• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 191 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.

UCC27712-Q1: Used as a charge pump to drive high side Mosfet in 120V BMS system

Subhash Dhyani
Intellectual 300 points
Part Number: UCC27712-Q1
Other Parts Discussed in Thread: UCC27712

Tool/software:

Hello CC,

We are using UCC27712-Q1 in one of the BMS application. Where this part is used as a charge pump.

When charge pump starts working and Main BDU Mosfet ON (with another UCC27712) then the charge pump frequency appears as ripples in the main 12V power supply.

We  have increased and decreased gate resistance of charge pump Mosfets and increase HS pin resistance but  this change heated up UCC27712.

We have to release board in couple of days but with this issue we cannot release for PCB printing. please support so that  we can send PCB tiringly for printing.

Driver is running at 50Khz. Dead time is given but device can itself manage the dead time.

Thanks and regards

Subhash Dhyani

  • 0
    TI__Mastermind 23155 points

    Hello Subash,

    I assume when you mention that when the charge pump is working that the 12V going to the driver VDD has ripple when the main BDU Mosfet is on. I assume the charge pump is used to generate a bias for another driver  circuit, can you clarify, and provide a schematic of the additional circut(s). Also scope plots of the ripple you mention, along with waveforms of the driver HO-HS, LO and HS pin will be helpful. I see that the Mosfets are low gate charge, so I dont see that the gate drive losses would explain the driver running at high temperature. 

    With the low gate charge and relatively low gate resistance, I would expect the switching to be fast, regarding the switch node dV/dt. I looked up the D15 and D17 diode and I did not see a recovery time spec. Can you confirm that these diodes are fast recovery, and try a fast recovery diode if these are not low trr?

    Regards,

  • 0 in reply to Richard Herring
    Intellectual 300 points

    Hello Richard,

    Thanks for your prompt response.

    Yes, there is another driver circuit (which is build up with UCC27712-Q) to driver main mosfets. please see the complete schematics below.

    .

    The main driver will drive 8 mosfets at a time. there is no switching involved only Mosfet ON and remains on till driving signal will be.low.

    Charge Pump is running at 50Khz (50% duty) and  buck converter ( to convert 100V DC  into 12C DC Power integration LNK3209) is running at 66Khz (fixed).I will provide different switching waveforms tomorrow to check the actual performance.

    Till the time please see the resistance values used here and check if we missed something here to add in the schematics.

    Thanks, and regards

    Subhash Dhyani

  • 0 in reply to Subhash Dhyani
    TI__Mastermind 23155 points

    Hello Subhash,

    Thank you for the more complete schematic, I have not seen this method of charge pump for the high side bias, although in theory it should work. Have you seen the app note on using a half bridge driver as a high side DC switch? There are customers using the method shown in this application note. https://www.ti.com/lit/an/sluaa58/sluaa58.pdf?ts=1744132833586&ref_url=https%253A%252F%252Fwww.ti.com%252Fproduct%252FUCC27282

    I am not sure if the DC voltage is too high for this approach. You could always use a separate driver for the charge pump in this configuration. With the UCC27712, you would need a separate driver as this part has interlock and will not work using the LO to drive the charge pump.

    Regards,