How to set TA1CCR0 in msp430f5438a

Your post doesn't really say what you are using TA1 for. Are we to assume that it is used for triggering an ADC conversion cycle?

If so, then the larger the number you use for TA1CCR0, the longer time between samples, and the better the chance of aliasing occurring.

I would recommend looking at the User's Guide (SLAU208) and making sure you understand how many clock cycles a single ADC conversion takes with all of your settings. Once you know that, then you can determine what rate you can actually sample at and compute the value for TA1CCR0 accordingly.

BTW.... to clarify for you, setting the ADC sample/hold time and other parameters determine the conversion time, not the sampling rate. The sampling rate is the inverse time between conversion starts.

Regarding sampling rate, can you please tell in bit more detail how to calculate sampling rate and on what all factors is it dependent on?

Also in the link below , a query was put up regarding sampling rate and I am doing same calculation as this guy did for his case

http://e2e.ti.com/support/microcontrollers/msp430/f/166/t/41263.aspx

 I am attaching the code I am using. Please if you could have a look at it and tell me where am  I wrong.

////////////////////////////////////////////////////////////////////////////////////////////////////////////////

char buffer[100];

static uint8_t SavedADC12MEM1; //To store ADC outputs

int16_t x_new,y_new;
float values;
float factor=0.000714285714; int volt;

/*************************************************************************************/
#pragma vector=TIMER1_A0_VECTOR
__interrupt void TIMER1_A0_ISR(void)
{
P1OUT ^= 0x01; // Toggle P1.0
ADC12CTL0 |= ADC12ENC; // Enable conversions
ADC12CTL0 |= ADC12SC; // Start conversion
}
/*---------------------------------------------------------------------------*/
#pragma vector=ADC12_VECTOR
__interrupt void ADC12_ISR(void)
{
ADC12CTL0 &= ~ADC12ENC; // Disable ADC12 conversions
REFCTL0 &= ~REFON; // Disable reference (if enabled)

SavedADC12MEM1 = ADC12MEM1; // Store the sampled data

__bic_SR_register_on_exit(LPM0_bits);
}

/*---------------------------------------------------------------------------*/
void accelerometerInit(void)
{
P7SEL |= BIT5; // Pin 7.6 and 7.7 are inputs from the accelerometer to the ADC => they are configured as inputs to the ADC
P7DIR &= ~BIT5;

UCSCTL8 |= MODOSCREQEN; // ADC is clocked by the ACLK in the UCS module
ADC12CTL0 &= ~ADC12ENC; // Disable conversion
ADC12CTL0 = ADC12ON + ADC12SHT0_6 + ADC12MSC; // Turn on ADC, Sample once every 1024 cycles, Multisample conversion
ADC12CTL1 = ADC12CSTARTADD_1 + ADC12SHP + ADC12CONSEQ_2 + ADC12SSEL_3;
ADC12CTL2 = ADC12RES_2; // Resolution=12 bits, 13 clock cycle conversion time

ADC12MCTL1 = ADC12INCH_13; //select channel 13(pin 7.5)

ADC12IE = 0x00;

__delay_cycles(200000); // Allowing reference voltage to stabilize
}
/*---------------------------------------------------------------------------*/
void accelerometerReadWithOffset(uint8_t *x)
{
*x = SavedADC12MEM1;

}
/*---------------------------------------------------------------------------*/
void adcStartRead(void)
{
ADC12IFG = 0x00;

ADC12IE = BIT1;
ADC12CTL0 |= ADC12ENC | ADC12SC;
// Enable conversion and Start conversion
}

/**********************************************************************************/
int main(void)
{
WDTCTL = WDTPW + WDTHOLD; // Stop WDT
// initialize Timer_A module
TA1CCTL0 = CCIE; // CCR0 interrupt enabled
TA1CCR0 =(52000);

TA1CTL = TASSEL__SMCLK + MC__UP + TACLR; // ACLK, up mode, clear TAR
//float voltages;

while(1)
{
accelerometerInit();
adcStartRead();
//accelerometerReadWithOffset(&x_new); // ADC output (digital data

//voltages=x_new*factor;


// initialize USCI module
P5SEL |= BIT6 + BIT7; // P5.6,7 = USCI_A1 TXD/RXD
UCA1CTL1 |= UCSWRST; // **Put state machine in reset**
UCA1CTL1 |= UCSSEL__SMCLK;
UCA1BR0 =9; //
UCA1BR1 = 0; //

UCA1MCTL = UCBRS_1;

UCA1CTL1 &= ~UCSWRST; // **Initialize USCI state machine**

int k=0;

sprintf(buffer,"%d",SavedADC12MEM1);

while(buffer[k])
{
while(!(UCA1IFG&UCTXIFG)); // USART0 TX buffer ready?
UCA1TXBUF = buffer[k++];
}

__bis_SR_register(LPM0_bits+GIE); // Enter LPM0, enable interrupts

}

}

///////////////////////////////////////////////////////////////////////////////////////////////

I did some changes as you told me to do, I sent a sine wave of frequency 10Hz from function generator and I was able to plot the wave on pc after passing it through msp430f5438a. I am even able to see the changes in plot as I change the frequency. My settings are:

ADC sampling rate: 40Hz

UART baud rate: 115200

My query is I am not able to obtain the plot for higher frequencies like for 100Hz , I changed the sampling frequency upto 1.1KHz but still no luck.What can be possible error?

Thanks

hello jeans,

can you please explain what do u mean by 

But let’s assume 7436hz sampling rate, this means you have 7436 results per second. You convert them into decimals with printf. So it can be 4 bytes you transfer for each result (4 digits, no delimiter). Means 40 bit per result. Or 297440Bd minimum UART baudrate or you’ll start skipping results. Your 115200bps aren’t by far not fast enough to transfer all samples. Between two sent samples, you always skip one or two. And depending on input signal waveform, this gives distortion or aliasing effects

thank u

It means what I wrote. You get a 12 bit binary value form the ADC. Then you convert it into a human-readable chain of digits with sprintf. Which takes up to four digits per value. This means, you’ll have send 4*10 bit through UART. So the UART baudrate needs to be at least 40 times higher than the sampling frequency – which you specified with 7436Hz. 7436*40 is 297440, which is your minimum UART baudrate then.
Now if you sample slower, of course you need a lower baudrate.
If you send binary data, you only need two bytes per sample. If you pack the data, you only need three bytes per two samples. Which will reduce the required baudrate to 1/2 or even 3/8 of this value. But then the data is not human-readable anymore.