• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 39 subscribers
  • Views 1129 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.

LM3150: About difference results between two times using webench tool

Hien Pham0
Prodigy 20 points

Other Parts Discussed in Thread: LM3150, CSD16321Q5, CSD17507Q5A, CSD17303Q5

Hi,

I use the same input 10V-21V and IC  LM3150, output 5V5A in two differents times, this time I cant not chose the mostfet M2 to be the same part as the previous.

I dont know why, pls support if you know the issue.5V-5A.pdf

  • 0
    TI__Intellectual 2155 points

    Hi Hien,

    How long ago did you make the previous design?  If there were changes to the FET selection since that design was created, it could impact the parts list.  Can you access the older design in your myDesigns view and share it?

  • 0 in reply to Kevin Ip
    Prodigy 90 points

    Hi Kevin,


    The issue where older FET's are no longer available in newer LM3150 designs also occurs in my case.

    Previously (december 2014) the Webench tool suggested two pieces of CSD16321Q5 as M1 and M2. This part number can (mid March 2015) no longer be selected.

    Another (undesired as I prefer the flatter less parabolic efficiency curve) is that the three parallel 22.0 uOhm Murata GRM32ER61E226KE15L Cin components are replaced by one Panasonic Aluminium Polymer radial capacitor. As a matter of fact the March 23 Webench 12 to 5VDC 8.0A LM3150 design requires a minimum Cin capacity of ± 76 uOhm. Which is more than the 66 microOhms in December 2014.

    It would be great if there are release notes available that document why these changes are applied.

  • 0 in reply to max mustermann
    TI__Intellectual 2155 points

    Hi Max,

    I am sorry for the troubles you are seeing with the differences in the BOM between the older design (Dec 2014) and the newer one (March 2015).

    The component passives library is dynamic with real-time pricing, updated hourly. WEBENCH looks at this library to choose to best component each time a design is created.

    Since each component is given a score based on factors including cost (also footprint size, efficiency, and availability), if there are changes in the price, the ranking of that component may show up differently.

    With the issue on the minimum Cin requirement, I don't see any updates to this component selection. The minimum Cin value is calculated using the operating frequency and the VinMin value. Can you check that the design inputs are the same?

  • 0 in reply to Kevin Ip
    Prodigy 90 points
    Hi Kevin,

    The issue is not that another component is chosen, the issue that the webench user can't manually override this selection because the earlier used CSD16321Q5 is no longer available in the alternate parts list.

    Regarding Cin, I don't know, today webench selects the three pieces of 22.0 uOhm Murata again.
  • 0 in reply to max mustermann
    Prodigy 20 points

    Hi Kevin,

    As the file I atached above 5V-5A, I used WEBENCH tool on April2014. Now with the same input and output parameters, I can not select M1 and M2 with the same part CSD17507Q5A?

    Tks&regards,

  • 0 in reply to Hien Pham0
    TI__Intellectual 2155 points

    Hi Max, Hien,

    I confirmed that we updated the Vds derating for the FETs recently. You should use the FETs in your latest designs to use the latest updates.

  • 0 in reply to Kevin Ip
    Prodigy 90 points

    Hi Kevin,

    With Vds derating for the FETs, do you mean that the FET is able to carry more drain current at higher temperatures?

    The formerly used csd16321q5 starts to derate from 75 degrees Celcius.

    Where the now used CSD17303Q5 M2 starts to derate from 125 degrees Celcius.


    PS Wish that TI would replace my current 10pcs prototype stock of Mosfets.

  • 0 in reply to max mustermann
    TI__Intellectual 2155 points

    Hi Max,

    We increased the safety margin for the FET voltage rating by a certain amount to protect against possible voltage overshoot for PC board layouts which have poor noise characteristics.  I don't necessarily think the past design will have problems, but it may not have as much safety margin if the board layout encourages ringing.