How to start with high output pin?

Futterama · Joined: 17 Oct 2005 Posts: 98

Hello,

Simple question: How can a PIC pin start as high output right after selecting the pin as output?

I use PIC12F629 and v.4.057.

I have this code in the beginning of main():

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

Specify the INTRC_IO fuse.

Futterama · Joined: 17 Oct 2005 Posts: 98

Ttelmah · Guest

Do three things.
1) Make sure you have fast_io. If this is not the case, the compiler will add TRIS control statements for you.
2) Update the entire output in one instruction.
3) Do the same for the TRIS.
On the 12, and 16 chips, there is not a separate accessible output latch, there is only a combined I/O latch. As such, updating any individual port bit, will result in the other bits being read at the same time, and the latch contents updated. When updating any single bit, there must be a delay before the next read operation, to allow the bit to really 'reach' the required state (this is known as the 'RMW' problem on the PICs - read, modify, write). Hence updating the whole port is much safer.
So, I'd perform:
#use fast_io(a)

output_a(0b110000);
set_tris_a(0b001100);

Separately, remember you need to ensure the other I/O functions on the pins are turned off. Personally, I'd always add large value resistors, to ensure the pins are biased to the 'safe' state, until the processor is awake. Otherwise it only takes the pin to be charged to a previous 'on' voltage, to result in unacceptable curent levels during boot...

Best Wishes

Futterama · Joined: 17 Oct 2005 Posts: 98

What happened to PCM programmers post?

I read it, and he was so right, I didn't think of that (setting on bit at a time with read modify write).

But when I wanted to quote his reply, it was gone Sad

PCM programmer · Joined: 06 Sep 2003 Posts: 21708

Just after I made my post about the read-modify-write issue, I saw that
Ttelmah had already posted an answer to it. (I wasn't quick enough on
the composition). Since I didn't want to give the appearance of stepping
on his post, I removed my last reply.

Futterama · Joined: 17 Oct 2005 Posts: 98

Ttelmah · Guest

Not at all.
PCM programmer, is one of the people here, who generally, is 'firing on all cylinders', and he is much better at some explanations than me.
Sometimes we end up posting multiple answers on the same thread. Usually I have started composing an answer, ge nearly finished, and then get sidetracked (phone call or something), come back and hit 'send', to find that multiple similar answers have appeared (I think the record, stands at seven!...).
Sometimes, it is really 'nice', to see just slightly different versions, since you then get the thing of seeing the problem from a slightly different 'angle', and can learn more stuff yourself. Smile

Best Wishes

avarachan · Guest

Hi,

I had the same problem of 12f629 lock out. I found a post about brown out fuse here. It solved the problem. I am writing this just in case any one who needs this info can get it.

Scene:I was using 12f629 with GP0 and 1 as unputs, 2 as output. When I power up, sometimes GP2 starts as high, which i dont want to. Bcz I use GP2 to trigger a switch when GP0 and 1 as comparators give a high out. Again, some times when i Power up, the chip locks out, and no response.

Solution: I was having NOBROWNOUT fuse set. I changed it to BROWNOUT. Every thing is fine now.