Post by thread:[PIC]pic16f737 and max7301ANI problems

Hello, I'm trying to connect a max7301ANI port expander up to the pic16F737 as a part of my senior project for college. So far everything has gone so smoothely up to this point but I can't figure out why the 7301 doesn't seem to want to talk to my uC. Here is how i Have it connected: 16F737 Max7301 ----------- ---------- | | | | | RC5|-----------------------------|Din | | RB6|-----------------------------|CS | | SCK|-----------------------------|SCK | | | | | | | | | | | | | ----------- ---------- I also have the two grounds on the 7301 connected together. They are connected to the ISET pin through a 40 ohm resistor and to the power pin through a 47 nF cap as is reccomended on the data sheet. I then have pins P16-P24 hooked to LED's through resistors to ground so that I can see the them turn on when I successfully change some of the pins to outputs and try to write values to them. I have not connected Dout on the 7301 because write now I am only concerned with getting data to write to the 7301 and not read from it. I have slowed down the chip as much as possible and when I connect an led to my chip select line I can see it blink. I also hooked up another LED to another port that is initially on but turns off when the writes to the 7301 are complete so that I could be sure I wasn't getting stuck in an infinite loop when I was transmitting data and waiting for the BF flag to get set. It seems to transmit the data fine. My Code: I'm just pasting the code relevant to the SPI because I have a lot of other non-relevant code that seems to work fine ;set P16-P19 to active high outputs banksel PORTB bcf PORTB, 6 movlw 0x0C ;address byte call spi_transmit movlw B'01010101' ;set the 4 ports to outputs active high call spi_transmit banksel PORTB bsf PORTB, 6 ; set P20-P23 to active high outputs banksel PORTB bcf PORTB, 6 movlw 0x0D ; address bytes call spi_transmit movlw B'01010101' ; set the 4 ports to outputs active high call spi_transmit banksel PORTB bsf PORTB, 6 ;Put some data on P16-P23 and watch some leds turn on banksel PORTB bcf PORTB, 6 movlw B'01010000' ; address byte call spi_transmit movlw B'10101010' ; write data to the desired ports call spi_transmit banksel PORTB bsf PORTB, 6 ; spi transmit function spi_transmit: global spi_transmit banksel SSPBUF movwf SSPBUF banksel SSPSTAT wait_for_transmit btfss SSPSTAT, BF goto wait_for_transmit banksel SSPBUF movf SSPBUF, 0 return ;I call this funciton earlier on ;---------------------------------------------------------------------------- ;initialize the SPI hardware spi_init: global spi_init banksel TRISB bcf TRISB, 6 banksel PORTB bsf PORTB, 6 banksel TRISC bcf TRISC, 3 bcf TRISC, 5 bsf TRISC, 4 banksel SSPSTAT bcf SSPSTAT, SMP ;input data is sampled at the end of data output bsf SSPSTAT, CKE ;transmit occurs on transition from active to idel clock state banksel SSPCON bsf SSPCON, CKP ;idle state fo sck is high bcf SSPCON, 3 ;set sck to fosc/4 bcf SSPCON, 2 bcf SSPCON, 1 bcf SSPCON, 0 bsf SSPCON, SSPEN ;enable spi return Also, when Im testing my code it gets stuck in the loop where I poll for the BF flag. Is there a way to tell the debugger I want to go out of the loop? I appreciate everyone's time a lot. -Marc Orr student @oregon state university

On 2/21/06, spam_OUTorrmTakeThisOuTonid.orst.edu <.....orrmKILLspam@spam@onid.orst.edu> wrote: > > > Also, when Im testing my code it gets stuck in the loop where I poll for > the BF > flag. Is there a way to tell the debugger I want to go out of the loop? > The only way you would be able to get out of that loop in the simulator (that is the debugger that you're referring to?) is by stimulating the register that contains the BF flag and actually manually setting the flag ( to a logical 1 in your case!). This can be achieved by 'editing' the value in the register after stepping into the loop. You can edit the registers value in the watch-window ( View --> Watch), or any other window that enables you to see what's momentarily in that particular register. Also: where you use: movf SSPBUF, 0 instead of the '0', you can use 'W', and instead of '1', you can use 'F'. It's a little more clear that way! I hope that I haven't been too vague? Sean, a fellow student from Europe.

I finally got it working. I guess you need to disable the "shutdown state" of the max7301 before you can write to any of the pins as outputs. I somehow kept reading over this in the data sheet. I do still have one issue I never could resolve. When I finally got access to an oscilloscope and looked at the signals I realized that my chip select wasn't bouncing. At the time I was using pin B6 as the chip select pin (remember this is on the PIC16F737). I eventually changed the chip select to be pin C0. I went back and tried to make the chip select pin B6 and it still never would work. The only thing special about that pin is it is used for programming the chip but I don't see how that would distract it from being a general I/O. -Marc Orr Quoting James Humes <james.humesKILLspamgmail.com>: > Hey Marc, > > Have you looked at your SDO line and SCK on a scope (together) to make > sure that data is being passed on the right edge and that you are sending > what you think you're sending? I've used the I2C version of this chip and > initially I wasn't pausing long enough between commands sent to the io > expander which was causing some havoc. > > James > -