USB Isolated Ground Plane?

Log in or register to post comments
Go To Last Post
5 posts / 0 new
Author
Message
iwoloschin
iwoloschin's picture
Level: Hangaround
Joined: Wed. May 11, 2011
Posts: 337 View posts
Location: Boston, MA
#1
Posted by iwoloschin: Fri. Oct 21, 2011 - 03:19 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

For a high speed USB 2.0 PCB, I've read that it should have a continuous, isolated ground plane. Since power is provided by USB that's pretty easy to ensure, just connect the ground plane to the ground pin on the USB connector. However, I'm using an LDO to drop the voltage from 5V to 3.3V (to power an SD card and UC3A3). I'm uncertain if the 3.3V ground plane should be connected (at one point) to the High Speed USB ground plane. Here's my reasoning:

Connect ground planes: If the ground planes are unconnected, the 3.3V ground is floating without a defined value. I'm concerned this could cause problems when D+ & D- and VBUS are connected to the UC3A3. Also, I'm not sure how the LDO would react to having a floating ground, I assume it would be ok as it's just going to separate Vout and GND by 3.3V (provided the input is an appropriate voltage/current).

Unconnected ground planes: This is technically what an "isolated" ground plane would be.

Tags:

Learning and Information, General Electronics
Top
Brutte
Brutte's picture
Level: Raving Lunatic
Joined: Thu. Oct 5, 2006
Posts: 3229 View posts
Location: Poland
Posted by Brutte: Sun. Oct 23, 2011 - 12:16 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
For a high speed USB 2.0 PCB, I've read that it should have a continuous, isolated ground plane.

Reference please. [Document, revision, paragraph, page]

No RSTDISBL, no fun!

Top
DocJC
DocJC's picture
Level: 10k+ Postman
Joined: Tue. Dec 11, 2007
Posts: 13933 View posts
Location: Cleveland, OH
Posted by DocJC: Sun. Oct 23, 2011 - 03:13 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Do connect the grounds.

There are options for a truely, totally isolated, USB interface. The generic USB interface isn't "isolated", and the voltage regulator and micro need to be tied to the USB ground reference.

JC

Top
iwoloschin
iwoloschin's picture
Level: Hangaround
Joined: Wed. May 11, 2011
Posts: 337 View posts
Location: Boston, MA
Posted by iwoloschin: Mon. Oct 24, 2011 - 03:15 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Brutte wrote:
Quote:
For a high speed USB 2.0 PCB, I've read that it should have a continuous, isolated ground plane.

Reference please. [Document, revision, paragraph, page]
I found an AVR appnote, AVR32787 - High Speed USB Design Guidelines. I'm using a copy I downloaded from Atmel's website, but I'm sure with the appnote number it'll be easy enough to find. This appnote also references an Intel USB design spec which I do have a URL for:

http://www.usb.org/developers/do...

The actual line in question is from the AVR appnote, Page 2, Section 2.1, Item 7 (second sentence in particular):

Quote:
7. Route all traces over continuous GND plane with no interruptions. USB ground plane
should be isolated from other ground plane.
Emphasis is mine.

DocJC,

Thanks, that's what I was thinking, it seemed odd to not tie a ground plane to an actual ground somewhere...even if it's all relative. Thanks!

Top
crwper
crwper's picture
Level: Resident
Joined: Fri. Aug 22, 2008
Posts: 588 View posts
Location: Calgary, Canada
Posted by crwper: Mon. Oct 24, 2011 - 03:59 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I ran into this app note recently, which you may find useful:

http://www.ti.com/sc/docs/apps/msp/intrface/usb/emitest.pdf

It doesn't include specific recommendations for high speed designs, but it does mention adding a ferrite bead between USB ground and the main ground. Perhaps this is the kind of isolation the Atmel app note is referring to?

Michael

Top