How to get/determine internal VDD voltage?

Guest · Posted: Mon Feb 09, 2009 8:36 pm

Hi everyone,

I'm working on a project and I need to determine the exact value of the VDD voltage of the PIC MCU without measuring through a Digital Multi-Meter. I need the exact value of the VDD voltage since it affects the computed temperature value of the temperature sensor that I am using.

Here is the working code that I am using which converts Analog voltage from the temperature sensor to a DC level signal for the PIC MCU. In my code, I assumed the value to have 3.28 volts.

FvM · Joined: 27 Aug 2008 Posts: 2337 Location: Germany

Can't work without a known reference voltage.

Guest · Posted: Tue Feb 10, 2009 2:47 am

Hi FvM,

"Can't work without a known reference voltage."

--> does this mean that it is not possible to determine the VDD value without using any reference voltage in the code?

Thank you! Smile

Ttelmah · Guest

At heart, yes.

Basically, there has to be something to compare with.

There are several trickish ways of finding your Vdd, but all will involve some form of 'reference'.

For instance, if you use the ADC, setup to use 'Vdd', as it's 'reference', to read an external voltage that comes from a source, whose voltage you _know_ (so the source is the real reference), you can then calculate Vdd.

Realistically, the best way for your hardware to work, would be to attach a known reference to the ADC, and do all your calculations from this. However,if you need to use the full supply range for your temperature sensor (some sensors use the full supply range), then the 'answer' is to attach a known voltage to another pin of the ADC, and use this to measure Vdd. For this, the 'point' is that if you have (for instance), a voltage of 2.5v, fed into AN2, and you read it, getting a value of 'X', using Vdd as the reference for the ADC, then Vdd, is:

(2.5/X)*1024

Best Wishes

Guest · Posted: Tue Feb 10, 2009 9:43 pm

Ttelmah

I think i would just use the hard way on measuring the VDD voltage using a Digital Multi-Meter then hard-code the measured value or just stick using '3.28' as the VDD reference.

Thank you for your time answering! Smile

Ttelmah · Guest

The problem is that a supply from a typical regulator, will change significantly with temperature, and often with time, for example, so the overall stability of readings your system will then be rather poor. Using a better reference, either as a way of 'testing' the supply every few minutes, or (better) as the Vref for the ADC, will give much more accurate long term readings.
You "pays your money".....

Best Wishes