This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

DS90LV049: LVDS chips burning out

Part Number: DS90LV049

An issue has popped up at my company on several different projects that use LVDS driver/receivers. This includes chips DS90LV049, DS90LV032, and DS90LV031. In a system, we generally connect a "master" PCB to several "slave" PCBs with a twisted pair LVDS communication cable and a separate power cable. In several instances LVDS chips have fried when a communication cable is plugged in between the hot master board and a cold slave board (i.e. power cable is not yet connected to slave). It may or may not occur after power is supplied to the slave board, we aren't sure. This has happened to all of the chips I listed above at different times, and also applies to chips at both ends of the communication cable in some if not all cases. And when I say burned out, there is no visible damage to the chip, and replacing the chip with a new one solves the problem. The issue has popped up so sporadically with different people that we have been unable to nail down the exact sequence of events that leads to the problem, so I'm sorry I can't give any more details. I have done some looking online to see if anyone else has had this problem and I can't find any documentation. Has anyone is this forum experienced similar issues, or understand why they might be occurring? Thanks!

  • Hi Stephen,

    This is commonly known as EOS or Electrical Over Stress.  When dealing with multiple system boards, power, and communications cables it is helpful to define a method to ensure a consistent power up and connection sequence.

    I would try to provide power to both boards initially, then follow with the data cable.  It is also important to ensure the data cable is not a "charged object" which can quickly discharge into the system and harm components.  Hopefully by establishing a common GND between the boards and cables before the data signals are connected you will not have anymore EOS events.

    Regards,

    Lee

  • Thanks Lee! Sorry for the late response.

    Your answer pointed me in the right direction. This document was very helpful with further explanation: www.cypress.com/.../download.