But I did not have this problem with previous series of PICs. Is there something different about the dsPICs and PIC32s which requires extra filtering on the AVdd pins? I have used the 10-bit ADC peripheral on 18F series PICs with AVdd pins connected directly to the power supply and obtained results which were consistently within 1 LSB of the correct value. Sean On Wed, Nov 23, 2011 at 11:08 AM, Isaac Marino Bavaresco wrote: > Em 23/11/2011 13:02, Electron escreveu: >> At 12.32 2011.11.23, you wrote: >>> Em 23/11/2011 09:03, Electron escreveu: >>>> Hello, >>>> on the PIC32MX1XX data sheet they tell me to give current to AVdd >>> and its bypass >>>> cap through a 10 ohm resistor (connected to Vdd). It's a LP RC >>> filter of course. >>>> Now, as the power supply noise will be high, and I need the best >>> possible precision >>>> from the analog part of the MPU, I decided to put a ferrite bead, too. >>>> >>>> My question is: since I put a ferrite bead, is the 10 ohm resistor >>> still needed, >>>> or does it become ~totally redundant? >>>> >>>> Thanks (and it's not to save 0.07 EUR on a resistor x board, believe >>> me, it's to >>>> learn something and make the circuit without redundant components, >>> that the space >>>> is of primary importance and is already a problem (may become too >>> big and force >>>> me to leave something out, it's not the case yet, though)). >>>> >>>> Cheers, >>>> Mario >>> >>> What I do is to put an additional 10uF ceramic capacitor between AVDD >>> and AVSS, besides the 100nF recommend by Microchip. >>> That reduces the cutoff frequency of the filter by a factor of 100. >>> >>> But don't worry too much about AVDD, Microchip's recommendation should >>> be enough. More important are the VRef pins. Feed them with really >>> stable and filtered voltages. >> I'm not using them (i.e. internal ref is analog vdd) so I guess AVdd >> must receive the same care You'd put in VRef....? > > > So you must tolerate the consequences of your choice. VRef pins are > there exactly because of that. They don't draw any substantial current > and don't introduce noise and spikes in the signals and are intended to > be compared with high precision with the input values to yield precise > results. > > AVdd also supplies some digital circuitry that introduce switching > noise. There's no chance of you getting the same results that you would > get with the VRef pins. > > > Isaac > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .