Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

High speed USB 2.0 eye pattern violation

Status
Not open for further replies.

curtis.lesmeister

Newbie level 1
Joined
Nov 14, 2012
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,302
We have an in-house USB 2.0 high speed device design with a captive cable that is showing violations of the Template 2 requirements when measured at TP2 (host board connector). Specifically, we are violating the D+/D- Level 2 parameter values of -525 mV. (See attached waveform.)

Some other details:
  • this same device, with designed with a USB connector instead of a cable, passes with flying colors
  • our captive cable measures about 8 inches, including the wire connection to the device
  • the USB wires are broken out from the cable for about the last 3/4" before being direct soldered onto the device
  • the USB lines do have ESD protection components, but there are virtually no line stubs
  • we have 0 Ohm resistors in line with the USB D+/D- lines right after the solder connection to the device board
  • the matched length of the D+/D- connection to the device component is about 1.5"
  • when we insert a 6 ft. USB cable between the host and the device, the eye diagram smooths out and there is no longer a violation of the Level 2 parameter
  • we very rarely see USB disconnects, and aren't sure if the ones we do see are unrelated to the eye diagram violation


After doing some on-line reading, I thought we might be able to correct it with a common mode choke, which I swapped in where the 0 Ohm resistors were. I tried several different values and used chokes specifically designed for applications such as USB 2.0, but none of them seemed to make a difference.

My questions are:
  • what is the cause of this eye diagram violation?
  • how do we fix it?
  • since we may not be seeing disconnects, is there any other problem or damage that may result if we cannot fix it?

I am attaching a waveform of the eye diagram. My differential probe isn't the best, but when measuring eye diagrams of other high speed devices it has proved adequate to give a thumbs up to their performance.

Also, if anyone has any idea as to why the leading edge of the eye is weaker than the falling edge, I will take that input as well. We are not violating the eye, but I don't understand why there is a difference.
 

Attachments


snail888

Member level 1
Joined
Dec 15, 2012
Messages
41
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,288
Location
china
Activity points
1,452
Hi,i've an idea for your information.it may be cause by impedance matching.
 

wowo1215

Newbie level 6
Joined
Nov 19, 2009
Messages
12
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Shanghai,China
Activity points
1,340
Have you try test the impendance of the whole D+/D- loop?
Maybe the impendance does not keep continual.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top