1
\$\begingroup\$

The setup I currently have is a Aluminum enclosure for an Android Tablet. Inside the enclosure, a shielded USB cable is routed into the micro USB port and out through a small opening. This cable is then plugged into a 5V charging port on a power supply.

If I tie the enclosure to earth ground, have USB plugged in and shock the enclosure with a Piezoelectric igniter crystal (to emulate an ESD shock), it causes the tablet to reboot between every couple shocks. With USB disconnected, shocking the enclosure does not cause restarts. I've also tried placing a 1M ohm resistor between the enclosure and ground but that doesnt help either.

The one thing I also noticed after tearing the tablet apart is that DC ground is tied to chassis ground (shield) internally. Is there anything that can be done to help suppress the shocks and not affect the USB circuitry which is most likely causing the restarts?

\$\endgroup\$
  • \$\begingroup\$ Have you tried a TVS? \$\endgroup\$ – Ignacio Vazquez-Abrams Jan 15 '16 at 22:07
  • \$\begingroup\$ You probably have a transition on an input pin. Maybe reset. Try adding a cap on reset, close to processor. \$\endgroup\$ – mkeith Jan 17 '16 at 1:38
  • \$\begingroup\$ @IgnacioVazquez-Abrams i was looking at mouser.com/pdfdocs/… but that would require the TVS to be inside the tablet(before the Micro-USB port and the IC), was hoping there was something that can be done outside the tablet. I'll look into regular TVS diodes and maybe connect it from the enclosure to ground? \$\endgroup\$ – EnG Jan 18 '16 at 19:05
  • \$\begingroup\$ @mkeith the tablet is bought fully assembled, will try to get the schematics on the reset circuitry \$\endgroup\$ – EnG Jan 18 '16 at 19:25
  • \$\begingroup\$ What is your ultimate goal here? It sounds like you have cobbled a bunch of stuff together and are simulating ESD using a non-standard technique, and trying to figure out how to address the resulting ESD failure. The easiest thing would be to simply avoid zapping the device with your piezoelectric igniter crystal. If you are trying to design a compliant system, you will have to buy a real ESD gun, because the way you are testing now may not produce the same results as a standards lab. Maybe it won't even fail. \$\endgroup\$ – mkeith Jan 18 '16 at 20:43
1
\$\begingroup\$

Something that looks worth checking: Whether the USB cable has a shield surrounding all of its wires, which should prevent that cable from acting as an antenna picking up transients from the crystal and sending them into the USB input ports.

\$\endgroup\$
0
\$\begingroup\$

Try making a modified USB cable with the shield tied to earth ground, and disconnected from chassis at the tablet.

This should shunt any secondary arcs going into the cable to earth ground and you're just left with the capacitive coupling to all of the pins (and hopefully not a tertiary arc).

The downside is that this may make the system non-compliant or non-functional if you try to run data through the modified USB cable.

There is no guarantee on this, but it's worth a try ¯\_(ツ)_/¯

\$\endgroup\$

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.