• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 410 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.

LMG5200: Efficiency verses Current

GREGG HOLLINGSWORTH
Prodigy 30 points
Part Number: LMG5200
Other Parts Discussed in Thread: LMG1210

I am using the LMG5200 in a class D configuration.  My application requires high voltage but not a lot of current.  I am wondering how much effect on efficiency will there be if I am running at 0.5A verses the higher currents this device is rated for.  I am looking to get about 15W of power out of the device.  Switching frequency is around 2MHz.  Right now, as I increase the voltage on Vin, the efficiency is getting worse.  I know I still have some layout issues to fix, but I dont think they would cause this problem.  I am driving into a fixed resistor for the moment.  Any insight would be greatly appreciated.

Gregg

  • 0
    TI__Intellectual 805 points

    Hello Gregg,

    It is normal to get lower efficiency as Vin increases. The reason is the parasitic capacitance charging losses, which increases as voltage goes higher. This is considered a part of switching losses and approximately equal to Qoss * Vin * fsw. This is the offset switching power loss. The second part of the switching losses is the current voltage overlap during switch node transition, which also involves load current in the loss mechanism. At 0.5A this is negligible.

    LMG5200 has 15mOhm FETs and you get the benefit of reliability and ease of use due to driver integration. However, 15W of output power may be too low for this part to be used. What you are seeing at the efficiency is probably right. If you still would like to get integration benefit of LMG5200 but desire higher efficiency, you need a soft-switching topology. For instance if you have 1W offset charging loss, this is 6.7% loss at 15W but only 1% at 100W. 

    I hope the above discussion helps.

    Regards,

    Serkan 

  • 0 in reply to Serkan Dusmez
    Prodigy 30 points

    Serkan,

    thank you for the information.  I am currently using a soft switching topology as I dont need as much linearity as that in a audio amplifier.  Do you have a suggestion for a set of parts that would work at the high voltage lower power(15W) even if it is not integrated as the LMG5200.  I need a switching frequency of around 2Mhz, soft switching topology, mainly would be driver and FETs. I am ssuming I need GaN to get the higher efficiencies.

    Again thank you for the help in this,

    Gregg

  • 0 in reply to GREGG HOLLINGSWORTH
    TI__Intellectual 805 points

    Hi Gregg,

    If it is a soft-switching topology, LMG5200 is not bad choice, and provide you easiness in layout. At hot temperture, Rdson of LMG5200 would approach 23mOhm. Even though you are doing Class D, if you drive only a resistive load, I guess you would be hard-switching. Can you make sure if you are soft-switching at all times? Because loss increase as Vin increases would probably only happen at hard-switching condition. Secondly, please check your dead-time to make sure you don't have too long or too short dead-time. If you would like to do discrete FETs, I would suggest 15 to 40 mOhm GaN FETs driven by LMG1210. If you would like to use this, I would open another thread under LMG1210 where experts can help. But as I said, LMG5200 wouldn't be a bad choice for 2MHz soft-switching application.

    Regards,

    Serkan

  • 0 in reply to Serkan Dusmez
    Prodigy 30 points

    Serkan,

    thanks for all the information.  I looked at the LMG1210 and if I decide to go discrete GaN fet approach that is a very good part for me.  Again appreciate the information and the timely replies.

    Gregg