• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 509 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.

Original question:

TPS54302: During Vin on/off, tps54302 often broke

TPS54302: During Vin on/off, tps54302 often broke (2nd thread)

1JK
Prodigy 40 points
Part Number: TPS54302
Other Parts Discussed in Thread: TPS54202, LM5088, TPS560430

Hi,

we observe the same problem as described in the related thread.
The pcb contains 5 buck converters:

1) TPS54302: 24V => 3,3V
2) TPS54202: 24V => 5V
3) TPS54202: 24V => 15V
4) LM5088: 24V => -15V
5) TPS560430: 24V => 12V

Assembly variant A contains 1) only.
Assembly variant B contains 1)-4) only.
Assembly variant C contains 1)-5) only.

In variant A TPS54302 brokes very often if hot plugged to a laboratory power supply  with 24V.
I have observed this while initial testing of 5 boards with no load at the output.

With other assembly variants it doesn't occur. I assume due to the other input capacitors in parallel.

After finding Shawns answer in the related thread hot plugging to 24V was avoided and the problem is gone.
It was plugged to 10V and than the voltage was raised to 24V.


In the real product it is powered with a 500W Meanwell power supply, which maybe can also have a high dU/dt while power up.

So I'm afraid, that i will breake again in the real product. Are there recommendations how to avoid this reliable?
Is it to be expected, that only a large spike on the input could be the root of the problem?
I don't expect a spike for the laboratory power supply with current limiter set to some 100mA.

Before testing that, which will most likely break it again i would like to know if there are other know issues.

Or in other words. Would you expect this problem if hot plugged to a 24V truck battery?

Thanks,
Jochen

  • 0
    TI__Expert 4035 points

    Hi Jochen,

    Hot plug would cause large spike voltage on the input side. That's the root cause for the damage. Usually we will need large input cap to filter the spike and add TVS to reduce the spike. Have you seen any damage with a lower VIN like 16V?

  • 0 in reply to Ryan Hu (BSR-MV)
    Prodigy 40 points

    Hi Ryan,

    "Hot plug would cause large spike voltage on the input side".

    Do you mean caused by the laboratory power supply maybe due to a too slow regulator, which can't deal with the sudden load variation by the inrush current of the input capacitor?

    That is why i have asked this: Would you expect this problem if hot plugged to a 24V truck battery?
    Or is it caused by the circuit itself (inductor).

    The question is, how to make it safe?
    Which input cap related to dU/dt or what ever will make it robust?

    Variants B and C with many other input capacitors from other converters did help.

    My observation was with a laboratory power supply, not with the Meanwell SMPS, which will be used in the target product.
    I will check the input voltage with just a capacitor as load, if there is a spike.

    In that case a 27V TVS with a max. breakdown voltage below 30V should be seen as part of the laboratory power supply in order to compensate the "maloperation" of this power supply.

    BR

    Jochen

  • 0 in reply to 1JK
    TI__Expert 4035 points

    Hi Jochen,

    Even hot plug with a 24V truck battery, it would also have voltage spike. I said that use large capacitance on your board at the input side terminal and add one TVS diode in parallel with the cap to improve the voltage spike.