synth-diy at mailhost.bpa.nl>
Date: Sunday, October 11, 1998 8:08 PM
Subject: Using 1K Tempcos with Electronotes designs


>Ian Fritz pointed out that I may have not been clear on the use
>of these 1K tempcos.
>
>The "correct" base-emitter voltage the log amp wants to see
>is approx 18.02mv/octave. (26 x ln2) There are 2 ways to do this easily.
>
>1) Use 100K input resistors and a 2K feedback resistor. This gives
>1/50 = 20mv/octave. Then a trimmer divider on the output gives
>the correct value.
>
>This is what Electronotes uses.
>
>2) Use 100K inputs and a 100K feedback. Then, use a 49.9K  1% resistor, in
series with
>a 10K trimpot and the 1K tempco (to ground)  as a divider. This gives an
adjustible range.
>
>This is the ASM-1 approach (sort of. I think a fixed 56K is used in some
versions).
>
>Note ARP used a 1.87K in the divider, which is even closer to "correct"
(for filters,
>no trimming is probably needed).
>
>So, if you are building an Electronotes VCO that has a 2K feedback resistor
in the
>input summer, you will have to use 49.9K input resistors (a standard 1%
value) instead
>of 100Ks. The resistors that summ in the "Coarse" and "Fine" tuning pots
will need
>to be halved as well.
>
>Paul Schreiber
>Synthesis Technology
>
>