Please add the AVcc filter

I never get 10-bit A/D performance with Arduino's - there's always too much noise. If I add the (Atmel datasheet) RC +AVcc filter, there is a huge improvement. Can you put these or at least provisions for these on Arduino designs? Even if there is a 0R resistor there, I can add it myself.

I totally agree. This filter is mandatory by Atmel, you have just to add an smd inductor of 10mH to satisfy Atmel requirements..

Can you also provide for the implementation of an on-board optional voltage reference.
A dip8 socket will be the best.
Each user will be free to choose the value.
This is not the place that is missing on the board.

Send your requests to The Arduino Team, they are listed on the home page.

I agree, an AVcc filter would be great.

Without the filter, accuracy depends a great deal on how you power the Arduino. I did some tests with a high quality 9V power supply http://arduino.cc/forum/index.php/topic,120004.0.html.

For a perfect ADC, the maximum error is 1/2 LSB and the mean error is 1/4 LSB.

For 5V Vref and a 10-bit perfect ADC the max error is 2.44 mv and the mean error is 1.22 mv.

My result for three Uno Arduinos with the high quality power supply was max error around 4 mv and mean error 1.4 mv.

So you can get very close to 10-bit accuracy with a good grounded 9V power supply and no modifications to a Uno.

The best solution would be a filter. Good 9V supplies cost a lot and good small three prong supplies are hard to find.

ADC calibration is absolutely necessary to achieve good results. The above link describes my calibration setup.

Good 9V supplies cost a lot and good small three prong supplies are hard to find.

I'm a little surprised that the onboard regulator and filters doesn't "improve" even "merely OK" external power supplies. How bad is a generic 9V supply (and how bad is average USB power?)

I've discovered the same thing, if doing analogReads, you need a decent PS, otherwise the noise puts the readings all over the place. Even then, I still have to do quite a bit of filtering/smoothing in sw to get stable and reliable readings.

An unscientific observation is that the old-style linear supplies may be quieter than the newer, more efficient switched-mode supplies. This is from just doing some trial and error with existing supplies I've had laying about. I wish I could identify a good, cheap supply for these applications. Perhaps the inductor is the next thing to try.

A good hardware reference about how it should be a noise immune design is here.

Ok i've mentioned it to the manufacturer, we'll be able to add it to a future batch of board.
I can't tell you when it will happen because it depends on how many are planned to be manufactured in the current batch.

@funlw65 thanks for the advertisement... I hope they are taking good care of you :slight_smile: