[sdiy] Aries modular schematics detail

[jh.]

> A diode only variation of the phase shift ladder (has anyone seen it
> before?)
>
> http://www.musicsynthesizer.com/Diode%20Phaser/Diode%20Phaser.html
>
I thought Jurgen would be amused and the diode only topology is rather
interesting. Opening the collectors increases the attenuation of the ladder.
Connecting the collectors to the base make no difference under simulation.

I think I have made simulations with diodes, but that was many years ago,
and
my memory isn't the best.

I *think* the diode version looked good for a single stage or even a couple
of stages,
but for many stages it's too far from APF to produce deep notches when mixed
with
the straight signal. I do remember a discussion with one list member
(probably
Joachim ?), to whom I reported that the diode version worked, but later had
to
correct myself.

Grant, now that you have set it all up in your simulator, please try a
longer
ladder to see if you get similar results - would be faster than me digging
up these
old Intusoft Spice files.

JH.





> > From: houshu at dsp.cl.nec.co.jp
> > Reply-To: houshu at luke.dsp.cl.nec.co.jp
> > Date: Mon, 21 May 2001 10:35:56 +0900
> > To: jhaible at t-online.de (jh.)
> > Cc: grichter at asapnet.net, synth-diy at node12b53.a2000.nl
> > Subject: Re: [sdiy] Aries modular schematics detail
> >
> > Hello JH and List,
> >
> > I was wrong.  JH is right.  Transistor ladder in Quadra phaser is an
> > almost perfect all pass filter.
> >
> > I checked my note and analyzed again in the weekend.
> >
> > First of all, my memory was wrong. What I analyzed was not phaser, but
> > low pass filter using Quadra style transistor ladder.
> >
> > Anyway, my analysis results are:
> >
> > * LPF based on EMS style diode ladder (EDL) has different transfer
> > function (frequency response) from LPF based on Moog style
> > transistor ladder (MTL). (Very famous as Grant mentioned.)
> > * Quadra style transistor ladder (QTL) can be an all pass filter
> > (APF).
> > * EDL can not be an all pass filter (APF).
> > * MTL can not be an APF.  At least very difficult.
> > * (QTL based LPF, which I imagined from Quadra, does not have the
> > same transfer function as MTL based LPF nor EDL based LPF.)
> >
> > I had a very interesting weekend.  Thank you so much.
> >
> >> Date:    Fri, 18 May 2001 11:32:54 JST
> >> From:    jhaible at t-online.de (jh.)
> >>
> >>>> I don't think the use of transistors is essential to the design, such
as
> >> the
> >>>> difference between the Moog transistor ladder and the EMS diode
ladder.
> >>>
> >>> I agree with you here.  I analyzed the transistor ladder phase shift
> >>> network using a simple transistor model (voltage controlled current
> >>> source), and found that it is equivalent to a diode ladder.
> >>
> >> I think I did the same analysis some years ago, but came to different
> >> results.
> >> But that's all I remember - I must look it up again.
> >>
> >>
> >>> Actually, I suspect that the transistor/diode ladder networks (not
> >>> Moog complementary transistor ladder for high pass filter) can not be
> >>> exact phase shifter (all pass filter), because each stage is not
> >>> isolated.
> >>
> >> If memory serves, *that* was the difference between using diodes and
emitter
> >> followers. Single stage would be the same, but transistors prevent
loading
> >> effects. But as I said, memory is very cloudy here, I must look again.
> >>
> >> Transistor ladder is still no perfect all pass filter, though. That's
why
> >> this
> >> phaser sounds much better with resonance than without I guess.
> >>
> >> JH.
> > - -- ---- ----------- --- --- - ----- -- --- --- - - ------ - - - --
> > Osamu HOSHUYAMA                    NEC Corporation  Kawasaki JAPAN
> > --------------------------------------------------------------------
> >
> >
> >
>
>