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.

TPS63700: power-on problem

Part Number: TPS63700

Hi all,

in our card we've used a TPS63700 converter to generate a -15V output voltage starting from an internal 5.25Vdc.

We've already produced 100 cards in the past, without any problem.

On the last production batch, without any hardware change in the DC/DC or in the load, we've seen that sometimes the converter doesn't complete the power-on sequence and the output voltage follows the attached waveform.Purple line is the -15Vdc output, while the red line is the input current.
Yellow trace shows the voltage on IN pin (pin 5) of TPS63700


Current scale is 1Amps=>0.33Volts, therefore the peak input current is around 0.98A and the current pulse duration is 10-12ms.
Standing to our test, we've seen that the output voltage reaches -16.4V, then the switching stops and the output voltage falls down exponentially.

We've repeated the test many times and we've seen that 90% of power-up procedures are successfull, but sometimes the converter does.t complete the power-on sequence.
The static load of the converter is negligible (<20mA@-15V), therefore we think that this behaviour could be related to the internal current limiting logic.
Total capacitive load is 66uF, plus DC/DC Cout (composed by 5 x 4.7uF).
If we reduced the external load from 66 to 44uF, the power-on sequence is always completed successfully.

Do you think the cause of the problem is I>I_LIM=860mA?
Could you provide any suggestion in order to solve the problem on the produced batch?

In attachment you can find our circuit:

  • Hi   Fabrizio,

    Our FAE team has emailed to let us know. 

    how about we finish discussion here? they we let you know once we have any update.

    thanks for your detailed description.

    Regards

    Tao

  • Hi Tao,

    thanks for your reply. Please contact me for any suggestion or any test you need to perform to clarify this problem.

    It is extremely urgent from our side to solve this issue, since we've already produced many cards.

    Regards, Fabrizio

  • Hi   Fabrizio,

    Could you help to let customer do follows to confirm?

    • Increase the input cap
    • Add a cap from EN to GND to make sure device start to switching when input is built already

    Regards

    Tao

  • Hi Tao, we've introduced an RC filter on the EN pin , so that the startup of the converter is 7-8ms delayed with respect to the original configuration.

    In this way everything seems to work ,even though the margin is not so big...

    I have another question regarding the UVLO and the current limit protection.

    What is the expected behaviour of the DC/DC? If an UVLO (or an overcurrent) event occurs during the VIN rising edge, does the TPS63700 try to restart again after some time or it never tries?

    Thanks

  • Hi Fabrizio

    according the description in DS,i am afraid UVLO will be locked down till EN restart.

    Regards

    Tao

  • Hi Tao,

    we've modified the input filter to patch the problem.

    Anyway, we'll never use TPS63700 in the future, since the behaviour of the protection is not acceptable for our products.

    Thanks for your precious support

  • Hi   Fabrizio,

    thanks for your reminder.

    we did this protection in order to avoid some really bad input source for device. we will check with our team internal and customer to make sure whether this way of protection is necessary. thanks for your advice again.

    Regards

    Tao