another RS232 issue

[sapy44]

Hello all,

I've been experiencing a problem receiving data on the HW RS232 port on the PIC16F877A. Basically I have a do-while loop that continue to receive data from the RS232 port.

The first time throgh the do-while loop works fine. But when I renter the loop the second time the getc() function seems to lock up. I ran the debugger and found that the microcode is stuck in an infinite loop checking for the data ready flag to be set on the Rs232 register.

Do I need to reinitialize the RS232 port for each cycle of the do-while loop ? Any thought's would be appreciated.

I also have a quick question about satnd alone programs. From my understaning all you have to do is remove the "#device ICD=TRUE: statement and the program should compile as a stand alone program. When i do this however my program only seems to run up to the first line of code in my main() function and then locks up. I also noticed that I can still use the ICD to execute the program even though it should have been disabled.

Thanks again for your help.

[Humberto]

sapy44,

Post your code and the whole info needed to give us a chance to help you.

Regards,

Humberto

[sapy44]

Ok, Humberto asked for some example code so here goes.

do {

do {
opr1=getc();
opr2=getc();
} while( !(opr1 == 'M' && opr2 == 'C');
/* loops until detecting "MC" in the incoming data stream */

for(i=0;i<63;i++)
data[i]=getc();
// after detecting "MC" then collect the next 63 characters

vel = atoi(data[23]);


if (vel < 2)
stop_counter++;

if(stop_counter > 10)
done = TRUE;

} while(done); //continues looping until stop_counter reaches 10


The first time throgh the outer do-while loop the getc() functions work fine. But after rentering the outer do-while loop the second time the program hangs on the first getc() inside the nested do-while. A look at the assembly code using the debugger shows that the programm hangs because when it checks the RX data ready ready bit it's not getting set. So it just it hangs because it cannot detect that bit getting set with the incomming data.


Do I need to reinitialize the RS232 port beteween loops ? If not, then what could be the problem that causes the getc() to hang the second time through the loop? Any suggestions welcome. Thanks.

[Humberto]

sapy44 wrote:

[SherpaDoug]

do {
do {
opr1=getc();
opr2=getc();
} while( !(opr1 == 'M' && opr2 == 'C');
/* loops until detecting "MC" in the incoming data stream */

This seems like a very fragile way to detect the start of your message. If you miss a single character in the serial input you may never recover. Instead try:

do{
do{
if (kbhit()){
if (getch() == 'M'){
if (getch() == 'C'){
<Do Stuff>
}
}
}

This way a lost character or a single extra character due to noise will caus a loss of one or two records, but you will eventually be able to rcover.
_________________
The search for better is endless. Instead simply find very good and get the job done.

[sapy44]

I'd like to as some more detail as to the operation of the RS232 port capture that I'm attempting. The data comming in over the RS232 is continous and periodic. Even if I don't catch the first "RM" it will repeat again in less than a second so I'm sure to catch it at some point.

I don realize now that I do need to content with hardware overflow issues, especially since data is alaways comming on over the RS232 port. Humberto's idea of using interrputs seems like a good solution. However, is the isr_rda interrupt routine defined for just the hardware UART or does it also apply to a software RS232 port ?

Also, if anyone has some idea as to why I can't get a program to run properly on it's on w/o having to use the ICD to run it please let me know. Thanks again.

[Humberto]

sapy44 wrote: