how read the battery voltage

[ilpaso 0]

hi,

I've a msp430g2253 board powered by a battery. I'd like to read the voltage of the battery.

I've found a function online (http://blog.elevendroids.com/2013/06/code-recipe-reading-msp430-power-supply-voltage-level/).

uint16_t Msp430_GetSupplyVoltage(void)
{
	uint16_t raw_value;
	// first attempt - measure Vcc/2 with 1.5V reference (Vcc < 3V )
	ADC10CTL0 = SREF_1 | REFON | ADC10SHT_2 | ADC10SR | ADC10ON;
	ADC10CTL1 = INCH_11 | SHS_0 | ADC10DIV_0 | ADC10SSEL_0;
	// start conversion and wait for it
	ADC10CTL0 |= ENC | ADC10SC;
	while (ADC10CTL1 & ADC10BUSY) ;
	// stop conversion and turn off ADC
	ADC10CTL0 &= ~ENC;
	ADC10CTL0 &= ~(ADC10IFG | ADC10ON | REFON);
	raw_value = ADC10MEM;
	// check for overflow
	if (raw_value == 0x3ff) {
		// switch range - use 2.5V reference (Vcc >= 3V)
		ADC10CTL0 = SREF_1 | REF2_5V | REFON | ADC10SHT_2 | ADC10SR | ADC10ON;
		// start conversion and wait for it
		ADC10CTL0 |= ENC | ADC10SC;
		while (ADC10CTL1 & ADC10BUSY) ;
		raw_value = ADC10MEM;
		// end conversion and turn off ADC
		ADC10CTL0 &= ~ENC;
		ADC10CTL0 &= ~(ADC10IFG | ADC10ON | REFON);
		// convert value to mV
		return ((uint32_t)raw_value * 5000) / 1024;
	} else
		return ((uint32_t)raw_value * 3000) / 1024;
}

It works but the first time I call the function it uses the 1.5V reference (Vcc < 3V ) and the second time it uses 2.5V reference (Vcc >= 3V).

The real voltage is 3,12V

What is wrong?

Thank you in advance.

 

Regards

 

ilpaso

 

[semicolo 39]

From the comments, it seems you should use a voltage divider (for example maybe 2 100K resistors), so you have VCC/2 on your analog input.

The first try is using the 1.5V internal reference, so it's good for voltage below 3V.

If the conversion returns 0x3FF, it means that VCC is >= 3V, so the software switches to the 2.5V reference that allows up to 5V for VCC (but of course you can't go higher than 4V on these chips).

[spirilis 1,265]

Shouldn't need a voltage divider, there is an on-demand one built in to the ADC (so power isn't wasted when you're not sampling). Will have to try this code later to see what's wrong.

 

Sent from my Galaxy Note II with Tapatalk 4

 

 

[spirilis 1,265]

Try getting rid of ADC10SR. You are running the ADC pretty fast and short so it should need the higher buffer strength available when ADC10SR=0 I think.

 

Sent from my Galaxy Note II with Tapatalk 4

 

 

[Register 3]

Hi,

 

Voltage without divider? Is there a way to do the same in energia?

 

Thanks

[semicolo 39]

 

It works but the first time I call the function it uses the 1.5V reference (Vcc < 3V ) and the second time it uses 2.5V reference (Vcc >= 3V).

The real voltage is 3,12V

What is wrong?

So are you asking why it's using the  1.5V reference and then the 2.5V one? But in fact it's working ok?

[semicolo 39]

Voltage without divider? Is there a way to do the same in energia?

Maybe with analogRead (11), this should set ADC10CTL1 correctly and reset ADC10AE0 completely.

Maybe I'll have some time to try it at lunchtime.

 

Maybe there's already a constant for this in energia.

[ilpaso 0]

Thank you for the replies.

It works in Energia but the first time the function uses the 1.5V reference cycle and the second it uses the 2.5V reference cycle and so on.

I'll try  with analogRead (11) this evening.

[semicolo 39]

Ok, then as I explained in my first response, it first tries with 1.5V reference to get a better reading, if it overflows it switches to 2.5V reference.

[semicolo 39]

It seems to be working, the following sketch gives me 3.5V with USB VCC

void setup() 
{ 
  Serial.begin(9600); // msp430g2231 must use 4800

  Serial.println("VCC value:");
  analogReference(INTERNAL2V5);
} 

void loop() 
{ 
  long data = map(analogRead(11), 0, 1023, 0, 500);
  Serial.print (data/100);
  Serial.print (".");
  Serial.println (data%100); // this is wrong if the tens are 0
  delay(1000);
} 

[nemetila 12]

Based on the MSP430G2553 datasheet, the 1.5V ref requires Vcc>= 2.2V and the 2.5V ref requires Vcc >= 2.9V to operate.

 

If your Vcc is below 2.9V you can't use 2.5V ref. So this code first uses the 1.5V ref to check Vcc.

 

[semicolo 39]

Makes sense, it'd be hard to generate a 2.5V reference with a 2.4V VCC.

[Register 3]

Makes sense, it'd be hard to generate a 2.5V reference with a 2.4V VCC.

 

Not if there a voltage regulator embedded?

 

And why pin 11?

[spirilis 1,265]

Not if there a voltage regulator embedded?

 

And why pin 11?

Voltage regulators can only lower voltages...

 

For analogRead the value isn't really the Energia pin# but rather the ADC channel (INCH_xx constant), 11 just happens to be an internal "(Vcc-Vss)/2" generated channel used for measuring the MSP430's Vcc voltage.

 

Sent from my Galaxy Note II with Tapatalk 4

 

 

[semicolo 39]

Well it's possible to generate higher voltages with a dc/dc boost converter for example, I just don't know if it's even possible to embed something like that in a chip without external components and keeping the power usage and costs low.