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.

DC/DC regulator - LM5116 or LM5085 ?

Other Parts Discussed in Thread: LM5085, LM5116, TPS2490

Here are the power supply specs.

Input - 50V DC (support up to 75V)

Output - 30V DC, max DC current 1.5A, peak current 15A (module should support high output transient current, that may go upto almost 8-10 times at switching) 

Output capacitive load of circuit - apart from the required low ESR output at module, the circuit have a dedicated cap load of total 1500uF (2x750uF) tantalum (http://www.digikey.com/product-detail/en/vishay-sprague/STE750-75T4KI/718-1687-ND/1587386) that has an ESR of 0.5 ohm per cap.

Ripple: Stable converter with minimum ripple (around 50mV) at output.

Need your help identifying suitable part?

Will LM5116 or LM5085 work? If so, which is better suited for the application? 

  • Hi,
    Any comments on above query?
    Let me know, if I need to raise this query on a different forum??
  • Let me see if I can get someone to look at this.
  • Prahlad,

    Since the maximum load current of the design is 15A I think the LM5116 is a better option due to it being a NFET controller instead of a PFET controller like the LM5085. The NFET will have lower RDSon resistance, saving on conduction losses as compared to a PFET.

    The transient response of the LM5116 can be tuned to minimize the output voltage disturbance by selecting the correct compensation components. This should be fairly easy to meet as there is quite a bit of capacitance on the output.

    50mV of ripple is a pretty tight requirement... however this is more based on the selected output capacitors. With such a large bulk capacitance the low frequency ripple should be small. If there is high frequency ripple ceramic capacitors should be used to help filter this out.

    Let me know if you have any more questions.

    Thanks,

    Garrett
  • Dear Garrett,

    I am Gaurav Agrawal, and our team is designing this project.

    An updated requirement from the customer is to have two stages of the capacitive load. 750uF directly on the DC DC output, and then 750 uF  after a sequencing circuit being implemented using TPS2490 hot swap controller. The mosfet used on the controller is IRF540NSTRL.

    We would like to know if you see any issues with this topology. Also if the same FET can be used on the LM5116?

  • Hello Gaurav,

    The IRF540NSTRL MOSFET RDSon (44mOhm) is large and might possibly have large conduction losses since the typical input /output voltage will be running at >50% duty cycle. When looking at MOSFETs for a switching power supply it is good to look from ones that note that they are optimized for switching converters.

    When using the hotswap to add in the extra capacitor bank there can be a large amount of inrush current. The current limit of the LM5116 will need to set low enough to protect the inductor from any over current condition.

    Thanks,

    Garrett
  • Hi Garrett,

    We are not able to realize the design in webench for LM5116. We have taken the following parameters:

    Vin 40 to 60V (Exact Input voltage is 50V DC)

    Vout 30V

    Iout 15A

    The error we see is that the Duty cycle exceeds the limit. The duty cycle for our case is 0.6.

    Can you please realize a webench schematic and share?

    Regards

    Gaurav

    +91 9873929391

  • Gaurav,

    This design calls for 450W(30V,15A) peak power and 150W(30V,5A) nominal power. I know if is only for a small amount of time but WEBENCH will take his as  steady state operation point and try to limit switching frequency to help with switching losses or will show a thermal flag.  I was able to get WEBENCH to create a design by lowering the ambient temperature to -39C. This also for WEBENCH to get around the thermal issues.

    Another opition would be to use the LM5116 in a multiphase configuration and spread the power evenly between the two phases. Here is a reference design that goes over that for the LM5116.

    Thanks,

    Garrett

     

  • Hi Garrett,

    We cant go for a multiphase configuration as we have limit to the real estate and cost on the board.

    When we talk of 15A, it definitely is the switching transient on the output, as the load is 2 of 750uF wet tantalum caps. I assume this means the output ramping to full voltage in around 3msec. Is it possible that we can reduce this transient by further slowing down the output ramp?

    Please provide a webench circuit with this, I am still not able to realize the design even at a lesser ambient temp.

    Kindly reply at the earliest as we need to close the design faster.

    Regards

    Gaurav