Oli Glaser wrote:
> [SNIP]
> Out of interest, why can't you just use a diode? 
>
>   
OK, so tell me if I am thinking correctly here.....

Under normal conditions (supply polarity is correct) the drain sees the 
positive supply voltage and the reverse body zener diode conducts, 
allowing current flow. In the "normal" implementation of this circuit 
(VSupply<VGS), you would connect the gate to ground and since it stays 
well below the source, the MOSFET junction also conducts, right? (not 
sure about this part...) I guess the advantage of this (if it actually 
happens) is that you get a lower on resistance than the diode alone.

But the gate cannot be allowed to become more than 20VDC negative with 
respect to the source, according to the data sheet. So the reverse 
biased 16V zener keeps the gate at 16V below the drain, and the source 
voltage = (drain voltage-VDS). The MOSFET is turned on, but the VGSMax 
spec is not exceeded.

Reversing the polarity of the input results in the gate seeing a 
positive voltage with respect to the drain. The source will be at 0V, 
(or floating; the rest of the circuitry acts as a pull down to ground) 
since the MOSFET is off....the gate must be kept within +/- 20VDC of the 
source. So the opposite polarity zener does not allow the gate to rise 
above 16VDC with respect to ground.

But, if the MOSFET junction does not actually turn on when the circuit 
sees correct input supply polarity, then I guess you don't need the 
zeners.....because the MOSFET never actually turns on, only the body 
diode does. But then, why attach the gate to ground, if not to turn on 
the MOSFET?

Does that make sense.......somebody straighten me out........

Mark


-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist