• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 1508 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.

UCC27211: Powering gate driver HB pins with isolated flybuck supplies powered by LM5017

Vishaal Varahamurthy
Prodigy 250 points
Part Number: UCC27211
Other Parts Discussed in Thread: LM5017, LM5107, LM5101A

Hello,

I am using 3x of the UCC27211 to run a 3 phase motor and am trying to use an isolated power supply for the high side rather than bootstrapping the VDD.

I am using a modified version of this reference design : http://www.ti.com/lit/df/snvr323/snvr323.pdf

My version creates 3x ~14V supplies on the secondary side and one 10.2V supply from the LM5017 on the primary side

Flybuck schematic:

Gate Driver schematic (single phase):

I am finding that a lot of the time either my gate drivers or the LM5017 will fail while switching the MOSFETs while driving a motor. The voltages on the transformer are pretty constant, but I was wondering if there is a possibility that the bootstrap diode within the gate driver is conducting due to some weird transients? The LM5017 failure manifests itself in a hole melting into the VIN pin and the gate drivers usually fail with some pins shorted to VDD or GND (not consistent) 

I've run with a thermocouple attached to my gate driver and the LM5017 and have not noticed any gradual overheating.

Are there any problems associated with choosing not to bootstrap from VDD and using an isolated supply at the HB pin instead? VCC_hiside_U (HB pin) is ~14V while VCC_lowside (VDD pin) is 10.2V so the diode should be reverse biased.

Thanks,

Vishaal Varahamurthy

  • 0
    TI__Mastermind 23155 points

    Hello Vishaal,

    Thank you for the interest in the UCC27211 gate drivers and LM5107. I support the high power gate drivers and can address the UCC27211 related questions.

    With the large capacitance on the HB to HS pins the internal boot diode may not be capable of charging this capacitance. There is an application note, SNVA083, that provides guidance on selecting external boot diodes in the case of high capacitance.

    I this case there is a couple of considerations that can keep the boot diode from conducting high current for long time periods. If the HB-HS bias is maintained above VCC the boot diode should not conduct from a DC standpoint. If there is negative HS voltage (below ground) when HS is transitioning low, the boot diode may conduct for the negative spike duration.

    Another suggestion to limit the charge that the boot diode must handle, is to have the boot capacitor close to the UCC27211s sized to address the charging of the MOSFET Qg. Then have some resistance from the local HB capacitance to the bias output bulk capacitance. The resistance only has to handle the average current of the HB bias. This way the boot diode will only have current pulses that is needed to charge the lower local capacitance. For guidance on choosing the HB capacitor, the LM5101A datasheet, section 9.2 provides guidance on component selection.

    I would confirm the HB bias is always higher than VDD to the driver as the main recommendation. And consider adding resistance from the local driver HB capacitance (lower value) to the bias bulk capacitance.

    Please confirm if this addresses the concerns, or you can post additional questions on this thread.

    Regards,

  • 0 in reply to Richard Herring
    Prodigy 250 points

    Thanks Richard,

    On the next spin of these boards I've included a small series resistor between VCC_lowside and the VDD pin, and between VCC_hiside (isolated) power and HB pin - this should let me switch over to a bootstrapped application by removing the series resistor on the HB side. I'm hoping this resistor will allow me to avoid putting another diode in parallel as there's not really much space on the board left.

    I used the reference http://www.ti.com/lit/an/slua887/slua887.pdf to select a smaller value for the bootstrap capacitor (the "detailed" version shows my Cboot is 100x larger than it needs to be right now, if that diode was in fact conducting)

    Interesting thought about HS going below the VDD ground reference - i'll have to confirm this next time I run.

    I'd like to keep this thread open until my 2nd spin of boards comes in and I get to test this again

    Thanks for the help

    Vishaal

  • 0 in reply to Vishaal Varahamurthy
    TI__Mastermind 23155 points

    Hello Vishaal,

    If you have additional questions after you get updated hardware you can follow up and we will address any additional concerns.

    Regards,