strange problem

[jimtr6]

I programmed a 12F629 and inserted it into my breadboard and it works fine, so I put it on my perfboard in a socket and it will not work, it has an internal oscillator so my connections are only voltage supply and ground, no other connections to the ports, the program simply turns a port on high when I apply a high to to an input. I took it out put it in the breadboard again and it works, I'm just using a logic probe at this point although on the breadboard it lights the LED if I hook it up. I'm baffled, I even made a second board thinking there was a strange issue with the first, still no go, I'm using two boards, one is perfboard the other is Radio Shacks Archer experimenter board made for TTL, neither one will work. No shorted or open pins, voltage and ground are on the pins of the chip, what could cause this??? much appreciated if I can get any help with this.

[medelec35]

Hello jimtr6. If it was my facing same issue, I will do two things.
1)I would measure continuity between each use leg of chip, and wire that is connected to it. Could be a simple connection issue.
2) Measure Voltage on the i/p of the chip. Make sure it is not floating (should be 0V if using a pull down resistor or 5V if using a pull up resistor if not using analogue part). If not used already you will require a pull up/ or pull down resistor. If works on one type of board but not other, I would say your software is OK. But to be on the safe side if you want to post the hex file It can be double checked for you.
Have you used a pull up or pull down resistor on the i/p?
Without switch connected, you could connect a flying lead to i/p. Then connect the other end of flying lead to +5V then to 0V to check functionality of chip.

[jimtr6]

Hello medelec, thanks for your advice, as you said these floating inputs can cause problems so I did use a pull down resistor, still no go, on the breadboard the only connects are vdd and vss, the input is open, the output will drive an LED or logic probe. This is strange to me, what i did was in process of elimination solder two wires to a socket, vdd and vss, no board here, just a socket with two wires, inserted into voltage and ground it works, next I tried a new socket on a new board soldering at as low heat as I could, the chip is not installed of course but the problem is pointing to the soldering process, the low heat solder seems to work, if this sounds strange I wish you could see this, I'm almost tempted to video this whole thing, it is baffling. But today I will test again to see if my latest soldering is still working. I did wonder if the board was contributing unwanted capacitance. My project by the way is my first, it is a device that lets me utilize one set of computer speakers with two computers feeding them, one computer is the main and is always on feeding the main speakers....except when sound is detected coming from the audio out of the laptop, this is done by feeding the audio of the laptop to an op amp and rectifying it to a positive DC voltage of 5 volts and sending it to port 5 which outputs a high to port 0 feeding a couple transistors to activate two low voltage stereo relays, it is as simple as it gets for a newbie, the program works great on simulate and breadboard......

[Benj]

Hello,

Just a thought but breadboards add a fair bit of capacitance due to their nature. Would it help if you added a 100nf and maybe also a 10uf or 100uf capacitor across the power rails right next to the PIC?

[jimtr6]

Ben, my thoughts exactly, something about the board, I added a filter cap right across the rails 47uF, I really thought that would do it, but still no go, if I added a smaller hi frequency cap like a pico or such wouldn't it still add up to the higher capacitance seeing their in parallel. Has anyone ever had to add small caps across port pins?

[medelec35]

Has anyone ever had to add small caps across port pins?

Only when using a high value (1Meg) pull down resistor. I had to use a 100nF cap across i/p to prevent strange behavior. But that was for analog i/p and not digital. This is an odd one as there is not a lot to go wrong. I would suggest going into 'Chip 'Configure.....', then switch to 'Expert config screen'. Make sure all settings are correct. Settings in attachment should work.

12fF675 Config.jpg (54.02 KiB) Viewed 7514 times

For instance you can have master clear enable set to external. Then on one board GP3 could be high, and on board which dos not work GP3 could be floating.

[jimtr6]

medelec, I went into the expert configure and found two problems at first, the setting was set to external on Master Clear Enable, plus my clock was external though that one I found at the beginning. I wentx back to real simple, a program with no inputs, just flash an LED, but I have learned the importance of tying down your inputs with a resistor to low. Thanks to both of you for your time

[medelec35]

Your welcome.
The reason for asking you to post your hex file was to check your configuration settings. At least all sorted now, up and running. That is good news.
Thank you for letting us know.