[sdiy] analog or digital ground?

[Seb Francis]

Hi Magnus,

Magnus Danielson wrote:

> Well, if my reasoning does not break, all the MAX306 powerlines are "digital",
> but in reality we can consider them analog, we want a quite powerfeed if we can
> get it. Now, if we run it on analog power (which means V+, GND and V- is tied
> to the analog power side), we still want to provide an AC path to the digital
> ground. I propose that this is done by tossing a pair of 10 nF caps from the
> digital ground, to the V+ and V-

I disagree here.  From the appnote it seems to say that, rather than the MUX injecting noise into the power rails from the digital part, the analog part (the switching FETs) are in fact sensitive to noise from the power rails due to the gate capacitance.

Therefore I would say it's rather better to decouple the power lines to the quiet analog ground which the analog signals are relative to.

In the appnote they acutally recommend an RC network, although I have used just a 100nF cap very close to the chip because I can't find a place on the PCB to squeeze in another resistor or ferrite bead for each power line.  Probably I should shuffle everything around to make a space but it's pretty tight, and the +/-15V rails ought to be pretty quiet anyway - they're only running properly decoupled (and relatively low frequency) analog chips.

> but not the MAX306 GND, since this actually
> have not much with the actual digital signal.

I agree with you the GND connection seems pretty arbitrary.  I went for digital in the end because in the MUX it seems to have slightly more to do with the digital signal than the analog.

> I assume that the digital source
> (the PIC in your case) is properly decoupled.

Yeah it has a 100nF cap to digital ground on the +5V pin.  The 5V supply has its own regulator (and electrolytic) that just powers the digital stuff (with the exception of the +5V DAC and gate output transistors which are separated from the digital stuff with a ferrite bead and another electrolytic).

Or do you mean decoupling of the digital logic signals themselves?  I have seen this done somewhere I think, but I haven't done that in this case.

>
> Hope you got things sorted out, or at least confused at a much higher level ;O)

A bit of both I think ;-)  That Maxim appnote was very useful though.

Seb