cfmd at swipnet.se
Mon Mar 6 02:01:11 CET 2000
From: "urosuros" <urosuros at bits.net>
Subject: Re: vcf5t update
Date: Mon, 6 Mar 2000 00:46:46 +0100
> Hi Tom and all others
> I'm looking at your schematic and I must say that I'm puzzled by how VC is
> implemented first thing both ends of initial trimming pot are conected to Vcc
> ( btw what is suply voltage ? )
I'd toast this at +/- 15V. It should not meet any major trouble.
> Are you shure this is conected right ?
It seems correctly hooked between +Vcc and -Vcc, yes.
> second thing both emitor and colector of transistor at which base is CV
> conected are conected to voltage suply across 10k resistors . Well this is
> definitly not typical tranistor amplifier stage . It seems to me that
> transistor wwould be in saturation region of operation . Am I right whit
> this ?
First, the up arrow is for +Vcc and the down arrow is for -Vcc, so then you
have a pretty classic transistor setup right there. It will basically operate
as a phase splitter.
Then, the initial setting will make the operating point just about right, so
for a rising input, the trany will conduct less and less. As the tranny reduces
its conduction the voltage of its emitter and collector will start to go each
way towards the power rail away from the ground between them. Due to the Beta
of the 2N3906 will the base current not make such a huge difference, so it will
look pretty symmetric.
Then, this voltage is being delivered to the diode setting thru the 100k
resistors, effectively working as current sources.
The audio input is stripped of any DC, with a gain of rougthly 1/220. The 100
ohm source will be powerfull relating to anything else near these diodes.
The input damping will make the input signal to use the small signal property
of the diodes which basically means that one is using the transconductance
property of the diodes. This property depend exponential to the applied
large scale voltage/current. The input voltage will be sourced into the
100nF over 10nF capacitor point where as the feedback term will experience
> I rarely encoutered saturation in linear electronic ( and my impulse/digital
> electronic course is yet to come ) so Im not shure how to model it . But any
> way since E-Vcc and C-Vcc resistors are conected across same voltage ( and
> Vce is very small , if transistor works in sat. ) one resistor will sink
> curent and other will source curent , right ? Or wrong ? I would appreciate
> if somebody explain how this works .
I think you started of with the power supply signs getting incorrectly
interprented. See my above comment. I think you understand the curcuit much
better when seening that there are infact two diffrent power supply voltages.
> Also , in old MS50 filter schematic from EFM we have 3046 array in diode
> conection. Are you shure Tom that 1N914 will work here ?
Notice that the 3046 transistors are wired up to act as diodes, only the
base-emitter NP-junction is being operational. The sourcing diodes has been
replaced by the 100k resistors.
> And two question not regarding this filter .
> First if transistors in ladder filter are not matched that would decrese CV
> rejection but what else will be affected ?
Yes, and it would also decrease the common mode rejection of the input stage,
to what ever degree that is usefull.
> It seems to me that poles would spread a little ( or a lot ) and that would
> decrese resonance and overall filter performanca , right ?
Yes, the unmodulated poles would not lie on the same place, so the initial
part they would be dragged together and only when the feedback would get high
enougth they would spread out as we are used to. I don't think the really
interesting thing is happening in the linear frequency plane (that is pole
positions) but rather in what happends in the non-linear range with operation
points of the individual transistors.
> Could this be neglected by simply increasing feedback around ladder ?
Nope, the feedback will raise the Q-value of two of the poles and move them
closer and closer to the jw-axis where as the other two are moving away from
the jw-axis. The poles are sitting on the 45 degree positions of an expanding
ring, the ring expands with the increased feedback.
> And second : dif amp with active load has best performance in CV rejecting
> in discrete VCF and VCA configurations ) but I need three mached pairs ( for
> mirror in emiter , for active load , and for difpair ) Is there any simpler
> configuration ( simpler = less parts and less matching ) with good Cv
> rejection ? The simplest would be one common colector transistor with
> controled curent source in colector but it would have terrible Cv rejection
> and could only work in VCA ( has anyone seen VCF built around this ) ?
Have you considered having another diff-pair to follow up? Let the inputs of
the second diff pair be at the collectors of the first diff pair. Just a
There are books describing such things, Gray and Meyer springs to mind.
More information about the Synth-diy