Analog In 0-5VDC

bstumm · ‎2011-03-04

Posted on March 04, 2011 at 22:46

Analog In 0-5VDC

Tesla DeLorean · ‎2011-05-17

Posted on May 17, 2011 at 14:26

They are limited to whatever +Vref is. And +Vref is limited to Vdda, which in turn has a max of 3.6V, per the manual.

So no 5V isn't going to fly.

Tips, Buy me a coffee, or three.. PayPal Venmo
Up vote any posts that you find helpful, it shows what's working..

raptorhal2 · ‎2011-05-17

Posted on May 17, 2011 at 14:26

The first step is to put an oscilloscope on the input signal. Your multimeter may be doing some filtering before displaying the value.

If that doesn't isolate the problem to noise on the input signal, then read AN2834 on Best ADC Accuracy. It provides requirements for, among other things, reference noise reduction and sample times.

If you are religiously following AN2834, have someone else look at your code. I had a problem with getting a clean signal on the CPU temperature channel until I found out I was intermixing the output from two ADC channels.

Hope this helps.

Cheers, Hal

bstumm · ‎2011-05-17

Posted on May 17, 2011 at 14:26

That is what I was afraid of.

Next question. Could someone explain sample time? Why use ADC_SampleTime_239Cycles5 over ADC_SampleTime_1Cycles5. I'm getting inconsistent values returned on the 0-5V signal (+/-50mV) when the multimeter gives a very consistent value (+/-2mV).

I know I could build a buffer and average values but our current processor technology is very old so just wanting to learn how to take advantage of this new stuff.