16F627A PortA.4 trouble?

[spcw1234]

I am not liking this chip! Maybe I should go back to my more familiar 18F parts!

New problem, I was trying to setup the capture module using timer1 and I accidentally configured timer1 to use an external osc which made me loose control of those pins as general I/O's. Problem is, now my programmer can't re-program the chip to change it. It doesn't detect the PIC. Not sure why an MCLR reset won't allow me to use the timer1 OSC pins as PGC and PGD, but it isn't working. Any ideas?

[spcw1234]

OK, I bricked 2 of these chips now... With this code?

Capture      VAR PIR1.2              
T1CON.5=1     '1:4 prescaler
T1CON.4=0     '1:4 prescaler
T1CON.3=1     'Timer1 OSC enabled
T1CON.2=1     'sychro clock
T1CON.1=0     'internal clock
T1CON.0=1     'start timer

It works fine until I add the above into my program, then it seems portb.7 gets fried and I have to replace the chip to program again.

[HenrikOlsson]

Hi,
Just a shot in the dark here, can it have anything to do with LVP, do you have it turned OFF in the CONFIGs?

What programmer are you using to program the chips? Once they are "bricked" can't you even erase them?

/Henrik.

[spcw1234]

LVP is off, the chip still works 100% except portb.7 (maybe portb.6 too, I don't have anything on that pin) I have an LED on portb.7, which works fine until I add the above code. The programmer is a PICFlash2. It does not recognize the chip at all, won't allow me to erase, read, write, etc... I really need to use the capture module, but I can't keep bricking these chips. Not sure how to proceed from here.

Thanks!

[HenrikOlsson]

Hi,
I think the pin stop working because you enable the oscillator for TMR1, the oscillator uses RB6 and RB7 so they can't then be used as I/O.

I see you're using internal oscillator, are you using internal MCLR as well by any chance?

Looking around there seems to be an issue with using internal oscillator and internal MCLR and initially setting PGD/PGC pins as outputs - or something like that. In that case the device apparently can't enter programming mode without the device programmer applying Vpp before Vdd which apparently not all device programmers can do - again, or something like that. See this for example.

Not saying this IS the case in your case but it does "smell" like it to me.

/Henrik.

[spcw1234]

I am using external MCLR pulled high with a 4k7 resistor. That does sound like what is happening, except I am not using the internal MCLR. I am also using an external VDD, and I didn't terminate a VDD pin on my ICSP connector.

It is a surface mount chip, and I replaced it again.

I plan to order an 18F1220 as that has the enhanced CCP module, which my code is already written to work with. I was just trying to cut costs, but it has been a headache. I need a pin compatible chip to the 16F627A since I already have the PCB's, assuming I don't get this working.

[Bruce]

Clear T1CON.3. If your programmer won't allow you to reprogram them with this bit set, at least you shouldn't have the same problem with new ones. You only enable TMR1 osc if you're using an external oscillator.