Once again The Trident!

[Bjorn Julin]

Hi Old Crow, Juergen and All!

In a second taught i was pleased to hear the "long"
beating range you had in your Trident,it sounded
really good, to good in fact to be true!

I want to rant about it once more!!!

I really believed that for a while but the more i was
thinking about it the more unreasonable it sounded,
so i did another test, this time i rearranged the
two trident CCO cores on my lab board and did a
precision trim of all possible offset voltages and i
also tried to match both CCO cores as much as possible.
The only cirquit i left out was the V to V expo converter
because we all know the behavior of such cirquit.
(but most, i didn't have room on my lab board!)

And my suspicions was verified i believe, the trident's
Linear VCO's (juergen language), muxed V to V converter
+ linear current regulator (BJ language)are almost
"beat free" in a sense, i achieved a beat periods of
several minutes long. Why! read on!

[Expo convereter:]

In a poly synth where you have one expo for each
VCO the most frequency alternating term is the temp
variation in the expo converter, so in a regular
poly synth the VCO's has a random behavior of
frequency drift = they drift in a specified direction
but random among each other.

But in a Trident the VCO's shares the expo converter
through a mux and this makes the major temp dependent
frequency drift shared, aka ,the major drift direction
is the same to all VCO's.

So what happens! Well, Old Crows "beat" test shows exactly
what happens, in a perfect "Korg" linear VCO cirquit the
beat dependent term are moved forward in the cirquit to:

1:Mux:
Not very likely.But the Ron changes with input voltage!

2:Current regulator:
Certainly some terms, offset voltage drifts in the OP
amp, the current transistor it self, and the main system
-Voltage regulator current reference, who in this case
are the system power supply.

3:The CCO it self and calibration:
Certainly, all components in the
core are more or less involved.

With this in our mind we go back to Old Crows
Beat test,what did it show? It showed exactly the same
what i found in my lab board test, that the Trident
Linear VCO´s has a peculiar frequency drift behavior!!

For example, when turned on it takes a second or two to
stabilize and when calibrated very close to each other
it can start to drift on one direction with a very slow
speed (minutes) and then suddenly change direction after
lets say 210 degrees, then starts to go back at 0
degrees but again suddenly change direction at
52,241351625891 degrees!!!!

And so it can keep going on for several minutes, and probably
hours and newer go the entire cycle (360 degrees)!!
It never beats, it just simply phases around!!

This time i was the one who was tricked for a while,
Old crows test does not show if the KOrg CCO are better
or worse in the soft sync area, as Juergen says it would.

It only shows the beat behavior, infact i was wrong to
i have changed my mind in respect to what i said earlier,
that the Korg CCO are "soft sync friendly", it turns out
that at high freq 10kHz and above the Korg CCO soft syncs
as bad as any regular CCO, and even worser in some cases!

Infact i had enormous problems to get the CCOs on the lab board
sync free, offcourse the lab board is not the best place to do
soft sync tests, but keeping an eye on details and doing some
speculations, Korg knowed this issues, and it might be one off
of the reasons to keep the frequency range short and (7 octaves)
and having "two separate +5V's" (+5VA and +5VB) to the cores.
And as Old Crow says "carefully laid out PCB" this might improve
the soft sync hassle and stress in the Trident!!

But if we take all those "found" issues and uses a different CCO
core,lets say we take a Moog /Source/Rouge/etc core and drive
it with the Trident muxed expo converter?

Im most interested and curiously intrigued to se the results!!!

Reg
BJ

_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com.

Share information about yourself, create your own public profile at 
http://profiles.msn.com.