rfPIC Serin problem

[onepiece]

I think I have the same problems with gavo. But currently I'm using assembly to do my programming. This rfPIC12f675 is different from other normal PICs where it has only one tris and gpio port. And the gpio reg is assigned to the pin itself.
Can you kindly explain to me about how can I read in data serially using assembly? I'm sorry that I'm not really familiar with the code you pasted before.
Can I just read in data bit by bit say into gp3 and then send it to gp2 for ASK modulation?
Or can I just put or connect my digital input to gp2?

I understand that the coding below is a Keeloq protocol

' Constants
PreAmble CON $A5 ' 10100101 preamble
Synch CON "~" ' 01111110 synch byte
N4800 CON 188 ' 4800 bps
N2400 CON 16780 ' 2400 bps
GUARD CON 5 ' 5mS guard time pause

As I'm not familiar with your coding, may I know what CON indicates and what N4800 CON 188 means?


Thanks again for your help!

[mister_e]

in PBP, CON is used to define a software constant, in asm it should looks like

YourConstantVariable = .25

or
#DEFINE YourConstantVariable .25

Using pre-defined constant here and there, allow to have a better readable code, and also allow to do some changes without having to scroll the whole code.

For a 12F675 asm example.. well, first google hit gave me
http://www.jbgizmo.com/page29.htm

[medusa2584]

hi people,

has anyone ever encountered a problem of stack overflow in their program?
rite now im doing a program to transmit a stack of data which contain 16 bit + 64 bit +

and the pic that im using is a pic12f675. what should i do to reduce this error?

[BobK]

Hi Medusa,

I had this problem once and it turned out to be a subroutine that had a GOTO at the end of it instead of RETURN. Changed the word--solved the problem.

HTH,

BobK

[medusa2584]

thanks bobk..i tried that, put return at the end of each goto instruction i have, but still it has not succeeded. is it possible due to overflow stack?

my program is supposed to have a stack of 8 byte output excluding the preamble and header and timeguard.

does anyone ever write a program that has a subroutine as follows

read input from port GPIO
move data from GPIO to 'input' register
take only data from bit 1
store the input (bit 1)to bit 0 at another register

repeat for eight times

what i already have in mind

movfw GPIO
movwf input
movlw 0000 0000
xorwf input,f
goto rotate

rotate
rrf input,f
rrf input,f

anyone has any idea to continue?

[BobK]

Hi Medusa,

That's not what I meant. If you are calling a subroutine then at the end of the subroutine you put a RETURN. GOTO does not go at the end of a subroutine. This was the problem I had with one of my projects (18F452). The stack overflow was causing my project to hangup when it got to the subroutine after the stack overflowed. This was the only point I was trying to make.

If you have both a GOTO and a RETURN then you might still be experiencing a problem.

HTH,

BobK

[hell_pk]

Hi Bruce!
I have tried ur codes on RfPIC kit. i can see the data being sent by transmitter on GP2 and received on PORTC.1 on receiver side but the LED's are not blinking. what might be the problem.plz help me troubleshoot.
thanks

[Bruce]

That's hard to say. It works perfectly on my rfRXD0420 module installed in PICkit 1 board.

Are you compiling the receiver code without any changes, and programming it into a
PIC16F676 on the PICKit 1 board with the rfRXD0420 installed per the instructions?

[hell_pk]

Hi
U have not included modedefs.bas in ur program.is that an issue?
I did that but to no avail.
maybe config fuses are causing something.they seem ok.they are as under when i read them using PICkit 1 software:
12F675
config word:0x31D4
osccal: 0x3480
bandgap: 0x3000

12F676
config word:0x31C4
osccal: 0x3440
bandgap: 0x3000

[onepiece]

Hi there,

I've altered the receiver_demo program to suit my application. - I took in the data from rfRxd0420 drom RC2 and try to take it out serially from RC0 of the pic16f676 so that I can
probe the data transmitted.

Currently, I'm simulating the program using MPLAB IDE. Pardon me if I'm wrong.
The receive_demo program that I understand, will check for header. But,
it does says that if the data is high before 6ms, therefore the header is valid.
What if the data is also 'LOW'?
I'm actually having problems with simulation coz it does not want to
come out of the loop - unless I put '1' in between.

When it came out of the loop, I should be able to record that and then output that
to the RC0 but it seems that the program didn't even jump to search for a guard time,
instead it continue the process to find for high or low pulse.

I just altered the receive demo of the dev kit. Can anyone help me please?
Or do you suggest a better way to simulate this program?

MAIN
call CLOCK ; to update TRMLOW and TMRHIGH
; TMRLOW ~=4us TMRHIGH ~=1ms

movlw HIGH STATEM
movwf PCLATH
movf STATECNTR, W ; Mask out the high order bits of
andlw B'00001111' ; STATECNTR (a noise guard)

addwf PCL, F ; The program clock (PCL) is incre-
STATEM
; check for possible start of data stream
goto BEGIN ; mented by STATECNTR in order
goto BEGIN1 ; to go to the appropiate routine
goto HEADER ; header is being ignored
goto HEADER1 ; just count up the time and then takes the data
goto HIGHPLSE ; times the width of Hi Pulse
goto LOWPULSE
goto RECORD ; record the 8 byte data
goto WAIT4END ; wait for guard time to be greater than 8ms

goto FIRST0 ; check whether DATA0 of bit 0 is 0 or 1
call SENDOUT ; to send out the serial data out through portc pin 0
goto RESET ; These RESET commands correct
goto RESET ; erroneous values of STATECNTR
goto RESET ; not caught by the mask above.
goto RESET
goto RESET
goto RESET

[medusa2584]

from the demo program, they used ASK modulation which use NRZ format modulation.
in this format, when the input is zero it would give you a pulse of 110 while if the input is one, it would give pulse 011, which means it will still give you pulse high for whichever input.