Hello Hardware freaks,
as an output driver I want to use a (dual) N-channel MOSFETs, a IRF7530, see http://www.irf.com/product-info/....
I have connected the gates of the MOSFET to general I/O pins of an ATtiny26. Datasheet for ATtiny26 is at http://www.atmel.com/dyn/resourc....
The drains should switch the loads to ground (source potential). The loads are connected on the other side to a higher voltage (12V) than the AVR (5V).
Since the gates of the MOSFETs are very high-impedant, I want to ensure that the MOSFETs are not conducting all at the same time, if
(1) there is no Vcc voltage for the AVR, but on the load
(2) if the AVR is in reset state.
(3) if the AVR starts up.
I tried, what happens, when there is no pull-down resistor:
in case (1) the MOSFETs are not conducting,
in case (2) the MOSFETs hold its last states.
in case (3) the MOSFETs are conducting for a short time.
Case (1) and (2) are both OK for me.
But can I know it for sure, that it will be like this under all conditions?
Case (3) can be avoided by a pull-down-resistor (I used 10kOhm). (Maybe also by reducing the start-up time specified by the fuse bits.)
It is said in the datasheet that the general I/O pins are set to inputs and pull-ups off after a reset => high-impedant.
But what happens during start-up with the general I/O pins (when the voltage rises to Vcc)?
If possible, I would like to omit the additional pull-down resistors between the gates and the sources / GND.
But I want to avoid any latch-up-problems (12V before 5V present), but also spent not more parts than necessary.
Is there a way to safe the resistor?
P.S: By the way: what is the correct adjective: high-impedance (http://www.leo.org) or high-impedant (datasheet)?