• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 62 subscribers
  • Views 475 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.

LM5118Q Startup below 5V in boost converter configuration

Ralph Engelhardt96963
Prodigy 20 points
Other Parts Discussed in Thread: LM5118, LM3478, LM3481

Dear Power Community,

I would like to apply the LM5518Q in a boost configuration, converting 5V to 5.5V at max. output current of 2A. The input will be driven by a not very stable power supply, specified to be 5V +/- 0.5V. In the worst case condition the LM5118 would have to start up with 4.5V. Now I see in the datasheet, that the LM5118 cicuitry can be designed to operate with VIN less than 4 Volts, but VIN must be at least 5.00V during startup. The data sheet describes a "workaround" by increasing the input capacitance to at least 100uF, but still requiring >5.00V for startup. However doing this will increase the inrush current to values, which I can not accomodate. I tried the design and it will start up at 4,85V with input capacitance of 33uF, but how reliable is this situation? Is the startup behaviour temperature dependent?

Using the WEBENCH I can design a power supply with my desired input conditions and it does not give me any warning about input voltage violation.

Would it be advisable to use an LM3478 or LM3481 for my problem? They are however not available in automotive grade, which would be highly desirable for my application.  

Thank you very much in advance for any comment.

Kind regards, Ralph.

  • 0
    TI__Intellectual 940 points

    Ralph,

    I would not advise using the LM5118Q in any design where there was at least 5V to start the regulator.  There is variation lot to lot which could cause your design not to start at some point.  My only advice would be to add a small boost converter on the input just to get the LM5118 started and let the VCCX   connected to the output power the regulator from that point.

    Regards,

    Ron Crews

  • 0 in reply to Ronald Crews
    Prodigy 20 points

    Dear Ron,

    thanks for your quick response. Due to reliability reasons I really would prefer a one chip solution instead of starting the LM5118 with another boost converter. I assume this boost converter needs to be able to supply the full LM5118 output current (2A) or even more for a short time to charge the startup- (input) capacitors of the LM5118? There must be some good chips around which can convert 4.5V reliably into 5.5V at 2A without any artificial/conflicting requirements?

    Thanks again and kind regards,

    Ralph.

     

  • 0 in reply to Ralph Engelhardt96963
    TI__Intellectual 940 points

    Ralph,

    You are welcome. 

    The boost would need to supply a transient peak current to charge up the capacitor but the average current wouldn't be too large.  I've never tried this approach so I'm not sure what other pitfalls might be there.

    Regards,

    Ron Crews