• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 458 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.

TPS92518HV: Managed to kill another part, need ideas on how to prevent in the future.

Brian McKee
Intellectual 405 points
Part Number: TPS92518HV
Other Parts Discussed in Thread: TPS92518, LM3409HV, LM3409

Hello.

I have a prototype hardware up and running. Love the part. Love the linearity. I'm making good progress.

In the process of debugging a python test script, I plugged in a load while the parts were trying to pump current. Dim duty cycle was around 40%. Analog Dimming was set very low (PeakCurrent set to around 120 mA).

I use a 350uH inductor in the circuit, which is needed for linearity at low currents, everything else is stock.

I had this problem before with removal of load while the output was on and added a diode from the output to Vin to protect the part from high voltage. That diode still works in the circuit that was damaged, so I suspect that there's some other process happening during plug in while the part is attempting to drive current.

Note, before I plugged in the failed channel and killed the TPS92518, I had plugged in 11 channels without incident.

I have a limited number of prototypes to reproduce this issue and since reproducing the issue kills two channels, I'd like to get some idea where to look before I start killing ports.

Any advice you have would be appreciated. I'll answer any questions you have as well.

Thanks.

  • 0
    TI__Mastermind 28655 points

    Hello Brian,

    It's probably still related to what the diodes prevent.  Are you running high input voltage?  If I remember I believe you are.

    Can you watch Vin and Vout when you are disconnecting and reconnecting the load?  What is your input capacitance?  Is this one TPS92518 per board when testing?

    What is the diode part number you are using?  Is the parasitic inductance of the diode loop small?

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Intellectual 405 points

    Input Voltage is 57.5 Volts.

    Input capacitance is a single 2.2uF ceramic + two 0.1uf ceramic caps.

    I have four TPS92518HVs per POC board. I have six boards total, one is already completely dead because I've been stealing parts from it.

    The protection diode is the BAS16LT1G.

    I think you are asking about the output power diode, which is the PDS5100-13. I chose that part because we've been using it in production with the LM3409HV.

    Above is the layout.

    The fat violet wire up the middle is 57.5 VDC. It feeds two fuses which feed each half of the circuit. The inductors L7 and L8 are 330 uH. Q32 and Q33 are not mounted correctly, see next paragraph. R111 and R112 are the sense resistors. D36 and D39 are the current ground feed. The gate resistor is 0 ohm. The over voltage protection diodes are D34 and D35 located on the bottom of the board.

    Also, there is a mistake in the board, I installed a PMOS FET instead of an NMOS FET because I wasn't paying close enough attention. I was expecting something like the LM3409. So the NMOS fet is installed 180 degrees rotated to connect the source and drain correctly. The gate is tied via a wire. I can send a picture of that if it will help. I don't think that will increase the inductance very much, but I'm no expert on parasitic inductances.

    Keep in mind that the cables connected to our engine are very long. In this case only a few meters, but in the field up to 100 meters.

    I can watch Vin and Vout on a scope when I attempt to reproduce the failure. I'm not doing that now because I have more important things to debug and don't want to wreck any more parts.

  • 0 in reply to Brian McKee
    TI__Mastermind 28655 points

    Hello Brian,

    There's not enough energy in the inductor to bump up the input voltage, just checked that so that's not it.  When you get a chance I would look at those two points, Vin and Vout.

    Asked another engineer and he stated the diodes always protected it so have to figure out what's going on.

    I don't see where D34 or D35 though I see a few bottom side components.

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    TI__Mastermind 28655 points

    Hello,

    I was informed of another possibility.  Do you have resistor between the output (LED voltage) and Vled pins?

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Intellectual 405 points

    There is no resistor between the output and the VLED pins.

    I will try to get to testing as soon as possible. I will keep you up to date.

  • 0 in reply to Brian McKee
    TI__Mastermind 28655 points

    Hello Brian,

    Another question, do you have any capacitance on the output?

    Let me know when you get a chance to look at Vin and Vout.

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    TI__Mastermind 28655 points

    Hello Brian,

    I haven’t heard back from you, I’m assuming you were able to resolve your issue.
    If not, just post a reply below (or create a new thread if the thread has locked due to time-out)

    Best Regards,

  • 0 in reply to Irwin Nederbragt
    Intellectual 405 points

    Hi Irwin,

    I attempted to recreate the issue today and found that at 100 mA (the previous settings) and at 700 mA, I was not able to reproduce it. Plugging and unplugging multiple times was fine

    I think I understand what happened. I'm pretty sure, I forgot to run my TPS92518 initialization code so toffmax was not set to 0xFF. I think inrush killed the part. I have since updated the software so that it will run the init code no matter what test I run.

    Thanks for your help.