• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 189 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

TPSM84A22: Soft Start behaviour

Catalin Palade
Prodigy 10 points
Part Number: TPSM84A22

we use the TPSM84A22 to provide 1.39 V at maximum 8A rail in our system. Please see the sch:

the total output capacitance is 141uF ( 3x 47uF ceramic) plus 0.1uF as seen above.

Note : the inductor seen in the PI filter at the output has no effect on the behaviour described bellow. (the inductor was replaced by a 0 Ohms resistors and the behaviour was the same)

we are seeing situations (so far we have 5 boards out of couple of hundreds behaving this way...) where the power module seemingly struggles powering into capacitive loads. Please see bellow for scopes:

and sometimes (1 in 5 tries) the module fails to power up and it seems entering a pseudo hiccup mode as seen below:

the module is still providing 0.5V as seen. The hiccup time is around 19ms, quite a bit shorter than the 32.8ms as mentioned in the datasheet.

The behaviour seen most likely seems to be related with the Soft Start Feature not behaving as it should. Verry sharp ramp ups (around 22us) can be seen. Per data sheet the module has a pre-programmed soft start time of 4.1ms (typ).

Removing the 3 x 47uF ceramic caps at the output (C14, C15 and C8) makes the module power up without issues but the ramp up is still as mentioned, way faster at 22us than the 4.1ms per datasheet.

Has anybody seen this behaviour?

Replies are much appreciated.

Thank you.

  • 0
    TI__Expert 4360 points

    Hi  

    My colleague will check it and give you a feedback later.

    xueliang

  • 0 in reply to Xueliang Zhang
    Prodigy 10 points

    Thank you Xueliang,

    I did some more investigations myself and it turns out that on these weird (I don't want to call them "faulty" yet...) modules the switching frequency has been changed to 7MHz... this explains the 18.7 ms (around 19ms seen in the scopes) for the hiccup time...: that is exactly 131,072 cycles at 7MHz (0.142857 us)…

    The datasheet mentions 32.8ms which is exactly 131,072 cycles at 4 MHz, the sw frequency for the TPSM84A22

    Regards,

    Catalin

  • 0 in reply to Catalin Palade
    TI__Mastermind 18370 points

    Hi Catalin,

    Your analysis makes sense, the inrush current causes the module to hiccup. 

    Indeed, something might have happened to those modules. 
    An internal component that is responsible for setting the soft-start time and the switching frequency seems to be shortened.

    Please look at the following guidelines for soldering tips with these modules.
    We are confident that tunning the soldering profile can fix this issue.

    https://www.ti.com/lit/pdf/snva853

    Regards,

    Yitzhak Bolurian

  • 0 in reply to Yitzhak Bolurian
    Prodigy 10 points

    Hi Ytzhak,

    Thank you for replying. Definitely something happened to these devices in the field.

    I will contact our assembly house and ask for feedback regarding the soldering profile. I will specifically point to the app note you provided.

    The assemblies these power devices were on failed at customer site part of a trial installation (part of CT scan / Xray machine) after roughly 8 months of working OK. If the soldering profile is suspected to be the cause of this issue, we will be looking at more failures to come. Your suggestion seems to imply that this time window of 8 to 10 months is the time window the soldering issue will start manifesting.

    Note: the customer reported these 5 failures happened virtually all at the same time or in very short succession one after other (minutes apart)

    I strongly believe TI should do a detailed FA on couple of "failed" power modules and provide a precise cause of failure. Could the exposure to Xray be causing the modules to fail?

    Regards,

    Catalin

  • 0 in reply to Catalin Palade
    TI__Mastermind 18370 points

    Hi Catalin,

    I was not aware of the fact that this happened after 8 to 10 months of operation in the field!

    Wondering if ESD, vibration, high humidity/vapor/chemicals, or any other external hazard could cause this.

    You are welcome to contact your local TI Fiels Application Engineer and ask for returning a couple of samples for FA and/or other analysis.

    Regards,

    Yitzhak

  • 0 in reply to Yitzhak Bolurian
    Prodigy 10 points

    Thank you Yitzhak,

    we will contact the local TI FAE  for a detailed FA on couple of failed power modules.

    Regards,

    Catalin

  • +1 in reply to Catalin Palade
    TI__Mastermind 18370 points

    Thanks Catalin.

    Since we are going to handle this return on a separate channel.
    Please set the status of this post to resolved.

    Regards,

    Yitzhak