View previous topic :: View next topic |
Author |
Message |
paddy
Joined: 04 Nov 2012 Posts: 15
|
MCP3204 ADC output fluctuation |
Posted: Tue Nov 24, 2015 6:58 am |
|
|
Hi,
Greetings of the day!!
I am using MCP3204 ADC interfaced with PIC18F4520. The reference voltage of MCP3204 is from voltage divider network of resistor and set to 500mV. Potentiometer value is 1K and resistor of 1.2K.
Input signal range is 0-500mV. I have used mcp3204.c driver file in code.
Even though the input signal and reference signal is stable, there is fluctuation in output of MCP3204. The fluctuation is approximately +/- 30 counts than expected.
I do not have oscilloscope to check the noise, but I have measured voltages and that are stable.
Please find attached schematic and suggest.
Thanks,
Paddy
https://www.dropbox.com/s/f4d3jcegzanmisl/Test.pdf?dl=0[/url] |
|
|
Ttelmah
Joined: 11 Mar 2010 Posts: 19513
|
|
Posted: Tue Nov 24, 2015 8:36 am |
|
|
Noise.
If you are using a standard DVM, this uses _integration_ to measure the voltage (hence 'dual slope integrating', in the 'title' of most simple DVMs). This gives good accuracy at low cost, but will completely miss noise signals that an ADC _will_ record.....
Unless you are building the design _really_ carefully, with well designed ground planes and a super stable reference (what is your reference voltage divided 'from'?. If this is the PIC supply, you can pretty much guarantee that this will have 10's of mV of noise already on it), the circuit will record large amounts of noise. |
|
|
asmboy
Joined: 20 Nov 2007 Posts: 2128 Location: albany ny
|
|
Posted: Tue Nov 24, 2015 10:45 am |
|
|
two things:
put a 100nF capacitor from the wiper of the ref-cal pot to ground.
metal film is ideal but even a ceramic cap will do.
this lowers the reference input impedance in AC terms and has been shown to improve noise .
this post may also help you :
http://www.ccsinfo.com/forum/viewtopic.php?t=50320&highlight=olympic
it is very easy to adapt...... |
|
|
|