<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On 26 December 2017 at 23:30, Neil Johnson <span dir="ltr"><<a href="mailto:neil.johnson71@gmail.com" target="_blank">neil.johnson71@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">David,<span class="gmail-"><br>
> Can you tell us briefly what these "welcome changes" are? I, for one, was<br>
> hoping that they'd put the stupid stability networks on the chip, saving a<br>
> pile of 560pF caps and 510R resistors.<br>
<br>
</span>For one thing, if they did manage to put four 560p caps on the die<br>
then you wouldn't be able to afford it - the die area would be so<br>
large the simple cost plus reduction in yield would push the price way<br>
up, far more than the savings of external components. So, no, you<br>
still need external compensation networks on each input.<br></blockquote><div><br></div><div>Also, the RC value of the compensation network needs to be changed a little bit if you use input resistances far from the typical values, so it might be good to have them outside after all.</div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Main improvements are: lower noise,</blockquote><div><br></div><div>Actually they are a tiny bit noisier (about 1/2 to 1 dB from the data I've seen so far) than the V2164, but a tiny bit lower noise (again about 1/2 to 1 dB) than the original SSM2164 if you swap them in the same circuit. But the reduced distortion can allow for a lower-noise circuit design for a given task.</div><div><br></div><div>/mr</div><div><br></div></div></div></div>