• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 80 subscribers
  • Views 894 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

SPI initialization problems

David Engineer
Intellectual 340 points

Other Parts Discussed in Thread: MSP430F1611

NOTE: I posted this in the wrong forum, this is a repost of the same question.

 

I have been trying to get the code below to work for over a week now with no success. I am building a datalogger with a MSP430F1611 and I am using a MSP-FET430UIF to interface with the controller. I am new to embedded programming and taking an incremental approach to the coding. I first made an LED blink based off of the Timer A interrupt and now I am trying to get the SPI to talk to a 1GB SD card. I have read through the data sheets and online forums and I am still stuck. I run my code and use an oscilloscope to monitor the UCLK pin and see it rise to a steady 3.3V. I know I am new at this but I thought that there is supposed to be a steady pulse emanating from that pin! I wanted to see if it was transmitting and I wasn't using the oscilloscope correctly so I set the LISTEN bit and ran the program. I checked the receive register and there was nothing there.

 

I just want this darn thing to send out a steady clock signal and some kind of data stream, I appreciate any help I can get. 

 

 

#include <msp430f1611.h>

#include <mmc.h>

#include <stdio.h>

#include "fatlib.h"

#include "HALayer.h"

 

// Global Variables 

 

int Working = 0;

unsigned int buffer[512];

 

//Function declarations 

void cpu_init(void);

void spi_init(void);

 

void main(void)

{

  cpu_init();

  spi_init();

 

  while(1)

  {

 

    U1TXBUF = 0xaa; //Dummy value

//  for(int i=0;i<0xFF;i++);                           //Delay for Tx

 

  }

}

 

void cpu_init(void)

{

  WDTCTL = WDTPW + WDTHOLD;      // Stop watchdog timer

  P5SEL |= BIT1+BIT2+BIT3;                   // Set pins P5.1-3 to peripheral

  P5DIR |= BIT0;                                        //P5.0 output

  P5OUT &= ~BIT0;                                  //P5.0 Low

}

 

void spi_init(void)

{

  U1CTL |= SWRST; //disable state machine 

  U1CTL |= CHAR+SYNC+MM+LISTEN;            //8bit char, SPI,USART is master,Tx to Rx

  U1TCTL |= SSEL0+STC;                                     //UCLK=ACLK

  U1MCTL = 0x00;                                                   //No Modulation

  U1BR0 = 0x02;                                                     //Divide ACLK by 2

  U1BR1 = 0x00;

  ME2 |= USPIE1;                                                  //Enable SPI module

  U1CTL &= ~SWRST;                                        //Initialize state machine

}

  • 0
    TI__Expert 6140 points

    David,

     

    Please take a look at the code examples for that device. They can be very helpfull when starting from scratch. There are a about 8 code examples that use the SPI configuration, ie. "fet140_spi0_03.c"

    http://www.ti.com/litv/zip/slac015n

     

    Try running them stand alone and then you can migrate them to your own code.

  • 0 in reply to David Racine
    Intellectual 340 points

    Thanks for the suggestion but I tried all of those on their own with no success. I didn't have anything connected to any pins when I tried it, thinking that if it worked a clock signal would be on the indicated pin. That never happened, I'm beginning to think there may be something wrong with the chip itself. I loaded someone else's program that worked before and that caused a clock signal on the right pin. However, whenever I attempted to write a new program using that same clock pin, the signal did not go away. I could not change it, I looked in memory and the right values were in the right registers to have the signal off, and yet, there it was. 

     

    Is there a way to clearly tell if a controller has gone bad? or if it is only a case of user error?

  • 0
    Guru 227245 points

    David Engineer said:
    I thought that there is supposed to be a steady pulse emanating from that pin!

    No. SPI works one clock pulse for each bit to transfer. No transfer, no clock. And transfers are always bi-directional (one clock pulse = one bit out and one bit in).

    So to get a clock, you'll have to send something. If you want to get something, you'll have to send something. Writing something to the TXBUF register starts sending and receiving and generates exactly 8 clock pulses after which a byte has been sent and a byte has been received (even if there is no slave at all: then 0xff has been received and the seinding went into nirvana)

  • 0 in reply to Jens-Michael Gross
    Intellectual 340 points

    I figured out a significant part of my problem, I did not have my compiler configured correctly. It was using a simulator instead of the JTAG, stupid me. I have the clock working just as Jens-Michael described and now I am trying to get the SD card to talk back. I suppose it is worth mentioning that my code has changed significantly since first posting. Thanks for the help.

**Attention** This is a public forum