Re: gEDA-user: basic anti-EMI design q

[DJ Delorie <dj@xxxxxxxxxxx>]

> Looking at the pcb w/ chips, optos and triacs.  It probably be good to get 2"
> between the chips and the triacs.

That would be tricky, since the box only allows a 3.5" board.

> Also, optoisolators need low impedance drive circuits.  To achieve
> hi speed with them they need power.

Low speed.  Cycles are on the order of minutes.

>  Even if you don't want them to fire quickly (can't
> tell if your software is running them as dimmers)

Nope, stricly on/off.  The resistor on the opt's driver is the minimum
I can get away with to limit current through the opto.

> there is a benefit to a forceful drive.

The new board will use a fet to drive the LED side of the opto, in
case the CPU can't drive it.

> It's that interference/surges getting onto the lines between the
> opto and whatever is driving it (your latch it appears) will get
> weighted down by the low impedance at this point ... making it
> harder for those surges/emi to swim upstream and latch your logic,
> or trip your latch.

Hmmm... would it help if I timed the opto transitions to occur at a
zero crossing?  I was planning on putting a sensor on the AC anyway,
as an additional clock and power detect, but I could use it to
syncronize the optos.  Remember, no dimming - stricly on or off, for
minutes on end.