[sdiy] Interpreting four-pole-w-feedback circuits

Ian Fritz ijfritz at comcast.net
Thu Mar 6 20:53:20 CET 2008

At 11:22 AM 3/6/2008, René Schmitz wrote:

>It doesn't really matter if they are, if you have an even number of them.

I count the input transistor 64 as one of them, so I count five total.

But, maybe I'm really missing something. Are we looking at the same patent? 
(4,011,466).  There are four transconductors plus the input transistor (64) 
in the feedback path. The patent says:

"At frequencies below cutoff there is a net sign inversion between the base 
of transistor 64 [70 in the text is a typo, since 70 is a resistor] and the 
output terminal 26c of the filter; thus the feedback to transistor 64 is 

This seems correct to me. (But of course I'm just an old man.)

This also agrees with the Verghese drawings, as far as I can tell, where Q9 
is labeled the "inverting
input" and also receives the feedback signal in the second diagram.

I just don't see any difference between the Arptech drawing and the patent 
drawing.  The correspondence seems to me to be:

64 -> Q9
30a -> Q1
30 -> Q3
30b -> Q5
30c -> Q7
26c -> Z1d p9 -> output 5
feedback to 64 -> feedback to inverting input 2

Sorry, but it all looks correct and consistent to me.  Did I misread something?


More information about the Synth-diy mailing list