0
\$\begingroup\$

I admit this is kinda "need a crystal ball" issue, however you might come up with some good hints that I can't figure out.

There's a Network Attached Storage unit (QNAP TS-219P+) that has a misfunctioning ethernet port (it doesn't connect to the the network, it's orange LED is on permanently regardless of whether the cable is attached or not). I noticed that the 1.8V LDO (U30 GS1117AX18) gets really hot after 5 secs after power up and the Ethernet IC (U22 Marvel 88E1318) gets warm as well.

I measured all capacitors on Vout of this LDO and they all show a resistance of 0.5Ohm, the same is for the resistance between Vout of the LDO and GND, that clearly indicates the short circuit. C286 and C287 are decoupling the Vin and they are ok (indicating ~10kOhm), that would confirm the short circuit is a Vout side.

The PCB is clear of any physical symptoms.

I could take off the Marvel 88E1318 to check if it's damaged and replace it with a brand new IC but maybe there's something more suspicious that I should check first?

enter image description here enter image description here

\$\endgroup\$
  • \$\begingroup\$ Tried measuring where the voltage between the rails is the lowest? or figure out which other components dissipate the energy? \$\endgroup\$ – PlasmaHH Mar 4 '16 at 13:12
  • 2
    \$\begingroup\$ Here's your crystal ball. electronics.stackexchange.com/questions/97689/… \$\endgroup\$ – Brian Drummond Mar 4 '16 at 13:15
  • \$\begingroup\$ What @BrianDrummond linked, or use a thermal imager to visually do pretty mich exactly the same thing...find the short by using the dissipated heat. \$\endgroup\$ – Robherc KV5ROB Mar 4 '16 at 16:18
  • \$\begingroup\$ If this is a commercial board I suspect the PHY is fried. \$\endgroup\$ – crasic Mar 4 '16 at 21:25
1
\$\begingroup\$

If the only parts that are getting hot are the LDO and the Marvel 88E1318 (Ethernet PHY) then it is probable that the PHY is faulty. If you have the tools then remove the PHY. While it is removed you could check for a short on Vout, you should find it has gone. You could also power up the board with the PHY removed and check Vout, it should now be 1.8V. Replace the PHY and hopefully the fault has gone!

| improve this answer | |
\$\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.