• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 249 views
  • Users 0 members are here
Support feedback
Options
Options

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.

LM5175: ISNS feature of LM5175 & Multi FET configuration & Thermal management

sakthi vel
Prodigy 160 points
Part Number: LM5175
Other Parts Discussed in Thread: , LM5176-Q1, , LM5176

Hello There,

We are designing a new product for automotive applications (like ECUs) that is powered by an input source of 6 to 36V, at an Ambient temperature of 65 Degree celcius Regulator 1 :-  Output = 12V 5A, Regulator 2 :- Output 5V 10A , Regulator 3:- 5V 15A, we found LM5175 series 4 switch buck boost converter will be suitable for our application. in that we are having few questions.

1. In the LM5175EVM-HP EVM there is an option for adding another Mosfet for all four FETs, Could you please confirm above which current ratting Two FETs configuration is needed? or what is the other parameter determines the multi FET approach is needed?

2. There is an option for disable the Current sense feature by shorting the ISNS- and ISNS+ to the GND, in that may i know what are the dis advantages of not using the current sense feature ? will there be any significant drop in overall efficiency of the converter ?

3. For the same Input and output configuration there is a dis similarities in the LM5175EVM-HP schematics and Work Bench schematics, (like,, Capacitor values for COMP, Rsense values, Bulk & De cap values..)could you please suggest by going by which approach would be more efficient? 

 4. For Higher current ratings, please advise the right approach for thermal management for FETs placement.

Thanks,
Sakthivel

  • 0
    TI__Guru 62975 points

    Hello Sakthivel,

    1. Normally the power dissipation defines when a second FET is necessary. The power dissipation needs to be low enough to get removed from the device and the connection of the device needs to be good enough to remove the temperature. So as the EVM is an existing circuit, it is easy to distinguish the moment when another FET is necessary: when the first one gets too hot.

    2. The disadvantage of disabling the average current sense feature is that now the maximum output current is dependent on the available input voltage. The peak current on a boost converter inductor is higher than the output current, so if the input voltage is low, the maximum output current at maximum peak inductor current is lower than at higher input voltage. The efficiency is slightly going up, as you do not sense the current through a resistor. If your output is shorted, the converter always goes to peak current limit which is not the case when using average current limit.

    3. Do you mean there is a difference between the LM5175 quickstart calculator and Webench? If yes, I would start with the result of the quickstart calculator. But the differences should not be too big. Please tell me if it is very big in one of your designs and which parameters are you using.

    4. This is heavily dependent on your board and cooling capabilities. In general copper helps. In addition, try to place vias below the transistor.

    The thermal management of this curriculum might be interesting for you: Experimental power electronics curriculum and reference | TI.com

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 160 points

    Hello Brigitte,

    Thanks for your reply,

    1. Please find the Schematics of LM5175-Q1 & LM5176-Q1 four switch Buck Boost converter, that we have drafted, kindly review and share your thoughts.
    2. Note that we have not added current sense resistor at the output side, if required please advise the resistor value.
    3. Please confirm that LM5175-Q1 and LM5176-Q1 are drop in replacement if the input and output voltages / current and the ambient temperature is same, 



    4 Switch Buck boost regulator sch.pdf

      Thanks,

  • 0 in reply to sakthi vel
    TI__Guru 62975 points

    Hello sakthi,

    Did you follow the recommendation of external components based on the quickstart calculator? See here: https://www.ti.com/tool/LM5175QUICKSTART-CALC or here https://www.ti.com/lit/zip/snvc208 

    LM5176 is almost a drop in replacement of LM5175. You need to change 2 external resistors:

    • Change the RT resistor to maintain the same switching frequency.
    • There should be a 2k resistor between the input voltage rail and VISNS pin.   It does not affect the loop and normal operation, but it protects the IC when the input voltage jumps above 40V during load dump.  This 2k resistor only applies to the LM5176 but not LM5175 (replace with o ohm resistor).

    And be aware of these differences:

    • LM5176 does not support DCM
    • For LM5175: It is recommended to add 7.8V or 8.2V Zener diodes from SWx to BOOTx to stay below the abs max rating of 8.5V (layout dependent)

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 160 points

    Hi Brigitte,

    Thanks for your inputs,

    Have you got time to review the schematics ? we have drafted schematics based on both web bench and the quick start calculator.

    Thanks,
    Sakthivel

  • 0 in reply to sakthi vel
    TI__Guru 62975 points

    Hello Sakthi,

    Why are there 2 scheamtics in the file you shared?

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 160 points

    Hello Brigitte,

    We are using both the regulators LM5175-Q1 and LM5176-Q2 for the output voltage of 12V and 5V, thats why there is two schematics.

    1. we need to know whether we need to proceed with the components value based on Design calculator or with the TI web bench values. Which is more accurate? 


    Thanks.

  • +1 in reply to sakthi vel
    TI__Guru 62975 points

    Hello Sakthi,

    Normally the design calculator and Webench should not differ too much from each other. Please tell me if you can see a big difference.

    I would start with the values from the design calculator, but as it is a calculator only, you need to test it when you have it in hand and might need to change some components.

    Best regards,
    Brigitte

  • 0 in reply to Brigitte
    Prodigy 160 points

    Hi Brigitte,

    Thanks for your inputs, both calculator and web bench values are nearly same.