[PIC] Maximum voltage for PIC's input

[burrow]

Hai,
My circuit is powered by an xformer and also a battery ( parallel) so when xformer turns off , the battery takes over. its a 12v xformer , it goes through rectifier and everything and a DIODE ( The xformer circuit has some other devices too, i dont want battery to feed them, hence using the diode ) and then to 9v rechargeable battery and then to 7805 ic.

so inorder to feed power status input to pic, i took a connection from a point before the diode and then fed into another 7805 ic and then to pic's pin. But it doesn't seem to work, it turns of the controller.

iguess since the main 7805 ic has more load its voltage might be around or around, but the voltage on the Second 7805 is negligible, so the pin might be actually getting higher voltage than mcu's vcc.

So is that causing it to freeze ?
will adding a resistor work ? , whats the allowable voltage range for an input pin , tolerance?

 

[esp1]

hi,
Pins on a PIC have clamp diodes to +V and to 0V, so when the voltage on the input pins exceeds +V + ~0.7V, the clamp diode will conduct.
Its important that the current flowing in to the pin, is resistive limited to less than ~20mA else the diode will blow and then the internal pin logic.

If the total load current drawn from the 2nd 7805 feeding the sample pin is less than the 7805 d/s specified minimum the 7805 will not regulate correctly.
E

 

[burrow]

Oh i see,
so what if i use a 1k resistor in series with that sample pin , also a 10k pull down

 

[esp1]

If the resistive voltage is chosen to give at least the High level threshold voltage, the PIC will see it as a High =1.
Its advisable to have a higher voltage than the High threshold as any 'negative going ' noise spikes will be seen as a Low=0

Your combination will give a voltage out of 5V * (10/[10+1]) ie: 4.54V which should be OK, I would add a 10nF cap across the 10K, which will help to reduce any 'noise' at the PIC input pin.

E