• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 281 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.

CSD19538Q2: High power dissipation at low currents while switched

Keshav Raj
Prodigy 200 points
Part Number: CSD19538Q2

Hello,

I am using the CSD19538Q2 as a low side switch for controlling a solenoid. I am driving it using an MCP1416 gate driver which can source/sink at 1.5A peak current with a drive voltage of 12.9V at a PWM frequency of 24kHz with an input voltage of 32V.

At a coil current of 300mA, the FET should only dissipate a few dozen mW, yet it appears to heat up to >70 degrees. I have tried altering the PWM frequency down to 4kHz and the end temperature is the same.

Below is the relevant parts of the schematic, initially I had a 10R resistor on the gate, but replacing it with 0R didn't make a difference.

I am unsure as to why the FET is dissipating more power than expected, any help is appreciated.

Attached in addition is the switching waveform at the FET gate, and the switched voltage at the drain. All captures are done at 24kHz, coil current 0.35A with a gate resistance of 0R.

Gate switch on:

Gate switch off:

Drain voltage at switch off:

Drain voltage at switch on:

  • +1
    TI__Mastermind 38190 points

    Hello Keshav,

    Thanks for your interest in TI FETs. It does seem like the CSD19538Q2 is getting pretty hot under your test conditions. I'm going to need additional information to try to help you resolve this issue. How and where are you making the temperature measurement? Are there other devices on the board near the FET that may be heating it up? Is there any way you can capture the drain current waveform? Is there adequate capacitance from VBUS to GND? There are two main power dissipation components for the FET: conduction loss due to the rms current thru the FET and switching losses when the FET turns on and off. There's also gate drive loss but that will be dissipated in the gate driver IC and gate resistor.

    Best Regards,

    John Wallace

    TI FET Applications

  • 0 in reply to John Wallace1
    Prodigy 200 points

    Hi John,

    I am using a FLIR thermal camera, but I do have onboard nearby to perform temperature measurements as well.

    The flyback diode (RB068MM100) and the 16V zener are close by to the FET and are conducting the decay currents of the coil and collapse the field faster. I'm going to try bypassing that zener as that might be the other major source of heat.

    I have a 10uF 100V ceramic capacitor, directly on VBUS. But I have an additional 47uF electrolytic and another 10uF ceramic before an overcurrent protection IC (MAX17527).

    Here is a snapshot of the local layout of the FET:



    Unfortunately I don't have an easy way to insert a shunt between the drain and the node where the coil and flyback circuit connect. Or at the source of the FET to ground.

    Thanks for your help, I'll edit this post with my findings after bypassing the zener.

    EDIT:

    After bypassing the zener the circuit runs much cooler. It's possible the zener was getting hot and due to the proximity it looked like the FET was getting hot.

  • 0 in reply to Keshav Raj
    TI__Mastermind 38190 points

    Hi Keshav,

    Thanks for providing the additional information.Maybe the zener diode is getting hot and conducting heat to the FET. They are electrically connected at SL1.  This also provides a thermal path to the drain pad of the FET which is the main path for heat to flow to/from the device. I know it's not always easy to measure the current without impacting the circuit performance.

    Thanks,

    John