[sdiy] Polyfuse Tripping Current values

[David G. Dixon]

> Not sure I explained myself that well - but the replies still make sense
> for the first question. Put the polys on the module/pcb/breadboard I'm
> testing - not the breakout module that will be powering it during testing.
> Fair enough.
> 
> Second question still stands though - what should the value of the poly be
> in relation to the current draw of the module it's protecting? Next size
> up? +10%? Double?
> 
> And I guess while I'm asking basic questions - what is the poly actually
> protecting against? Is it protecting the circuit from the power supply or
> the supply (and other modules?) from the circuit?
> 
> I've changed a whole bunch of fuses - but I've never added one to a
> circuit, so apologies if this is on the verge of newb.


I just went through this very same line of enquiry, and it seems to me that
the big thing a polyfuse protects against is improper power pin connection
and power rails shorting to ground.  As such, it is important to add a
back-biased rectifier diode (such as 1N4001) between each rail and ground
after each polyfuse.  The rectifiers protect the circuit against improper
power connections during the time it takes the polyfuses to pinch off.  The
polyfuses protect the rectifiers from burning out due to the excess current.

Concerning the trip current, this is a good question for John Blacet as I
believe he employs polyfuses on all of his modules, but I would guess that,
as a matter of cost, one would probably stock a single value which is large
enough to accommodate any module (100mA, perhaps).  When it comes to current
protection, it's probably more of an all-or-nothing proposition than a
matter of degrees.