Is there a difference in RF noise between internal vs external resonators?

wapo54001

Senior Member
I'm finding it worthwhile to move from a 20X2 to a 28X2 chip but I want to keep my 32MHz clock frequency. Since the project is audio-related I'd like to know if I'm taking a step up in radiated RF noise if I start using an external resonator as required by the 28X2 for 32MHz operation.

If there is, I might consider trying to use the 16MHz internal clock instead. Would like to know if anyone has studied this question or has information that would be useful to me on the subject. Thanks.
 

hippy

Technical Support
Staff member
I would imagine there would be an increase simply because the drive signals are exposed to the pins when they would otherwise not be.

How much of an increase and whether it would be detrimental is hard to say. Many audio systems include crystal or resonator controlled chips within them so it seems any potential problems can be overcome.
 

Goeytex

Senior Member
MCU development boards geared towards audio, commonly use external crystals to no ill effect. You may get lots of theory & opinions but I suggest you just try it and see what happens. Unless you have a network or spectrum analyzer there is no way see RF noise, harmonics, spectral scatter ,etc.

Personally, I think you are worried about nothing of consequence.

EDIT: On the other hand ..... It is probably a good idea to use a ground plane on the PCB and not to route audio traces near the crystal.
 
Last edited:

wapo54001

Senior Member
Thank you, it's good to have some reassurance. My board has three separated ground planes (for power, processor, audio sections) to keep current flows separate and connected together at a single location at the regulator output capacitor. The resonator is at the opposite end of the board from the audio components and I'm planning to use .2" or .3" traces between processor pins and resonator pins. I guess I'll be OK and won't worry about it further.
 

John West

Senior Member
That sounds just fine, wapo. Fortunately, the crystals are operating with such low currents that good bd layout practices are sufficient to make the RF signals of no concern, and it sounds like, (without seeing the bd layout,) that you've done a good layout.
 

geoff07

Senior Member
32MHz should not make much impact on audio circuits, given good equipment practice as seems to be the case. RF pickup from long wave would worry me more (but not a lot).
 
Top