• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 637 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.

LM5170-Q1: Very different current levels in different directions.

Kipton Moravec
Expert 1545 points
Part Number: LM5170-Q1
Other Parts Discussed in Thread: LM5140-Q1, LM5170

I have project that requires bidirectional power.

From the High Voltage (battery) to the Low Voltage we are looking at 250A. From the Low Voltage to the High Voltage we are looking at a peak of 6A, and more likely 3A, with a possibility of going lower as the batteries charge.   

When we were looking at the LM5170-Q1 our first thought was we might be able to do it with 3 chips, 6 phases to get us the 250A. Does that seem reasonable?

The concern is the other direction. Assuming we can disable two of the chips by pulling UVLO low, can we generate regulated 3A or less with HW designed for 83.33A? 

My experience with other power supplies says we have to go to a different mode at really light loads, and I do not know how the LM5170-Q1 deals with light loads. I doubt it was designed for that. 

Would it be better to have two separate power supplies? A multiphase Buck for the 250A and a single phase boost for the 3A nominal (6A peak)? 

If so what would be a recommended chip for the Buck, so we can have multiphase? 


Thanks, 

Kip

  • 0
    TI__Mastermind 21720 points
    Kipton,

    The LM5170-Q1 should work for this application. For the buck operation (high voltage battery to low voltage battery) 250A of current is achievable with 6 phase operation. That would meant that the average current per phase is approximately 42A. Also note that LM5170-Q1 also allows for phase shedding and adding (pins EN1 and EN2) so it is possible to have only one phase in operation which will allow for more accurate setting of the current in boost mode (low voltage to high voltage). The LM5170-Q1 is still accurate at light loads (discontinuous conduction mode) due to the average current mode control architecture that it designed around.

    For buck mode operation the LM5170-Q1 is the best bet for handling 250A load current. This is due to it's ease of stacking phases and strong gate drivers. So even if a separate boost controller is used the LM5170-Q1 is going to be the best option for buck operation.

    For boost mode I just need a little more information. When 6A is mentioned in the boost direction are you referring to output current (current flowing into the high voltage battery) or the input current (current being sourced from the low voltage battery)? What are the battery voltages that this system is being designed around?

    Thanks,

    Garrett
  • 0 in reply to Garrett Roecker
    Expert 1545 points
    The 48V or 56V battery is on the high side and the low side is an uninterrupted battery backed power supply at 12V, not a battery.

    6A max is on the battery side. But when I was verifying, they told me that it could go as low as 100 mA to 200 mA when balancing cells. Can we get good regulation at 100 mA or 200 mA. I do not think 6A is the issue, it is regulating the lowest current. I am guessing it was not designed to operate at such low current in one direction and such high current in the other direction. We are now in the range of 400:1.

    If not would the LM5140-Q1 work just as well for Buck only? It also has two outputs and a sync pin. And is about $2.50 cheaper...
  • 0 in reply to Garrett Roecker
    Expert 1545 points

    For 6 phases I do not see much of a difference between this chip and any other that has 180˚ difference between the 2 phases. 

    I can see the fancy stacking phases is better for 3 phases and better for 4 phases, but not better for 6 phases. 

    You still have to generate off chip phases at 0˚, 120˚, and 240˚ at the frequency you want, or at 0˚, 60˚ and 120˚ for the frequency you want. 

    (I found errors in your data sheet in this area, listed in another post.) If you would separate the OPT into a separate switches for Phase Shift between external clocks for multiphase interleaving, and CH-2 Phase Lagging VS CH-1, so one could be 0 while the other 1 (or vice versa), it would work for 6 phases, without external HW. 




  • 0 in reply to Kipton Moravec
    TI__Mastermind 21720 points

    Kipton,

    For the buck mode  operation (250A) I would still recommend the LM5170 instead of the LM5140:

    1. LM5170 has 5A gate drivers

    -Allows multiple MOSFETs to be driven helping reduce conduction losses

    2. LM5170 is uses average current mode control

    - Increased noise immunity as compared to peak current mode. This will help with current balance from phase to phase.

    3. LM5140 cannot disable error amp

    - need for external buffer. (See reference design PMP30059)

    For the boost mode operation since there is a need to regulate the output current to a few hundred mA, a loop to  the output current to the battery would be needed. This will make the output current accurate through whole range (100mA - 6A). An addition current sense resistor and current sense amplifier will be needed to implement this idea.

    If you have more questions please message me directly.

    Thanks,

    Garrett

  • 0 in reply to Garrett Roecker
    Expert 1545 points
    You have sold me on the LM5170-Q1. Thanks for the help.

    I do not have any questions at the moment, but how do I message you directly?

    How do I mark the question as answered?

    Kip