• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 404 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.

LMZ21701: webench output upper resistor value??

Michael T Schneider
Expert 3790 points
Part Number: LMZ21701

Team, 

I was double checking myself for the LMZ21701 design on Webench.  The datasheet for a 5.0V output has a 242k upper resistor.  Webench recommends using a value 1.116-1.185Mohm.  Why?


  • 0
    TI__Guru 50120 points

    Hi Michael,

    You mention you are simulating 5.0V but the Webench design shows output 5.9V at 0.31A. Can you confirm which one you are trying to simulate?

    If you simulated 5.9V first, then the feedback resistors are set to Rfbt(1.1MOhms) and Rfbb (180kOhms). When you try to create a new custom part, the top resistor will be recommended between the range you mentioned above because it assumes your application is still 5.9V and Rfbb is still 180kOhm.

    You can notice when you go to change the output from 5.9V at 0.31A to 5.0V at 0.31A, update the design, and try to create a custom part for the Rfbt. Now the recommended resistor will be between 924.4kOhm and 981.6kOhm.

    What I noticed is the simulation has the Rfbb fixed at 180kOhm for all designs. You will notice this when you update design for different output voltage. Even if you change Rfbb to a different value (in your case is 42kOhm), the Rfbt is still being calculated based off the previous Rfbb of 180kOhm. This may be the reason why you see the Rfbt range is 1116 - 1185kOhm for 5.9V output even if you changed the Rfbb value. 

    Regards,

    Jimmy

  • 0 in reply to Jimmy Hua
    TI__Expert 3790 points

    Asking the question differently: 

    Why does Webench select Rfbb = 180k, resulting in a very high Rfbt?  Is there a reason the datasheet picks a lower value at ~40k and Webench chooses 180k?  Is it for compensation?  There is a Cff capacitor integrated in the part. 

     

    I am simulating for 5.9V, and I figure I would use similar values to the datasheet realizing that the ratio will change to program in the right voltage.  When Webench “recommends” ~1M for the top and 180k for the bottom, that raised red flags. 

  • 0 in reply to Michael T Schneider
    TI__Guru 50120 points

    Michael,

    Since the divider requires a minimum 2uA of current, this sets the maximum limit on the bottom Rfbb to no more than 400kOhms.

    Any combination up to this point would provide a valid component selection. If anything, if you choose a smaller set of feedback resistor values you will unintentionally sink more current ( ie. For 5V, Rfbt = 232k and Rfbb = 44.2k will draw more standby current than Rfbt = 945k and Rfbb = 180k).

    It may be the case that the Webench design is trying to optimize for high efficiency. If your worry is that changing the resistor divider values will affect stability since there is a Cff integrated in the part , you can always run a transient simulation of the two resistor divider combinations using Webench and check for stability.

    Regards,

    Jimmy