Fastest baud rate w/PBP?

[jrprogrammer]

So I suddenly need to go faster than 38,400 with a PIC16F883 and a 20 MHz xtal. I'm sending out on 4 pins, with no MAX232 chips involved. The good book says a max of 38,400, but haven't I seen postings that say (up to)115,300???.. Do I need to write a routine in assembler to do this?
Thanks for listening,
JR

[HenrikOlsson]

If you're using SEROUT(2) then I think the answer is no. There's a formula that tells you how to specify the baudrate bits and for 115200 the result of that formula comes out negative = no go. I'm not sure but it's possible that DEBUG can work with higher baudrates, it might be worth a shot.

With that said, that chip you're using has an USART, can't you use that?

/Henrik.

[jrprogrammer]

Thanks for the reply. Yes the PIC has a USART, but I need to send serial data out a total of 4 pins... If you use Mister E's Multicalc, 115200 does give me a number. I have a more precise xtal coming. Right now my frequency counter says 19.84 MHz which would really screw up the baud rate. Granted, my 10:1 scope probe could be pulling the frequency that I read.
I'll report back, once I get the new xtals.
JR

[tumbleweed]

If you use Mister E's Multicalc, 115200 does give me a number

What number does it give?
If you look at the table of numbers for SEROUT2, it pretty much maxes out at 38400.

I have a more precise xtal coming. Right now my frequency counter says 19.84 MHz

You're loading it with the scope probe. Better to measure it a different way.

Even so, while that's not a great number for an xtal it's less than 1% error so the serial baud rate would probably be fine (if that table produces accurate numbers to begin with). I don't think a better xtal is going to do much for you.

Since you're interested in the serial out data, perhaps just transmit a 'U' character ($55) and look at the individual bit times.