• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 62 subscribers
  • Views 151 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.

TPS57040-Q1: TPS57040 maximum switching frequency

Doug Vetter
Prodigy 200 points
Part Number: TPS57040-Q1

I'm using the TPS57040 in an automotive application.  One of the reasons I selected this part is the adjustable switching frequency.  I intend to switch around 2Mhz.

I tried to use Webench to come up with some numbers to compare to my own calculations derived using the formulas in the datasheet but the first thing I noticed is I can't select a switching frequency higher than 1.3Mhz.  Is this a bug in Webench?

  • 0
    TI__Mastermind 22299 points

    Hi Doug,

    On page 22 of the TPS57040-Q1 datasheet  figure 41 and equations 12, you can see the maximum frequency is limited by VIN.

    I checked in WEBENCH, as I decrease the input voltage and the maximum frequency increases.

    Can you check with equation 12 to see if your desired frequency below the maximum to prevent skip mode?

    Otherwise can you share the WEBENCH report? (Export->Print Report)

    Thanks,

    -Orlando

  • 0 in reply to Orlando Murray
    Prodigy 200 points

    I've moved on to another part, but wanted to follow up for the record.

    Despite application's nominal system voltage being 13.8 with the vehicle actively charging, my system design limits maximum voltage (worst case) to 16V, so that's what I used for this application.  Looking at Figure 41 it's clear that the part should be able to handle 2Mhz, while the webench model clearly limits it to 1.13Mhz.  I've seen this issue on similar devices when I've tried to use webench to validate my math, so it looks like a bug in webench or at least incorrect model data fed to it.

    One other point I'd like to make about the datasheet for this part, and many like is that TI needs to step up its game is in providing example data with its functions.  I'm a designer who looks at countless parts every day and if I have to search all over the datasheet to provide the data to plug into the formulas to validate the formula itself against the results provided then I'm wasting time.  I don't know where any of you went to school, but if I didn't show my work I didn't get credit.  TI needs to show its work.

    This is particularly relevant with the formula to determine the maximum switching frequency.  Just looking at it you're including a bunch of values in those formulas with conflicting or incompatible units, so even if I could figure out the correct data to plug in there without looking all over the datasheet I would not have high confidence in the result.

    Please update this datasheet (and indeed ALL datasheets) with example data and all work needed to arrive at the result.  This will simplify every designer's life and lead to more design wins.

  • 0 in reply to Doug Vetter
    TI__Mastermind 22299 points

    Hello Doug,

     

    Sorry that you have decided to move on without providing more details of your requirements in order for us to help you through the design stage. We appreciate your feedback and are looking into issues like these. It becomes difficult to find a balance between explaining every computation in the datasheet and also keeping it a manageable length.

    Regarding the units in that equation, everything is compatible. The fractions become a ratio of voltage which cancel out and you are left with the inverse time. The fdiv is a constant from 1 to 8. I suppose the equation shows "Rhs" and below shows "RDS(ON)" [both meaning the drain-source resistance of high-side FET] but other than that the variables are explained as they should be.

     

    Should you wish to reconsider this device please feel free to reopen this thread.

    Anything else feel free to contact us,

    -Orlando

  • 0 in reply to Orlando Murray
    Prodigy 200 points

    For what it's worth I moved on to another TI part that was less configurable (meaning no compensation network was required), but it still met our design requirements.  The new part placed a few extra constraints on our design but these were easily quantified and implemented.

    I appreciate your clarification of the formula inputs and result, but I must reinforce that had I not found a better TI device you would have lost a design win simply due to an opaque datasheet.  Provide clear examples and "show the work" as my professors used to say.  I don't have a lot of time to evaluate parts as I evaluate dozens a day.  It takes far too much time to ask for help here and get a response.  48 hours might as well be 48 days. The better your datasheets are with respect to providing clear examples the more likely I am to select your part.  It really is that simple.