| View previous topic :: View next topic |
| Author |
Message |
PICoHolic
Joined: 04 Jan 2005
Posts: 224
|
| PIC18F4550 bugs again & again !? |
 Posted: Wed Jul 04, 2007 6:28 am |
 |
|
Hey to all,
I've noticed that PORTB[0] behaves weirdly. when i write a value to PORTB, RB0 sometimes is correct some other times is not. At first i thought that the PIC is damaged until i tried another one. Same problem!
Going through the listing i've noticed that the instruction used is MOVFF
Any comments?  |
|
 |
Ttelmah
Guest
|
|
| Posted: Wed Jul 04, 2007 8:29 am |
|
|
Every pin on portB, on that chip, has other devices multiplexed onto it. Are all these disabled?...
Best Wishes |
|
|
PICoHolic
Joined: 04 Jan 2005
Posts: 224
|
|
| Posted: Wed Jul 04, 2007 8:44 am |
|
|
Yes of course...
| Code: |
#fuses HSPLL,NOWDT,NOPROTECT,NOLVP,NOBROWNOUT,NODEBUG,USBDIV,PLL5,CPUDIV1,VREGEN, NOPBADEN, ICPRT
...
setup_adc_ports(NO_ANALOGS|VSS_VDD);
setup_adc(ADC_OFF);
...
set_tris_b(0x00);
|
Anything else?
10x |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Wed Jul 04, 2007 1:48 pm |
|
|
1. Post a short, compilable test program that demonstrates the problem.
Show all #include, #use, #device, and #fuses statements.
Show all data declarations.
2. Post your compiler version. |
|
|
PICoHolic
Joined: 04 Jan 2005
Posts: 224
|
|
| Posted: Thu Jul 05, 2007 12:36 am |
|
|
Compiler: 4.029
| Code: |
#include <18F4550.h>
#device adc=8
#fuses HSPLL,NOWDT,NOPROTECT,NOLVP,NOBROWNOUT,NODEBUG,USBDIV,PLL5,CPUDIV1,VREGEN, NOPBADEN, ICPRT
#use delay(clock=48000000)
#byte PORTA = 0xF80
#byte PORTB = 0xF81
#byte PORTC = 0xF82
#byte PORTD = 0xF83
#byte PORTE = 0xF84
#byte TRISE = 0xF96
// USB
#include <usb_cdc.h>
///////////////////////////////////////////////////////////////////////////////
#define SetData(X) PORTB = X
#define GetData() PORTB
#bit SS = PORTE.0
#bit SG = PORTE.1
#bit RW = PORTE.2
#bit DR = PORTC.2
#define isSPI() bit_test(Buff[1],7)
#define Finalize() (!bit_test(Buff[1],6))
#define isREAD() bit_test(Buff[2],7)
///////////////////////////////////////////////////////////////////////////////
void SetCSR(int16 CSRData)
{
PORTD = CSRData & 0xFF;
PORTA = CSRData >> 8;
}
///////////////////////////////////////////////////////////////////////////////
#use delay(clock=48000000)
///////////////////////////////////////////////////////////////////////////////
void main()
{
unsigned int8 Buff[132];
unsigned int8 BuffPointer=0;
unsigned int8 DataLen=0;
unsigned int8 i;
PORTA = 0x00;
PORTB = 0x00;
PORTC = 0x00;
PORTD = 0x00;
PORTE = 0x01;
set_tris_a(0x00);
set_tris_b(0x00);
set_tris_c(0xFE);
set_tris_d(0x00);
set_tris_e(0x00);
setup_adc_ports(NO_ANALOGS|VSS_VDD);
setup_adc(ADC_OFF);
setup_spi(FALSE);
setup_wdt(WDT_OFF);
setup_timer_0(RTCC_INTERNAL);
setup_timer_1(T1_DISABLED);
setup_timer_2(T2_DISABLED,0,1);
setup_timer_3(T3_DISABLED|T3_DIV_BY_1);
setup_comparator(NC_NC_NC_NC);
setup_vref(FALSE);
setup_low_volt_detect(FALSE);
setup_oscillator(False);
delay_ms(1000);
//////////////////////////////////////////////////////////////
usb_cdc_init();
usb_init();
while(!usb_cdc_connected()) {}
while (TRUE)
{
usb_task();
if (usb_enumerated())
{
Buff[BuffPointer++] = usb_cdc_getc();
if (BuffPointer == 3) // check for data length
DataLen = Buff[BuffPointer-1] & 0x7F;
if ((BuffPointer == DataLen + 3) || ((BuffPointer == 3) && (isREAD())))
{ //end of buffer?
switch (Buff[0])
{
case 0xFF: // Data <-->
if (isREAD()) //read
{
set_tris_b(0xFF); // input
RW = 1;
if (isSPI())// SPI
{
SetCSR(0x100 | (Buff[1] & 0x3F));
SS = 0;
for (i=0;i<DataLen;i++)
{
SG = 1;
delay_us(3);
SG = 0;
while(!DR); //wait for DR
usb_cdc_putc(GetData()); // send data to pc
}
SS = 1;
}
else // I2C
{
SetCSR(Buff[1] & 0x7F);
SS = 0;
while(!DR); //wait for DR
for (i=0;i<DataLen;i++)
{
SG = 1;
delay_us(3);
SG = 0;
while(!DR); //wait for DR
usb_cdc_putc(GetData()); // send data to pc
}
SS = 1;
}
}
else //write
{
set_tris_b(0x00); // output
RW = 0;
if (isSPI())// SPI
{
SetCSR(0x100 | (Buff[1] & 0x3F));
SS = 0;
for (i=0;i<DataLen;i++)
{
//SetData(Buff[3+i]);
PORTB = Buff[3+i]; // PROBLEM in RB0 ///////////////////////////////
SG = 1;
delay_us(3);
SG = 0;
while(!DR); //wait for DR
}
if (Finalize())
SS = 1;
}
else // I2C
{
SetCSR(Buff[1] & 0x7F);
SS = 0;
while(!DR); //wait for DR
for (i=0;i<DataLen;i++)
{
//SetData(Buff[3+i]);
PORTB = Buff[3+i]; // PROBLEM in RB0 ///////////////////////////////
SG = 1;
delay_us(3);
SG = 0;
while(!DR); //wait for DR
}
SS = 1;
}
}
break;
case 0xAA: // Command (reserved)
break;
}
BuffPointer = 0; // reset pointer
DataLen = 0; // clear data length
}
}
}
}
|
Everything works poerfect except for the lines commented with "// PROBLEM in RB0 ///////////////////////////////"
Last edited by PICoHolic on Thu Jul 05, 2007 1:29 am; edited 1 time in total |
|
|
PCM programmer
Joined: 06 Sep 2003
Posts: 21708
|
|
| Posted: Thu Jul 05, 2007 12:44 am |
|
|
| I wouldn't call that a short test program. |
|
|
PICoHolic
Joined: 04 Jan 2005
Posts: 224
|
|
| Posted: Thu Jul 05, 2007 12:53 am |
|
|
Dont worry about all the code.
Just the commented one.
10x |
|
|
Ttelmah
Guest
|
|
| Posted: Thu Jul 05, 2007 3:15 am |
|
|
I would add a couple of things:
CCP2C1
Otherwise CCP2 is multiplexed onto portB
Also note that the USB peripheral is multiplexed onto RB2 & 3, while RB4 is the SPP CS pin. I'd probably explicitly clear SPPEN to ensure this is not causing a problem.
Also, if you want to control the TRIS yourself, and write directly to the port, use #use fast_io(B) otherwise the compiler will still think that _it_ has control of the port direction register, and may override what you are doing.
As a comment, why not use the output_b instruction?.
Best Wishes |
|
|
PICoHolic
Joined: 04 Jan 2005
Posts: 224
|
|
| Posted: Thu Jul 05, 2007 3:29 am |
|
|
1- CCP2C1 is added by default:
| Code: |
Configuration Fuses:
Word 1: 0E24 NOIESO NOFCMEN HSPLL PLL5 CPUDIV1 USBDIV
Word 2: 1E38 NOBROWNOUT NOWDT BORV20 PUT WDT32768 VREGEN
Word 3: 8100 NOPBADEN CCP2C1 MCLR NOLPT1OSC RESERVED
Word 4: 00A1 STVREN NODEBUG NOLVP NOXINST ICPRT
Word 5: C00F NOPROTECT NOCPD NOCPB
Word 6: E00F NOWRT NOWRTD NOWRTC NOWRTB
Word 7: 400F NOEBTR NOEBTRB
|
2- SSPEN is cleared:
| Code: |
.................... setup_spi(FALSE);
10CC: BCF FC6.5 <---
10CE: BCF F94.7
10D0: BSF F93.0
10D2: BCF F93.1
10D4: MOVLW 00
10D6: MOVWF FC6
10D8: MOVWF FC7
|
I still have to try the output_b function ..
10x anyway |
|
|
Ttelmah
Guest
|
|
| Posted: Thu Jul 05, 2007 4:04 am |
|
|
SPPEN, not SSPEN
The latter is the SPI connection. The fmer is the synchronous parallel port CS.
Best Wishes |
|
|
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
