• State Answer Suggested
  • Locked Locked
  • Replies 11 replies
  • Answers 1 answer
  • Subscribers 34 subscribers
  • Views 566 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TRS3243E: About the cause of the failure.

hisashi yoda
Intellectual 640 points
Part Number: TRS3243E

Tool/software:

We are investigating the cause of a part failure and would like to ask you for your advice.

We have two devices that implement an RS-232 Line Driver or Receiver.

One of the devices (a third-party driver) has a cable connected and is powered on.

The other device (TRS3243ECPWR) is powered off, and we have connected the cable and turned the power on multiple times.

Performing the above operations has resulted in a part that is no longer able to communicate over RS232C and has a different TRS3243ECPWR state.

The number of parts that have failed is 3 out of 650.

The failure rate is 100%.

We will explain the condition of good and defective parts.

Resistance between pins 3(V-) and 4(VIN1) when power is off

　Good parts：Around 650kohms

　Defective parts：Around 100ohms

Voltage between pins 4(VIN1) and Ground when power is turned on

　Good parts：Around 0V

　Defective parts：Around -5V


From this phenomenon, can you infer the cause of the problem and what is happening inside the component?

 

Regards

Yoda

  • 0
    Guru 316690 points

    This sounds as if plugging in or powering up generates voltage spikes that exceed the absolute maximum rating of ±25 V.

    How are the bus lines protected against ESD or transient voltages?

  • 0 in reply to Clemens Ladisch
    Intellectual 640 points

    Hi Clemens:

     

    Thank you for your reply.

    For both devices, the housings are connected to GND, and the cables are designed so that they come into contact with the chassis when connected.

    A varistor is installed in the wiring of the device on the power-off side.

    Even in this state, is it possible that a voltage spike could occur?

    As for the part in question, it's possible that there was something wrong with the part itself even before it was mounted on the board.

    We would like to hear your comments based on this information.

     

    Regards

    Yoda

  • 0 in reply to hisashi yoda
    Guru 316690 points

    What is the clamping voltage of the varistor?

  • 0 in reply to Clemens Ladisch
    Intellectual 640 points

    Hi Clemens:

    The maximum clamping voltage is 44V.

    The component used is the AVRM1005C270KT101N manufactured by TDK.

    Could you please comment on this information?

     

    Varistor part information

    product.tdk.com/.../info

    Regards

    Yoda

  • 0 in reply to hisashi yoda
    Guru 316690 points

    That 44 V happens at 1 A; for 1 mA, it's 27 V.

    The transceiver can be damaged by voltages above ±25 V. It looks as if you need to use a different varistor. But a TVS/ESD diode acts faster and might be a better solution.

  • 0 in reply to hisashi yoda
    TI__Guru 72711 points

    Hi Yoda,

    I agree with Clemen's comments.

    What you could try to do is place a scope probe on the pin you think is seeing the over voltage (I assume it's pin 4 since you mention it's shorted to pin 3) and set a trigger when the device goes above 25V and repeat your testing to see if the over voltage is occurring on the suspected pin and to what degree.

    -Bobby

  • 0 in reply to BOBBY
    Intellectual 640 points

    Hi BOBBY,

    Thank you for suggesting a way to check.

    I placed a probe on pin 4, triggered it with 25V, and hot-plugged the cable about 50 times, but the trigger did not react.

    Please let me know if there is anything else I should check.

    Regards

    Yoda

  • 0 in reply to hisashi yoda
    Guru 316690 points

    Arcing happens only when a voltage is being applied.

    I do not know how well your connectors are protected, but if it is possible to touch the pins, then you need to do ESD tests.

  • 0 in reply to hisashi yoda
    TI__Guru 72711 points

    Hi Yoda,

    You have 3 out of 650 units that are failing so the event that's causing the failure doesn't seem to happen very often. I'm not sure if it's the way your plugging is happening or if it's unrelated to the actual hot plug event. Maybe you can lower the trigger lower to like 10V and see if the hot plug is causing overvoltage. If it isn't then maybe we're looking in the wrong place.

    Clemens may be right, if the hot plugging isn't causing the issue. It could be some kind of ESD event that is occurring. If the connectors are exposed and people can physically make contact with it then it atleast opens the door to that possibility as well. 

    Do you know if the parts are working before you begin your testing with the cable connecting? I'm wondering if the damage could be from soldering or something else prior to the testing.

    -Bobby

  • 0 in reply to BOBBY
    Intellectual 640 points

    Hi BOBBY,

    Thank you for your comment.

    Here are the results of my check.

     

    We lowered the trigger to about 10V and checked for overvoltage, but no overvoltage occurred.

    The plug is not designed to be touched by a human finger.

    I did not check the function of the parts before connecting the cable, so I do not know.

     

    Regards

    Yoda

  • 0 in reply to hisashi yoda
    TI__Guru 72711 points

    Hi Yoda,

    I'm not entirely sure where else the damage could be coming from then. 

    Do you have any idea if the devices were functional before you began testing them? The last thing I can think of is some kind of damage when the parts were being populated/soldered. 

    -Bobby