LM5106: LM5106 is burning.

무성 김

Prodigy 50 points

Part Number: LM5106
Other Parts Discussed in Thread: LMG1025-Q1

Hello,

LM5106 is being used for the motor driver.
The problem is that LM5106 is burning.

There is a problem that GND is shaken because there is a mistake in the artwork.

By the way, I'm not sure if the reason why lm5106 is riding is because GND is shaking.

I need your advice.

I have an inquiry.
Should I give the input PWM a Dead Time?
How much time do you need?

Thank you.

over 2 years ago

0 무성 김 over 2 years ago

Prodigy 50 points

The operating frequency is 16 kHz
RDT resistance value is 10K.

0 Daniel W over 2 years ago in reply to 무성 김

TI__Genius 13216 points

Hi,

It should not be necessary to add dead time to the PWM. GND shaking is an issue. Can you send a copy of your schematic?

Is the included scope shot the shaking on GND?

Thanks,

Daniel W

0 무성 김 over 2 years ago in reply to Daniel W

Prodigy 50 points

Using the step motor, use a total of 4 drivers.

The scope shot above is a picture of a 12V line.

This is a picture of 12V and GND.
1ch is 12V and GND is 2ch.

+1 Daniel W over 2 years ago in reply to 무성 김

TI__Genius 13216 points

Hi,

I suspect that the instability in the power source is leading to the device burning. Are there any other devices on the board overheating and burning?

Thanks,

Daniel W

0 무성 김 over 2 years ago in reply to Daniel W

Prodigy 50 points

There is no problem with other circuits.

I checked that the gate driver was the first to fail, then FET and then to the sensing resistor.

Gate Driver -> FET -> Sensing Resistance

+1 Daniel W over 2 years ago in reply to 무성 김

TI__Genius 13216 points

Hi,

Having a sense resistor between LO and VSS can cause issues with the FET that can affect the driver as well. Please see the Typical Application section of the LMG1025-Q1 datasheet for sense resistor and ground bounce recommendations.

https://www.ti.com/lit/gpn/lmg1025-q1

Thanks,

Daniel W

0 무성 김 over 2 years ago in reply to Daniel W

Prodigy 50 points

I found out that not only the GND noise problem, but also the sensing resistor could be a problem.

The sensing resistor is 0.5 ohm.
The data sheet you sent says to use 2 ohms or more. Could this part be the problem?

I saw it as a mistake in the design of the PCB layout and is in the process of redesigning it.
But I need to check if there are other issues before the PCB design is complete.

Thanks.

0 Daniel W over 2 years ago in reply to 무성 김

TI__Genius 13216 points

A lower value sense resistor should not be an issue. The part of the guidelines that I want to highlight is connecting VSS to the drain of the low side FET instead of the board ground.

Thanks,
Daniel W

0 무성 김 over 2 years ago in reply to Daniel W

Prodigy 50 points

It was confirmed that there is a level difference between VSS and FET drain due to artwork error.

We are aware that this could be an issue.

I'm making a new board right now. Upon completion, the test will proceed.
It was very helpful.
thanks

Power management

Power management forum

LM5106: LM5106 is burning.