LM5023: LM5023 light load issue

Jimmy WANG

Genius 4495 points

Part Number: LM5023

Tool/software:

Hi team,

customer tesed LM5023 and got some issue here.

Red is VGS

Grees is VDS.

Is this waveform normal?This waveform is different from what I know.

Condition 100VAC Vo=15V 0.1A or less

over 1 year ago

0 Mike O' over 1 year ago

TI__Guru* 86286 points

Hello,

Your inquiry has been received and it will be addressed in the order that it was received.

Regards,

0 Mike O' over 1 year ago in reply to Mike O'

TI__Guru* 86286 points

Hello,

Your design seems to be going from 50% duty cycle to a minimum duty cycle. I am not sure what is causing this bursting. It could be noise on the QR causing the FET to turnoff or and instability.

Could you probe Comp, CS, VCC Out during this condition with an oscilloscope? It is important to have them all on the same plot.

If you want to share a schematic, I will review it for you as well?

Regards,

0 Jimmy WANG 11 months ago in reply to Mike O'

Genius 4495 points

Hi team,

100VAC

264VAC

0 Mike O' 11 months ago in reply to Jimmy WANG

TI__Guru* 86286 points

Hello,

I think you might have noise on the QR pin make the controller think zero current is reached and turning off the gate drive.

In your evaluation you might want to swap VCC for the QR pin to verify?

If it is noise on the QR pin then you should be able to resolve it by adding some capacitance to the QR pin. You might even see that just probing the QR pin removes the issue. If it does than 10 pF of capacitance would fix the issue. This is because the scope probe capacitance is generally around 10 pF.

Regards,

0 Chen Alex1 11 months ago in reply to Mike O'

Genius 4265 points

hi Sir

add the 220p cap between the QR Pin, still cut off

0 Mike O' 11 months ago in reply to Chen Alex1

TI__Guru* 86286 points

Hello,

You might want to study QR, Comp, CS and VDD with an oscilloscope to see what they look like to verify there is no noise on the QR pin. I would study before and after adding any capacitance.

If the noise is not present from the waveforms than the next place to look at is loop stability. You will need to check it with a network analyzer to make sure you enough phase margin at cross over (> 45 deg) and enough gain margin less than – 3 dB.

Regards,

Power management

Power management forum

LM5023: LM5023 light load issue