| View previous topic :: View next topic |
| Author |
Message |
Miniman
Joined: 30 Apr 2007
Posts: 44
|
| RS232 receive help! |
 Posted: Thu Nov 29, 2007 3:54 pm |
 |
|
Hello
I'm desperate. I have worked hours with this simple thing. I cant get the PIC to receive anything from the PC via rs232. I have done this before with exactly the same external hardware with a 16f870 but now I'm using a 18f4550. I se a signal which appers to be good, on my oscilloscope but the receive part does not work thou sending to the PC from PIC works great. All I have done is this
| Code: |
#use rs232(Baud=19200, Xmit=PIN_C6, Rcv=PIN_C7)
void main()
{
char c;
while(1)
{
c = getc();
putc(c);
}
}
|
Does anybody know what could be the problem? It all seams very strange to me and I simply can't take this any more..
please help me!
/miniman |
|
 |
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu Nov 29, 2007 4:14 pm |
|
|
1. Post the #include, #fuses, and #use delay() statement.
2. Post your compiler version.
3. If you're using a factory built board, post the manufacturer and
part number of the board. If you built the board by yourself, then
post a link to a schematic, if possible. |
|
|
Miniman
Joined: 30 Apr 2007
Posts: 44
|
|
| Posted: Fri Nov 30, 2007 1:42 am |
|
|
This is my include and delay
| Code: |
#include <18F4550.h>
#use delay(clock=48M) // 48MHz External Clock
|
I do not use fuses, the configbits are set once upon programming in the bootloader.
My compilerversion is 4.013.
I'm using a homebuilt board and I don't have any schematic, sorry.
I'm also using decoupling capacitors on both sides of the PIC (two Vdd and two Vss).
Any suggestion what the problem could be?
/miniman |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Fri Nov 30, 2007 2:19 am |
|
|
| Quote: | I do not use fuses, the config bits are set once upon programming in the
bootloader. My compiler version is 4.013.
I'm using a homebuilt board and I don't have any schematic, sorry. |
I don't have 4.013. The closest I have is vs. 4.014. I compiled your
program with that version and compared the .LST file to vs. 4.064.
They're very similar. Vs. 4.604 uses BRGH mode for the Baud Rate
generator. I used the PicBaud baud rate calculator and it says the
SPBRG register value for 19200 baud with a 48 MHz oscillator is 38.
That's the same as 0x26 as shown in the .LST below. See this thread
for a link to the PicBaud utility program:
http://www.ccsinfo.com/forum/viewtopic.php?p=88734
So, assuming that vs. 4.013 produces the same code as 4.014 below,
I don't see a problem with the code. What does that leave ?
1. The bootloader fuses may not be correct for 48 MHz operation.
2. The external crystal frequency may not be correct for the existing
fuses.
3. There may be something wrong with the hardware such as missing
the pull-up resistor on MCLR (assuming that MCLR is enabled).
4. I suspect something is wrong with the serial port connections
which includes the MAX232 chip and the connections to the PC.
Part of the .LST file for vs. 4.014:
| Code: | .................... void main()
.................... {
0004: CLRF FF8
0006: BCF FD0.7
0008: CLRF FEA
000A: CLRF FE9
000C: BCF FB8.3
000E: MOVLW 26
0010: MOVWF FAF // SPBRG
0012: MOVLW A2
0014: MOVWF FAC // TXSTA
0016: MOVLW 90
0018: MOVWF FAB // RCSTA
001A: MOVF FC1,W
001C: ANDLW C0
001E: IORLW 0F
0020: MOVWF FC1 // ADCON1
0022: MOVLW 07
0024: MOVWF FB4 // CMCON
.................... char c;
....................
.................... while(1)
.................... {
.................... c = getc();
0026: BTFSS F9E.5 // RCIF bit
0028: BRA 0026
002A: MOVFF FAE,06 // RCREG
.................... putc(c);
002E: MOVF 06,W
0030: BTFSS F9E.4 // TXIF bit
0032: BRA 0030
0034: MOVWF FAD // TXREG
.................... }
0036: BRA 0026
.................... } |
|
|
|
Miniman
Joined: 30 Apr 2007
Posts: 44
|
|
| Posted: Sat Dec 01, 2007 3:34 am |
|
|
It seams like there is a software problem..
If have tested this code now:
| Code: |
#INT_RDA
void rda_isr()
{
putc('R');
rcvChar = GetC();
printf("%u",rcvChar);
putc(' ');
}
|
With this code I get "R146 R146 R146 ".. printed all over again till i shut the pic down. but if I replace the GetC(); with RCREG; i get 3 chars in the begining when i send my chars out, but not later when i push a key. Do I have to clear the buffer? I thought it would clear itself when I read RCREG, but I also thought that I wouln't need to use registers... Shouldn't Getc() work? I'm getting crazy over this...
I also need help with the configbits, I think the these gonfigbits will work:
But wich fuses to use to get them? I have never used fuses before...
/miniman |
|
|
Miniman
Joined: 30 Apr 2007
Posts: 44
|
|
| Posted: Sat Dec 01, 2007 7:14 am |
|
|
I have been able to get it to work a little bit now.. :/
If i use this code, no interrupts:
| Code: |
while(1)
{
while(!RCIF);
rcvChar = RCREG;
printf("-%u-",rcvChar);
rcvChar = RCREG;
printf("-%u-",rcvChar);
rcvChar = RCREG;
printf("-%u-",rcvChar);
if(FERR)
putc('W');
if(OERR)
{
putc('E');
CREN = 0;
delay_ms(100);
CREN = 1;
}
}
|
I then get respons to a key-press. when I press a key I get -97--255--255-E It _ALWAYS_ overrun the buffer, the OERR. And if I in my program only read RCREG and prints it ONCE it does not work att all, then it just prints -255-E every loop without stopping and does not respond to a key strike... Gahh What is happening? Could sombody help me?
/miniman |
|
|
Ttelmah
Guest
|
|
| Posted: Sat Dec 01, 2007 10:32 am |
|
|
What you are describing, is exactly what would happen, if there was a hardware problem, and the receive line into the PIC was held low.
You need to check with a scope, what atually is on the receive input (or even just with a voltmeter, that the line is at 5v, when nothing is being sent...
Best Wishes |
|
|
Miniman
Joined: 30 Apr 2007
Posts: 44
|
|
| Posted: Sat Dec 01, 2007 10:44 am |
|
|
I have checked that - It is 5v when nothing is sending.. It all works "fine" when I use the strange code above..
/miniman |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
