Looking for Atmega with more then 10Bit ADC

[narindia16]

Hello Friends,

I am working with atmega32 and using its internal 10bit adc for reading 4-20mA signal. But my desired value to be read is of 6 digits so i am not able to read it with 10bit ADC. Kindly suggess any AVR controller with above bit ADC.

Regards,
Narinder

 

[nandhu015]

As far as i know there is no avr part with built in adc above 10 bits. If you need more resolution better use external adc with I2c or any other input with 12 bit or 16bit.

 

[Tahmid]

Hi,
There's the Atxmega which has 12-bit ADC, but not Atmega (as far as I know).

Atmel Corporation - Atmel AVR 8- and 32-bit - AVR XMEGA

Hope this helps.
Tahmid.

 

[narindia16]

Thanks nandhu for your post. I am already getting same response from some other sources too. but my problem is i am willing to get higher value from atmega32 ADC, value higher than 1023. How i can get that.

if any one having piece of code to do this kindly post.

As far as i know there is no avr part with built in adc above 10 bits. If you need more resolution better use external adc with I2c or any other input with 12 bit or 16bit.

 

[nandhu015]

Thanks nandhu for your post. I am already getting same response from some other sources too. but my problem is i am willing to get higher value from atmega32 ADC, value higher than 1023. How i can get that.

if any one having piece of code to do this kindly post.

There is no chance of getting resolution more than 10bits with atmega32. You have to use external adc chip.

 

[narindia16]

that i know i cann't get better resolution. Actually as my single varring from 4-20mA is giving me value in 6 digit, if i still use atmega32, can i use that to display this value????

 

[alexan_e]

Actually you can get a higher resolution using oversampling but the sampling rate will drop accordingly
http://www.atmel.com/dyn/resources/prod_documents/doc8003.pdf

Check the table at page 8

In simple words an oversampling of 4x will give 11bit in 1/4 of the sampling rate:
You take four 10bit samples and you add them to give a 12bit result, then the 11 MSB are your 11 bit result so you just need to shift right one

for 12bit you need 16x oversampling so you add sixteen 10bit results to get a 14bit result and then your result are the 12MSB, again shift right 2 to get them as LSB, the sampling rate will be 1/16

The source that you measure shouldn't change faster that your sampling rate.

Alex