10 bit or 8bit AD Converter

[itachi012587]

Hello!

Which is better to use in AD convertion, the full 10 bit resolution or the 8 bit? If 10 bit, why? If 8 bit why? By the way I'm using PIC16F877.

Thank you.

 

[nandhu015]

As you can see 10 bits means more accurate than 8 bit.

Nandhu

 

[amol_shah]

depends upon your requirement, sometimes even 8 bit is more than sufficient
what do you want to measure & what is the minimum voltage that you want to measure

 

[rkodaira]

Sometimes you do not need more precision, and it is easier to handle only one 8 bits register to make calculations (instead of 10 bits divided in two registers) or you have speed restrictions in your processing (i.e. you need a higher sampling rate associated to the processing routine), so you should sacrifice precision to get faster sampling.

 

[ramshankarsingh]

what is your application? it depends on the application whwre you are using adc.if accuracy is important you can go through 10 bit . If not 8- bit is ok. i can give you the asm file too in 8-bit

 

[Faizan Jawaid]

Yes...definitely it all depends upon ur precision...The more no. of bits the more accurate ur reading is...if u wanna just use it for ur hobby project i recommend the 8-bit is enough...

 

[bld]

Hello itachi012587.

What signal do you want to measure?
If we assume the range of your signal to be [0..5V] if you use 8 bit resolution you will have a precision of ~19.5mV (5V/256); if you use 10 bit resolution you will have a precision of ~4.88mV (5V/1024).
So depends on your application and the signal you want to measure, which resolution you may want to choose.

 

[swapgo]

8 bit means the input signal is divided in to 256 parts.
10 bit means the input signal is divided into 1024 parts.

so if you want to deal with more accuracy then go for 10bit.

Gopi

 

[cool.man]

If u are doing this just for experiment then u should check both options 8-bit and 10-bit.this will clear ur mind.

 

[itachi012587]

what is your application? it depends on the application whwre you are using adc.if accuracy is important you can go through 10 bit . If not 8- bit is ok. i can give you the asm file too in 8-bit

I'm using LM35 connected to the analog pin of PIC16F877 then I want to display the temperature up to 100 degree Celcius. By the way I'm waiting for the asm file.

Thank you.

Added after 23 minutes:

8 bit means the input signal is divided in to 256 parts.
10 bit means the input signal is divided into 1024 parts.

so if you want to deal with more accuracy then go for 10bit.

Gopi

Actually I'm quite confuse. You mean, correct me if i'm wrong for example I have a result of 0x3ff and I use 8 bit then I need to devide it by 256 and when I use 10 bit I need to devide the 0x3ff by 1024?

Thank you.

 

[amol_shah]

Well if it’s a 8-bit ADC then you won’t get 0x3ff, since in 8-bit system the maximum value can be 0X0ff i.e. 0-255 which means 256 steps, whereas in 10 bit system the maximum value will be 0x3ff i.e. 0-1023 which means 1024 steps.

 

[jhbbunch]

just remember that the resolution needs to be better than the noise levels and the slop in the components. Op amps can have pretty hight voltage offsets for example.
Filters are usually the most crucial element in ADC design, that is a quiet circuit.

 

[bobsanjose]

Hi, really seems that you are a little confused ;-) If you have a result of 0x3ff you are using 10 bit by definition because the max value you can achieve with 10 bit is exactly 0x3ff. If you are using 8-bit your max possible result is 0xff = 11111111 binary = 8 bits

Actually I'm quite confuse. You mean, correct me if i'm wrong for example I have a result of 0x3ff and I use 8 bit then I need to devide it by 256 and when I use 10 bit I need to devide the 0x3ff by 1024?

Hope this helps clearing 8-bit / 10-bit a little bit

Bob