 |
 |
| View previous topic :: View next topic |
| Author |
Message |
blak3r
Joined: 11 Jan 2004
Posts: 45
|
| Using 2 UARTS @38400 = PROBLEM, @9600 = OKAY |
 Posted: Thu May 11, 2006 9:01 am |
 |
|
Target Details:
18F6527
Clock = 20Mhz Xtal
I am testing my board and i have two serial ports controlled by the pic. To verify they both work i connected them together appropriately. (One is a DTE and one is DCE so i don't need null modem) The UART on PIN G1 & G2 is solid. It always works. But, the one on C6 and C7 isnt.
The code below kept failing - so i hooked up a DLA and found that even though the baud rate was set to 38400 the actual baud rate the uart on c6&c7 was putting out was 10,000bps (period of each bit was 100uS). I then changed the rs232 baud rate to 9600 for both and the code ran successfully.
Any ideas?
| Code: | #use rs232(baud=38400, xmit=PIN_G1, rcv=PIN_G2, ERRORS, STREAM=DB9)
#use rs232(baud=38400, xmit=PIN_C6, rcv=PIN_C7, ERRORS, STREAM=RJ45) |
is how i have them setup.
This is the complete code in my main - variable declarations:
| Code: |
lcd_init();
lcd_putc("Starting loop Tst\r\n");
// CLEARS EACH BUFFER - just in case
while(kbhit(RJ45) )
fgetc( RJ45 );
while(kbhit(DB9 ) )
fgetc(DB9 );
// TEST READING CHARS ON - RJ45
fputc('A', DB9 );
choice = fgetc( RJ45 );
if( choice == 'A' )
lcd_putc("SUCCESSFUL READ");
else {
lcd_putc("FAILED READ: ");
lcd_putc( choice );
printf(lcd_putc, ", 0x%X", choice );
}
lcd_putc("\r\n");
// TEST WRITING CHARS ON THE RJ45 Serial Port
fputc('T', RJ45 );
choice = fgetc( DB9 );
if( choice == 'T' )
lcd_putc("SUCCESSFUL WRITE");
else {
lcd_putc("FAILED WRIT: ");
lcd_putc( choice );
printf(lcd_putc, ", 0x%X", choice );
}
lcd_putc("\r\n");
|
_________________
"Everything should be made as simple as possible, but not one bit simpler" -- Albert Einstein
http://www.blakerobertson.com |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu May 11, 2006 9:42 am |
|
|
| Quote: |
Even though the baud rate was set to 38400 the actual baud rate the uart
on c6&c7 was putting out was 10,000bps (period of each bit was 100uS). |
Post your compiler version. |
|
|
blak3r
Joined: 11 Jan 2004
Posts: 45
|
|
| Posted: Thu May 11, 2006 10:22 am |
|
|
PCM Version 3.184
_________________
"Everything should be made as simple as possible, but not one bit simpler" -- Albert Einstein
http://www.blakerobertson.com |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu May 11, 2006 10:56 am |
|
|
| Quote: | Target Details:
18F6527
PCM Version 3.184
|
You must mean PCH, because you're using an 18F PIC.
I don't have PCH vs. 3.184. The closest I have is 3.187.
I installed that version and tried to compile a test program for the
18F6527 and I get this error message:
| Quote: |
Error 24 "C:\PROGRAM FILES\PICC\devices\18F6527.h" Line 2(9,19):
Unknown device type "PIC18F6527" |
That version of the compiler doesn't support it. I tried PCH 3.188
and it doesn't support it either. |
|
|
blak3r
Joined: 11 Jan 2004
Posts: 45
|
|
| Posted: Thu May 11, 2006 11:02 am |
|
|
You're right,
PCH 3.222
I also have PCM. My mistake.
18F6527 is supported in 3.222
_________________
"Everything should be made as simple as possible, but not one bit simpler" -- Albert Einstein
http://www.blakerobertson.com |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu May 11, 2006 12:15 pm |
|
|
The closest version I have to yours is PCH vs. 3.231. The .LST file
looks OK for that version. They are setting up both ports to use a
16-bit baud rate generator. The baud rate formula for that mode is:
| Code: |
Fosc
-------------
4 * (n +1) Note: n = the 16-bit SPBRG value
|
which gives:
| Code: |
20 x 10e6
------------- = 38461.53 baud
4 * (129 +1)
|
That's very close to the desired value of 38400.
Compile this test program and see if the generated code in the .LST
file matches the .LST file shown below it. Specifically look at the values
that are put into these registers:
BAUDCON1
SPBRG1
SPBRGH1
TXSTA1
Also look at the generated code for fputc(0xAA, RJ45). Make sure
it looks similar to the code shown below. It could be that your version
is creating a software UART instead of using the hardware UART.
| Code: |
#include <18F6527.h>
#fuses HS, NOWDT, NOPROTECT, NOBROWNOUT, PUT, NOLVP
#use delay(clock=20000000)
#use rs232(baud=38400, xmit=PIN_G1, rcv=PIN_G2, ERRORS, STREAM=DB9)
#use rs232(baud=38400, xmit=PIN_C6, rcv=PIN_C7, ERRORS, STREAM=RJ45)
//==============================
void main()
{
fputc(0x55, DB9);
fputc(0xAA, RJ45);
while(1);
} |
| Code: | .................... void main()
.................... {
0004: CLRF FF8
0006: BCF FD0.7
0008: CLRF FEA
000A: CLRF FE9
000C: BSF F7C.3 // BAUDCON2 Use 16-bit SPBRG
000E: MOVLW 81
0010: MOVWF F6F // SPBRG2
0012: MOVLW 00
0014: MOVWF F7D // SPBRGH2
0016: MOVLW 26
0018: MOVWF F6C // TXSTA2
001A: MOVLW 90
001C: MOVWF F6B // RCSTA2
// Setup register for the UART on pins C6 and C7:
001E: BSF F7E.3 // BAUDCON1 Use 16-bit SPBRG
0020: MOVLW 81
0022: MOVWF FAF // SPBRG1
0024: MOVLW 00
0026: MOVWF F7F // SPBRGH1
0028: MOVLW 26
002A: MOVWF FAC // TXSTA1
002C: MOVLW 90
002E: MOVWF FAB // RCSTA1
0030: MOVF FC1,W
0032: ANDLW C0
0034: IORLW 0F
0036: MOVWF FC1 // ADCON1
0038: MOVLW 07
003A: MOVWF FB4 // CMCON
003C: CLRF 05
....................
.................... fputc(0x55, DB9);
003E: MOVLW 55
0040: BTFSS FA4.4
0042: BRA 0040
0044: MOVWF F6D
....................
.................... fputc(0xAA, RJ45);
0046: MOVLW AA
0048: BTFSS F9E.4
004A: BRA 0048
004C: MOVWF FAD |
|
|
|
blak3r
Joined: 11 Jan 2004
Posts: 45
|
|
| Posted: Thu May 11, 2006 2:46 pm |
|
|
See lines 0024 and 0036, i believe that's the cause of the bug.
| Code: |
0004: CLRF FF8
0006: BCF FD0.7
0008: CLRF FEA
000A: CLRF FE9
000C: MOVF FC1,W
000E: ANDLW C0
0010: IORLW 0F
0012: MOVWF FC1
0014: MOVLW 07
0016: MOVWF FB4
0018: MOVF FB4,W
001A: BCF FA1.6
001C: BSF F7C.3
001E: MOVLW 81
0020: MOVWF F6F // SPBRG2
0022: MOVLW 00
0024: MOVWF F70 // != SPBRGH2 (should be F7D)
0026: MOVLW 26
0028: MOVWF F6C
002A: MOVLW 90
002C: MOVWF F6B
002E: BSF F7E.3
0030: MOVLW 81
0032: MOVWF FAF // SPBRG1
0034: MOVLW 00
0036: MOVWF FB0 // != SPBRGH1 (should be F7F)
0038: MOVLW 26
003A: MOVWF FAC
003C: MOVLW 90
003E: MOVWF FAB
0040: CLRF 05
.................... fputc(0x55, DB9);
0042: MOVLW 55
0044: BTFSS FA4.4
0046: BRA 0044
0048: MOVWF F6D
....................
.................... fputc(0xAA, RJ45);
004A: MOVLW AA
004C: BTFSS F9E.4
004E: BRA 004C
0050: MOVWF FAD
|
_________________
"Everything should be made as simple as possible, but not one bit simpler" -- Albert Einstein
http://www.blakerobertson.com |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu May 11, 2006 3:50 pm |
|
|
According to the data sheet, the SPBRGH1 and SPBRGH2 registers are
set to 0x00 upon power-on reset. So that bug probably isn't a problem.
It's writing 0x00 to the CCP5CON and PSPCON registers by mistake,
but those registers are also set to 0x00 upon power-on reset.
Just to be sure, you could add the following code to your program
to test it.
| Code: | #byte SPBRGH1 = 0xF7F
#byte SPBRGH2 = 0xF7D
void main()
{
SPBRGH1 = 0;
SPBRGH2 = 0; |
Also, can you try the small program that I posted ? Does it also
cause the problem with the incorrect baud rate on Pin C6 ? |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
