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.

LM5105: Full bridge application with CSD87588N FETs

Part Number: LM5105
Other Parts Discussed in Thread: CSD87588N, , CSD87333Q3D, CSD88584Q5DC, CSD87353Q5D, CSD87334Q3D

I am investigating using 2 of the LM5105 half bridge gate drivers for a full bridge coil drive circuit (see schematic sketch).  For the bridge FETs, I'm looking at the CSD87588N dual N-channel part.  Some of the parameters of the circuit are as follows.

  • Vdd and bridge supply voltage = 14V
  • PWM input = 327kHz (0 - 100% duty cycle)
  • Output drive waveform = sinewave
  • H-bridge load = 22uH to 160uH wire wound ferrite core inductor
  • H-bridge currents = 18A pk-pk
  • H-bridge power dissipation = 1.5W
  • This circuit is used to generate an AC magnetic waveform

Questions

  1. What determines whether you need to add the optional external bootstrap diode?
    1. The forward voltage drop of the internal diode is fairly high.  If we have room on the board, is it preferable to add a fast low voltage drop external diode?

  2. Is the CSD87588N FET a good choice for this application?
    1. I am not as familiar with dual N-channel FET parts where the high and low sides have different specifications.
    2. We currently use a Vishay Si7218DN dual FET in a similar H-bridge application, but this part is currently listed as EOL.
  3. Do you see any issues with driving the gate of the CSD87588N FET with 14V.  The maximum gate drive voltage (Vgs) listed in the datasheet is only 16V.

Thank you for your help -

Scott

  • Hi Scott,

    For your bootstrap question, please refer to this document: www.ti.com/.../snva083b.pdf

    As for your question about the best FET for your application, let me ask one of my colleagues who works in the FET group to recommend the best part for your application.
  • Hi Scott,
    Thanks for your interest in TI power block. The CSD87588N has been optimized for low duty cycle, synchronous buck converter applications. Because of this the high side and low side FETs are sized differently. High side is smaller to minimize switching loss and low side is larger to minimize conduction loss. It can probably work in your application but the performance may not be great due to power loss. I checked the Si7218DN datasheet and it uses symmetric FETs. There shouldn't be any problem driving the CSD87588N with 14V as the abs max is rated at 20V. However, you will not be getting any lower rds(on) @ 14VGS and will be dissipating additional gate drive power in the LM5105's.
  • Thank you both for your replies.

    Does TI offer any dual symmetric N-channel FET packages that might work well as a replacement for our existing Si7218DN FET that is going EOL? I've been trying to find a balance between low total gate charge and low Rds(on). At our PWM drive frequency of 327kHz, I think prioritizing low gate charge over on resistance makes sense. Although, the strength of the LM5105 gate drive might make the amount of gate charge less critical. Ideally, I'd like to find a dual FET package that is smaller than an SO-8.

    Thanks for your help -

    Scott
  • Hi Scott,
    The only dual part we have with symmetric FETs that may work for your application is the CSD87333Q3D. It's in 3.3 x 3.3mm QFN style package. Please note, the VGS is rated at 10V abs max for the FETs with a recommended max operating of 8V. Therefore, you would need another, lower voltage such as 5V to drive the FETs. We do have a 40V (CSD88584Q5DC) and a 60V power block (CSD88584Q5DC) with symmetric FETs but they are optimized for low frequency (20kHz) motor drive applications and as such are very low rds(on) and high charge devices.
  • Hi Scott,
    A couple of other options for symmetric FETs: CSD87334Q3D (3.3 x 3.3mm QFN, 30VDS, 10VGS) and CSD87353Q5D (5x6mm QFN, 30VDS and 10VGS). As with the other option, you will need to drive the gate with 8V or less.