> Interestingly, if /MCLR is grounded when power is applied the LCD doesn't
> initialise. Nor does it when you release the button for /MCLR to go to it=
s
> usual high. Using the LATB,0 markers to view progress, it seems code is n=
ot
> being executed at all. The oscillator (HS crystal) is going though. The p=
ower
> needs to be cycled with /MCLR high to get the s/w running as you'd normal=
ly
> expect it to.
>=20
> So the button is doing "something". I must be missing a h/w directive
> somewhere but the path between pin and internal reset signal is pretty
> straight-forward

This suggests that the rise time on the reset pin may be too long. Use a sm=
aller cap or lower value resistor to Vdd and see what happens.

It may be that when powering up the board the internal brownout circuitry i=
s holding things reset long enough for the /MCLR pin to get to a high enoug=
h voltage for this permutation not to be affected.

Haven't looked at the data sheet to see what rise time is required, but I k=
now it can't be too long on some chips.



--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.