[sdiy] Problems with linearized SSI2164 ?

Ian Fritz ijfritz at comcast.net
Thu Feb 14 03:27:34 CET 2019


There is an old saying hereabouts: "Weirdness is caused by hf 
oscillations." This is obviously one of the first things to consider in 
a situation like this. Especially when a change has been made that 
reduces the stability by a factor of 5.

If you make a change in a design and it doesn't work reliably, then 
guess what? You should change it back.


On 2/13/2019 4:53 PM, Neil Johnson wrote:
> Hi David,
>  > The datasheet says (in large boldface type):
>  >
>  > "FEATURES: Improved Direct Replacement for SSM2164/V2164"
>  >
>  > So, the datasheet is literally screaming that it is a direct replacement.
>  > In my book, that means a drop-in replacement.
> You can drop an SSI2164 into a properly designed circuit, and it will 
> likely work as-is, or with minimal tweaks.  But this is analogue 
> electronics, not Lego.
>  > In any case, I believe that the circuit Dan was using it in is 
> exactly the
>  > one given by Phillip Gallo in the datasheet, and it ain't working.  That
>  > tells me that the ICs are defective.
> In this case, no - the circuit as built by Intellijel was not as Irwin 
> or Gallo designed, as the loop compensation capacitor was wrong but a 
> factor of 5ish (22p vs. 100p).
>  > 2) If the SSI2164 is NOT a drop in replacement for the SSM2164 (or 
> it's true
>  > drop-in replacement, the Coolaudio V2164), then I would submit that it is
>  > the most facking useless IC in the history of ICs.  There are 
> hundreds (if
>  > not thousands) of designs using the 2164, and if these all have to be
>  > changed, that's an enormous cost and a major headache for all involved.
> That is rather harsh and unfair when the circuit as built by Intellijel 
> was not the one published by Irwin or Gallo.  Details matter, and when 
> you get the details wrong, you don't get the results you wanted.  To 
> blame the chip vendor is churlish.
>  > I suspect that this thread alone will cost SSI a major chunk of 
> business, at
>  > least in the short term.  If it were me and I were reading this thread, I
>  > wouldn't touch these shitty ICs with a ten-foot pole.  Good job SSI!
> Really, David, is that language necessary in a public forum?  The 
> SSI2164 is a great chip, packs a lot of useful functionality, is 
> actively supported by the manufacturer (including a very detailed 
> datasheet), who is actively engaging in the analogue synth community, 
> and is a definite improvement over the SSM2164.
> But you're right about one thing: /Good job SSI!/
> Cheers,
> Neil
> -- 
> http://www.njohnson.co.uk
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy


More information about the Synth-diy mailing list