When disassembling a failing USB 2.0 flash drive, I noticed a feature on the USB data pins, that I do not understand from the perspective of a student just getting started with PCB layout:

enter image description here

They're definitely traces and not a hole in the solder mask (it's more visible on the phyical device than in the picture), and are much thinner than the actual D+/D- traces running to the ICs on the flash drive itself. I was under the impression that stubs of any sort, and any unnecessary trace length on the D+ and D- pins would be poor in terms of EMI/EMC/noise, due to reflections at the ends of the stubs, and the fact that they appear to be radiators, yet they're included on the drive (produced by a major flash memory maker). This flash drive doesn't have a metal shield around the USB port, as it's one of those connectors where just the PCB fits into the area between the receptacle's body and the receptacle's pins. The only part of the drive not pictured is a plastic body that protects the PCB itself and wraps around the sides and back of the connector for correct physical dimensions for a snug fit.

Could someone enlighten me as to the benefit of these traces?

  • \$\begingroup\$ Is it usb 2.0 or 3.0? \$\endgroup\$ – Gregory Kornblum Oct 17 '15 at 13:43
  • \$\begingroup\$ @GregoryKornblum This is a USB 2.0 device. \$\endgroup\$ – Reinstate Monica - ζ-- Oct 17 '15 at 13:43
  • \$\begingroup\$ Then i don't believe they cause any radiation, not inside the metal connector enclosure. It looks most like a printed filter, but i really don't know :) never seen that before \$\endgroup\$ – Gregory Kornblum Oct 17 '15 at 13:45
  • \$\begingroup\$ @GregoryKornblum What I forgot to mention, but might be key, is that this is one of those flash drives without the metal shield around the USB port, but just the half-width connector that fits into the bottom half of the port. I'll edit the post now. \$\endgroup\$ – Reinstate Monica - ζ-- Oct 17 '15 at 13:46
  • 4
    \$\begingroup\$ My guess is that they connected to something that is now cut off, and were used in some production or test step, maybe in the gold plating, or in some electrical test. \$\endgroup\$ – Wouter van Ooijen Oct 17 '15 at 13:57

I believe they are just connections that are removed by the milling tool that defines the PCB outline- probably for hard gold electroplating.

Notice that all four gold contacts have the connections running off the PCB, not just the D+ and D-. They probably all short together on the PCB panel.

enter image description here


This question is already answered, but my guess is that they're for testing the PCB while it is still panelized.

Check out this episode of EEV Blog where Dave talks about mass production testing using this method.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.