[sdiy] Quick query about 2164 VCAs

Tom Wiltshire tom at electricdruid.net
Sun Mar 17 01:06:47 CET 2019

Hi Neil,

What makes you think that:

A) this parallel-inputs-independent-outputs arrangement only works if the CVs are identical?


B) the outputs have to be paralleled too?

I’ve had a look at the SSI2164 datasheet, and I’m not seeing anything that suggests this shouldn’t work, although like Matthias said, I’m not seeing anything that confirms that it should either. It really seems to depend on whether the input current divides into the two VCAs equally or not. If it does, I don’t see a problem. If it doesn’t, or if the division of the current is affected by the CV (and hence would only divide equally with equal CVs) then it won’t work.


> On 16 Mar 2019, at 22:15, Neil Johnson <neil.johnson71 at gmail.com> wrote:
> Hi Matthias,
>> Another quick answer: I don't see why.
>> How could the current know that it should be divided into two equal parts?
>> This reminds me of the discussion last year about paralleling 2164s. I'm using three 2164s in parallel with a separate Rin for each VCA, and somebody said that those three resistors could be swapped against one common resistor feeding all the VCAs. I was then asking how the current into a virtual earth (or even worse, three less than perfect virtual "earths" that sits on three slightly different voltages) would know that it would split into equal halves, but IIRC there was never an answer to that. :-)
> This kind of ended up at the current version of the SSI2164 datasheet
> which shows how to connect multiple cells in parallel.  Yes you can
> connect the inputs in parallel, and you scale the compensation network
> accordingly.
> I suspect, however, that this arrangement only works if all the
> paralleled gain cells are working at the same CV and that all the
> outputs are also paralleled.
> Neil
> -- 
> http://www.njohnson.co.uk

More information about the Synth-diy mailing list