PIC18F got some weird problem?

[NatureTech]

Thanks a lot steve,doh!!..i tried i many times but sometimes get rush!!

But steve,sometimes it doesn't work?.

[mister_e]

When you say sometime it doesn't work... what do you refer to?
programming or the blink?

make sure you have set the EPIC software to program the config fuses. I know there's an option somewhere.

when you program it, make sure you (or the programmer) tie/load the PGM pin to gnd.

and the usual, Both VSS and VDD connected, 0.1 uF close to the PIC, MCLR to Vdd.

No loose connection between the PIC and the crystal+Caps. And short distance between the PIC and the crystal.

[NatureTech]

I also found that it resets itself.Could it be the perfboard problem?.

I just made SPI work between 2 PIC ( 18F452 & 18F458 ).It worked but sometimes,it resets and return to main loop although i had set BOR to lowest and make PWT enable + placed 2 npo 7 multilayer ceramic and 2 tantalum cap as near as possible to PIC.The Vreg uses LDO with good quiecent current which i belive would handle load very well.So why sometimes it resets itself?

[NatureTech]

The software part.It sometimes reset itself.I face no problem programming it as i do play with the fuses cuz sometimes you need to activate PLL and so on.So no problem with programming.Everything is ok.But,i face no idea why it don't want to for example:blink a led every 1 second..it's no "go".Maybe i spent so much time on CCS...hehehe...

[mister_e]

Without the whole code and schematic it's hard to say...
Stack-Overflow reset?
Faulty MCLR contact?
BAD config fuses programming (LVP_ON could do some strange thing sometimes)
messy supply line may cause some BOD reset as well
etc etc etc

To me scoping the VDD rail to see if it's clean and monitor the MCLR pin would be my first test...

Well, you could still use CCS to see if the results are the same It have to be!

[NatureTech]

So,you belive this problem nothing to do with the EPIC programmer nor the software,that would be PBP version 2.47.My first thought is that the problem might come from the EPIC programmer as it was stated in the melabs website recommending a 0.1 uf cap between the power line?.Is that really neccessary?.I don't face any problem with the programming.So it's 50-50 in my mind.

Actually , i made this circuit in perf board(donut board) or point-to-point soldering board with 16mhz crystal and a 4.7k pull-up for mclr.That's all.

[mister_e]

it's not a PBP problem.. it's working here on an EasyPIC 4 board.

Is this a EPIC problem.. maybe, i don't have it. And yup the Cap while programming should help. And yup always plae one close to your PIC and, at least, close to all other IC. I always place a 10uF tantalum close to the PIC too.

Which capacitor value did you use around your crystal?

[NatureTech]

I use 22pf npo 7 grade cap.Mmm..it's fine..

Hey,you mentioned about the stack overflow right.Mmmm..Does that always come in a way to this sort of problem.I don't think so.I always set LVP off as this is not the "J" type PIC.

[mister_e]

F... weird. Did you also erase the PIC before programming it?

Once programmed, read it back, and look the config fuses haven't change to something else.

There's some days like that...

Did you also tried to program your PIC in circuit?

I always set LVP off as this is not the "J" type PIC.

Sorry but it has nothing to do with the J type... or, HEHE i misunderstood something awhile back