Analog GND and digital GND clarification needed

Go To Last Post
3 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

 

Doing a project with SPI and MP3 player using the VS1003B MP3 decoder chip.  I've found a lot of projects on the net that I can try out.  One such project is posted on the VSLI (the maker of the VS1003B decoder) site.  Attached is the schematic.  Also attached is the VS1003B datasheet.

 

As far as I am concerned, the two GNDs (analog and digital) are basically the same GND from the power source?  I have gotten it working on the breadboard, which is very noisy, because I can hear a little tiny buzzy sound in the background.  I figured that once I have the PCB made with the ground plane, things will be better.  However, I would like know if putting a low-pass filter (a resistor and a ceramic cap) on each of the LEFT and RIGHT output pins of the VS1003B will eliminate the buzzing sound.  I am about to try but just wanted to see if there is a better solution.

 

And additional quesitons:

 

-I use the attached SMD 3.5mm earphone jack.  (See the attached footprint image.)  I figured out that pin #2 is for the "common buffer for the headphones."  Pins #3 and #4 are for the LEFT and RIGHT outputs from the VS1003B.  I am not sure about pin #5.  When I try to ground it, the buzzing is louder, and I am not sure leaving pin #5 unconnected (to whatever it's supposed to be connected) is causing the tiny buzzing sound in the background.  Unfortunately, I don't have the entire datasheet except the attached footprints.

 

-This is my 1st time dealing with very sensitive analog stuff.  Wow!  The breadboard is full a noise.  By just moving even one of the shielded wires, the thing won't work.  I have used so many 0.1uF caps.  Besides the ground plane and all analog components to one area and digital components to one area on the board, are there other suggestions that I need to follow to ensure stability in the entire circuit as shown in the attached schematic? 

 

Thank you!

Attachment(s): 

Last Edited: Fri. Jan 2, 2015 - 08:06 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,

both analog and digital grounds are at the same potential. It does not means, indeed, they should be physically single ground. The issue (and the root of evil noise) is that digital ground path  carries return currents that can induce voltages on analog input grounds. Whatever,

A good practice is to split and bypass separately supplies and grounds; both grounds  ideally should have only one common point as close to the power supply as possible. Some designs use direct high current (read digital) ground connection to PSU; sensitive small current ground is connected via small value resistor.

Two side PCB with two ground planes only on one side should be a wise solution.

IMHO earphone socket shall have #2 (common), ##3 and 4 connected. Reference design suggests that. Personally would test #5 with a tester and earphones connected: shall this be shorted to GND - join it with #2. Leave floating otherwise.

Last Edited: Fri. Jan 2, 2015 - 09:48 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kas wrote:

Hi,

both analog and digital grounds are at the same potential. It does not means, indeed, they should be physically single ground. The issue (and the root of evil noise) is that digital ground path  carries return currents that can induce voltages on analog input grounds. Whatever,

A good practice is to split and bypass separately supplies and grounds; both grounds  ideally should have only one common point as close to the power supply as possible. Some designs use direct high current (read digital) ground connection to PSU; sensitive small current ground is connected via small value resistor.

Two side PCB with two ground planes only on one side should be a wise solution.

IMHO earphone socket shall have #2 (common), ##3 and 4 connected. Reference design suggests that. Personally would test #5 with a tester and earphones connected: shall this be shorted to GND - join it with #2. Leave floating otherwise.

 

Ok. Thanks.  This is more than meet the eyes for me.  I guess I could put the VS1003B on the bottom side of the board, including the SMD headphones jack, and all my digital ICs, which all I have is the ATMEGA, on the top side, including the LDO regulators (AMS1117-3.3V and AMS1117-2.5V) and the SD card.  Then I make a separate ground plane for the top and bottom side of the board.  If it requires more than this, I will have to do some research on separate ground planes, even how to do that with EagleCad.  Attached is a document I found on the net.

Attachment(s): 

Last Edited: Sat. Jan 3, 2015 - 02:11 AM