Post Go back to editing

Problem with ADG1606 protection

Hello,

We use an ADG1606 MUX as the final stage of our circuit to connect to external units (Hot unplug condition is normal in our application). From datasheet, I understood that "Overvoltages at IN, S, or D are clamped by internal diodes. Current should be limited to the maximum ratings given" and from engineerzone I know that ADG1606 is protected against ESD.

We have implemented following items as protection for the MUX (Our switch is powered from a single supply):

1- A blocking 1N4148 diode is placed in series with supply to VDD pin. (The circuit is not subjected to negative voltages and we eliminated blocking diode in GND side.)

2- A current limiter circuit is implemented in D channel and prevents currents above 60mA to pass S channels to D channel (S->D). (D ->S is under control and over-current is not possible.)

Before we implement protection methods above, we had different kinds of failures, but recently we had only one type of failure (It is rare, but it happens); A single channel is shortened to GND. It seems an INTERNAL Zener diode of channel is burnt out and the result is short-circuit to GND. If the MUX is protected against over-voltage, why does it happen? What is the solution?

Look forward to your reply,

Best regards

Parents
  • Hello Joe,

    Thank you very much for comments and the document. I think this should solve the problem. But, believe me, the datasheet was not clear and I remember when I first read that part, I thought it must be a mistake of datasheet. Now that you clarified its intention, I guess we have two choices to implement:

    1- Clamp any over-voltage using Schottky or Germanium (is Germanium diode alright?) diodes and leave the current-limiter circuit adjusted for 60mA

    2- Limit the current, passing through S channels, to max 30mA and let the over-voltage conditions happen (I prefer this solution)

    By the way, over-voltages are only positive and almost in 50% of cases, it is normal to have over-voltages in our application.

    Look forward to have your comments on the solutions above or please let me know any other solution you may recommend.

    Best Regards,

    imanjohn

Reply
  • Hello Joe,

    Thank you very much for comments and the document. I think this should solve the problem. But, believe me, the datasheet was not clear and I remember when I first read that part, I thought it must be a mistake of datasheet. Now that you clarified its intention, I guess we have two choices to implement:

    1- Clamp any over-voltage using Schottky or Germanium (is Germanium diode alright?) diodes and leave the current-limiter circuit adjusted for 60mA

    2- Limit the current, passing through S channels, to max 30mA and let the over-voltage conditions happen (I prefer this solution)

    By the way, over-voltages are only positive and almost in 50% of cases, it is normal to have over-voltages in our application.

    Look forward to have your comments on the solutions above or please let me know any other solution you may recommend.

    Best Regards,

    imanjohn

Children
No Data