• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 353 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.

TPS40170: Pin16 ( SW ) absolute max limit

Norman Oliva
Prodigy 210 points
Part Number: TPS40170
Other Parts Discussed in Thread: LM5145

(upper Fet )Specs says, Pin16 ( SW pin )  has absolute max limit = Vin. Basically once the upper Fet switches ON then the SW pin will see Vin and therefore the said Pin will always be at the 100% stress level.

This pin is also tied at the low side Fet where-in due to leakage inductance from PCB trace can potentially create additional spikes on top of Vin and will therefore see >100% stress.

Does in mean in order to use TPS40170  the  design should be perfect/spikes is not allowed on the Vds of the lower FET because it should be equal to Vin, how do we do that??.

Even if the Body diode  ( upper Fet ) clamps the spikes to Vin during OFF state ( lower fet ) it will still be "Vin + Body diode drop"  and still > 100% stress.

Our company has internal voltage derating guidelines including stresses on every pins of the IC and 100% stress is not acceptable due to reliability reasons.

I think something is not right in how the absolute max is written for SW, can you check with TI experts on this please, I can't believe that it does not have upper tolerance limit, if there is upper limit, please advise the level  such as absolute max  for SW=Vim + given value? so we can really know the stress level on our side.

In our application, we are seeing "Vin + 5v for 5nS ) is this +5V/5nS spike on top of Vin enough to kill the part?

Please advise.

Thank You.

What can potentially fail in SW ckt if it sees >Vin??

Below is the specs:

  • 0
    TI__Guru 74052 points

    Hi Norman,

    In general, the SW node should not exceed the 60V DC rating. However, ringing of short duration (like 5ns) and relatively low amplitude as mentioned is usually acceptable as it is largely sustained across the parasitic bond wire inductances, not affecting the silicon itself.

    For higher voltage capability, please consider the LM5145 75V controller.

    Regards,

    Tim

  • 0
    TI__Guru 74052 points

    Norman,

    My understanding is that there is no reliability concern here.

    If you would like to attenuate the SW voltage overshoot and ringing, consider adding an RC snubber from SW to GND, e.g. 2.2 Ohms and 220pF.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Prodigy 210 points

    Hi Tim,

    Our internal voltage derating for controller/pins is 80% based on absolute max given by specs.

    In this case we will not be able to pass our internal  reliability guidelines  because the part specs itself explicitly says  "SW pin absolute max = Vin".

    Therefore we will have reliability issues internally unless TI gives us guarantee that this is not a concern.

    Fyi:

    In our application, below is what we are measuring:

    Vin = 30Vdc

    SW pin = 35Vmax, that 35Vmax lasted for around 5nS then settles back to 30V whenever the low side fet turns-off.

    As far as our internal QAV is concern, pin16 ( SW )  is  116.6% voltage stress ( 100 x 35/30) with respect to  SW  absolute max rating which is Vin or 30V.

    We already have some sample units sent to customer and we wanted to assure our customer that what ever we measured in SW pin is non-issue and will not cause any failure. 

    On the other hand, even if we are able to remove the spike via adding snubber ckt, we will still be at 100% stress level and will never pass in anyway our 80% internal derating guidelines.

    Unless:

    1. We get a guarantee from TI that SW=35V/5nS spike will not cause any reliability and/or potential issue for Vin=30V. Then our QAV will wave the voltage derating issue we have for the part and then we can continue using it.

    2. Or, if TI can issue a document to us that explicitly say, "SW absolute max" can be 37.5Vmax ( even for a given specific duration let us say 10nS or so) then we can give that document to our QAV so they can adjust the stress level from 116.6% to 80% then we will be good and will continue use your part.

    By the way, we would like to set a short call with you, some of the guys here may have some more clarifications, kindly advise if this is possible so we can close this issue as soon as possible. If you are okay for a call, I can send the invite just kindly advise contact details.

    Rgds,

    Norman

  • 0 in reply to Norman Oliva
    TI__Guru 74052 points

    HI Norman,

    We're checking the abs max spec now and should have a reply shortly.

    Regards

    Tim

  • 0 in reply to Norman Oliva
    TI__Guru 74052 points

    Hi Norman,

    I confirmed that the ABSMAX rating for SW is 62V. It is only referenced to VIN because of the MOSFET's body diode present from VIN-SW. There is no concern here from a SW pin perspective. 

    SW going 5V above VIN for 5ns is within the expected operation of the device. This is not a concern at 32VIN.

    Regards,

    Tim

  • 0 in reply to Timothy Hegarty
    Prodigy 210 points

    Hi Timothy,

    Thank you for your confirmation that SW pin absolute max is 62V.

    Is there any plan to update the specs to reflect the abs. max rating of the device for SW pin to 62V max  just like the other devices? This is just to make the specs clear and avoid wrong interpretation/assumptions.

    Please advise.

    Rgds,

    Norman

  • 0 in reply to Norman Oliva
    TI__Guru 74052 points

    Norman,

    Yes, I will plan to update the datasheet accordingly. It should appear online in 4-6 weeks.

    Regards,

    Tim