• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 39 subscribers
  • Views 472 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.

Webench not cooperating with LM2854

TI Lady
Genius 4025 points
Other Parts Discussed in Thread: LM2854

When I used Webench to create a design for the LM2854 1000KHz given our requirements, it came up with component values that did not give a working implementation.  The output was not the correct voltage and was unstable and erratic. 

 

We calculated the component values by hand using the data sheet and app note, and this design worked.  This working design was stable and the correct voltage, but the ripple was larger than expected.  I reduced the value of the compensation capacitor and the switching waveform and ripple were improved.

 

We now have what appears to be a good implementation, but is very different than what the online Webench tool suggested.

  • 0
    TI__Mastermind 41310 points

    TI Lady,

    Can you give us more information about the design that did not work and what values needed to change to make the design function properly?

    Was the design built and tested in the lab or was the design verified in simulation before fabricating the design?

    Thanks for any additional information you can provide.

     

  • 0 in reply to Britt Brooks - Online Design Tools
    Genius 4025 points

    First, I tried the eval board and it worked great with the original 1.2V setting and after changing the feedback resistor to set the output at 3.3V.

    Second, I modified one of our fully populated boards (Bill and I tried modifications on an unpopulated board so we could eliminate layout and other load issues) and, while the waveform is stable throughout a range of loads, the voltage output seems to drift up from about 3.0V when cold to the proper voltage over the course of about 15-20 minutes. I measured the output of the switcher under light load (connected to the rest of the board) and recorded the following values:

    t=0min (board is powered up; supply is fed from 4.0V supply which measures correctly), V=3.31 but drifts down within several seconds to 3.29

    t=2min, V=3.282

    t=6min, V=3.291
    t=16min, V=3.298
    t=22min, V=3.300
    t=31min, V=3.302
    t=40min, V=3.303
    t=48min, V=3.306

     

    1) How critical is the low pass filter on AVIN? The eval board has it, but the reference designs in the datasheet do not show it (indicating it is optional).

    2) We cooled parts down rapidly and the output dropped drastically to about 2.8V. When I put my finger on the part nothing seems to happen, but when I touch the inductor (cooling it down) the voltage drops. From this, I hit the area with hot air and the voltage recovered within a few seconds but settled around 3.31. How much does inductor temperature affect the switcher?

    3) Have you ever experienced anything like this? Can you give us an area to focus on to try to eliminate the problem (input, load, components, etc)?

    Just to avoid any confusion, the data above is accurate, but doesn't completely represent the problem we've had. As I mentioned in my paragraph, when the board hasn't been powered on for a while, the supply usually starts out around 3.0V.