• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Answers 7 answers
  • Subscribers 63 subscribers
  • Views 2072 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® Tools/TPS55340-Q1: TPS55340-Q1 vs TPS55340 Web bench frequency recommendation

Marinus Yates
Prodigy 170 points
Part Number: TPS55340-Q1
Other Parts Discussed in Thread: TPS55340, , TPS55332-Q1

Tool/software: WEBENCH® Design Tools

Hi,

I was using Webench Power Designer to recommend a power supply for our requirements which are as follows:

Input: DC 3.2 V - 4.2 V
Output: 28.27 V at 0.5 A
Temp: 30 °C
It recommended the TPS55340 and the TPS55340-Q1.
However, the recommended switching frequency is 1.27MHz for TPS55340-Q1 and only 
778Khz for the  TPS55340.
From what I understand this difference will affect the Inductor size and therefore the current through the Switch.
What is the reason for the difference in Frequency?
I would ideally like to keep the switching frequency below 1.2 MHz because of the following note:
"When setting the switching frequency higher than 1.2 MHz, TI recommends using an external synchronous clock as the switching frequency to ensure that the pulse-skipping function works at a light load."
We currently use the TPS55340 with a 22uH inductor and designed for a switching frequency of 600Khz
But that was designed only for 0.2A max load. 
Best regards,
  • 0
    TI__Mastermind 40080 points

    Hi Yates ,

    I don't know what cause the frequency difference,I'll check with the Webench team later on. From design point of view, the two IC can be of the same frequency with the same Vin, Vo, Io, and inductor L.

    you can keep using the 22uH inductor if the current rating is enough, what is the current rating of this inductor, could you please attach the datasheet?

  • 0 in reply to Helen chen
    Prodigy 170 points

    Hi Helen,

    I don't think our inductor will cut it. 

    It is the 22uH one below.

    I'm not too worried regarding the choice of Inductor. I can go with the one that was recommended.  

    I was more concerned about the difference in frequency with regards to the Webench recommended values for the output voltage and max load current. 

    I was under the impression I was missing something since TPS55340-Q1 had a higher switching frequency compared to the TPS55340. 

    I would ideally want to use the  TPS55340-Q1 as our input voltage can go below the 2.9V Vin(min), of the TPS55340, momentarily(5-10ms).

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points

    Hi Yates ,

    Actually, the operation frequency doesn't matter much. The optimized value can be of a range. because the inductor can be chosen based on 20-40% current ripple at big load condition, but if the output current is small comparing to the maximum output capability, the current ratio can be much bigger , like 50%-100%.  So the inductance can be a range based on the experiments of the engineer.  In the reverse, when the inductor is fixed at 22uH, the operation frequency can be also a range.  So please tell me your Vin, Vo, Io, I can optimize the frequency for your.  The different result of Webench simulation maybe because the algorithm is done by different person. They help develop the software but don't understand the power electronics.

    Actually, the minimum input voltage of TPS55340 and 55340Q is the same.

  • 0 in reply to Helen chen
    Prodigy 170 points

    Hi Helen,

    You are correct I was thinking of the TPS55332-Q1.

    Vin: 3.2 - 4.2V

    Vin(nominal): 3.7V

    Iout: 0.1 - 0.4A

    Iout(nominal):0.25

    Iout(max): 0.5A

    We adding 25% headroom to our output current in case our power budget is off.

    We would rather over design slightly. 

    Thank you for your help. 

     

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points

    Hi Yates,

    Thanks for the update! TPS55332-Q1 is very expensive, sugggest you still use the TPS55340-Q1 instead.

    Good luck!

  • 0 in reply to Helen chen
    Prodigy 170 points

    Hi Helen,

    I can't seem to get my calculations to tie up to what the Webench Power designer is giving me.

    I've attached my schematic along with my TPS55340 Boost design calculator excel document. 

    My calculations are quite similar to that of Excel document but not exactly.

    The webench tool target crossover freq is quite a bit higher than the frhpz_mod/3.

    Could you tell me if my design is stable? Or would I need to modify my compensation feedback. 

    TPS55340_MY_DESIGN.xls

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points

    Hi Yates,

    Your calculation is stable. you can keep your parameter.  I have two suggestion :

    1. please use a 4.7uH inductor in your design, 4.7uH inductor can have smaller DCR and therefore much smaller loss;  in your application, with a 4.7uH inductor, the current ratio is around 20%, which is a very good choice. 22uH is really too high.

    2. the effective output capacitance be  higher than 10uF. please check the C/V derating curve for the exact value. maybe you need 1-2 more 10uF ceramic to make sure the Co_eff >10uF. Or there will be stable problem because of the un-enough capacitance.

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points
    The crossover frequency should be less than frhpz_mod/3, usually, we let it around frhpz_mod/5 or even less.
  • 0 in reply to Helen chen
    Prodigy 170 points

    Hi Helen,

    I'm not 100% sure what you mean by current ratio?

    The main reason I chose 22uH because the inductor has a current ripple is 0.2A compared to the 1A of 4.7uH.

    The internal resistance is 37mOhm which would be about  1W conduction loss at max load. 

    I have added an additional 20uF to the output as well. Thanks for that. 

    Best regards,

    Marinus

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points

    Hi Yates,

    What I mean the inductor current ripple ratio, which is usually 20%-40% by design to optimize the loss.

    The DCR of the 22uH inductor is much bigger than the 4.7uH one, so the loss will be much much bigger. 4.7uH inductor is enough in your application.

  • 0 in reply to Helen chen
    Prodigy 170 points

    Thank you Helen.

    I've made the changes it does appear better in the simulations.

    Best regards,
    Marinus

  • 0 in reply to Marinus Yates
    TI__Mastermind 40080 points

    Great! Good luck :)