LMR14030-Q1: Switch-node ringing at No Load

tateo yamashiro

Part Number: LMR14030-Q1

Hi,

We got a question from customer about LMR14030-Q1.
Could you help us?

[Question]
When the IC turns on at Vin=16V and no load, the ringing occur on SW. Please refer to below screenshot.
They are concerned that this phenomenon is abnormal of the IC. Why does this ringing occur?

- Condition -
Vin=16V
Vout=5V
Iout=0A
fsw=2.1MHz

*When Vin=12V, there is no ringing on SW.

Best Regards,
tateo

over 6 years ago

0 tateo yamashiro over 6 years ago

Guru 10890 points

Hi, We need your help. We would appreciate it if you could give us your feedback.

Best Regards,

tateo

0 Vincent Zhang(BSR) over 6 years ago

TI__Expert 4795 points

Hi tateo,

It seems there exist a short period unexpected HSON at 16VIN conditon. Could you provide the schematic and layout for reference? Meanwhile, do you have the zoom out waveform of 16VIN? (which includes two or three ringing waveform in it.).

Best regards
Vincent Zhang

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Thank you for your reply.
They measured the waveform by EVM. The board setting is below.

L1=2.2uH
R3=10k
J3=short

I will send the zoom out waveform of 16VIN by next Monday.

Best Regards,
tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi Tateo,

Got it. I can test the condition on EVM today. Also waiting for your waveform so that we can double check the performance from buth sides.

Best Regards,
Vincent Zhang

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, Here is the zoom out waveform. Could you check it?

Best Regards,

tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi Tateo,

I test the condition on EVM and also duplicate your waveform. I think this phenomeon is because during light load condition, IC enter PFM mode and the pulse is not even, but produce two or three pulses and waiting for a period. In your waeform, HS on for a short period then the SW rise to VIN, but in this condition HS is not turned on completely so that it's couldn't reach VIN. I attach 16VIN5Vout Iout from 20mA-0mA, you could see the first pulse become smaller and smaller until like what you test on EVM.

20mA

10mA

5mA

0mA

Best Regards,

Vincent Zhang

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, Thank you for your support.
They would like to know what is the cause of not turning on completely and whether this phenomenon has a possibility the IC will be damaged or not.
Could you give us your opinion?

Best Regards,
tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi Tateo,

No, there is no possibility to damage the IC in this condition.
I think here HS MOSFET couldn't turn on completely is because the drive current is not enough.

Best Regards,
Vincent Zhang

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, So you mean the drive current is enough at Vin=12V and not enough at Vin=16V? We would like to know why the drive current is not enough at Vin=16V, no load.

Best Regards,
tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi Tateo,

I think the internal LDO drive capability will weaken in PFM mode, but I need to check with Design engineer. I will feedback to you once I get the results.

Regards,
Vincent

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, Thank you for your support. I look forward to your any update on it.

Best Regards,

tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi tateo,

After discussing with design guys, I have to correct my assume. It's not because of Hide side not completely turned on but with the boot cap charge/discharge circuit.

See the circuit below. Inside the IC, there is a boot cap charge circuit paralleled with HS. IC will sense to boot cap voltage the turn on the left FET. When FET turned on, there exists a route from VIN-Boot -Cap-inductor, that's why the inductot increase.

Regards,

Vincent

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, Thank you for your support. You mean the boot volatge is not enough before high side SW is ON, and the boot circuit will be active, so this phenomenon occurr.
Is my understanding correct? And we would like to know one more thing. Why does this phenomenon occurr at 16V input voltage and doesn't occurr at 12V input voltage?

Best Regards,
tateo

0 Vincent Zhang(BSR) over 6 years ago in reply to tateo yamashiro

TI__Expert 4795 points

Hi Tateo,

Yes, you're right. The difference between 12V and 16V input is: IC will sense internal boot cap voltage(VAB in previous block diafgram)to turn on/off the internal FET. Once it below than 5.5V, the internal FET will be turned on. The VAB is relavtive to input voltage, frequency, inductor and load current. In this circuit, 16V input makes the VAB below than 5.5V to make the boot circuit active.

Regards,
VINCENT

0 tateo yamashiro over 6 years ago in reply to Vincent Zhang(BSR)

Guru 10890 points

Hi, Sorry for the delayed reply. Thank you for your support.
They are thinking cause of this phenomenon. Comparing the waveform, on pulse at 16V input is shorter than 12V input and the conduction period of the diode at 16V input is longer than 12V input. So they are thinking 16V input is better boot charging condition than 12V input.
They can't think of the cause that 16V input makes the VAB below than 5.5V. Please tell us the conceivable causes for that.

Best Regards,
tateo

0 tateo yamashiro over 6 years ago in reply to tateo yamashiro

Guru 10890 points

Hi, Is there any update on that? I would appreciate it if you could give us your feedback.

Best Regards,

tateo

0 Andy Chen over 6 years ago in reply to tateo yamashiro

TI__Mastermind 27045 points

Hi Tateo,

Sorry for late!

We will check this ob EVM and get back to you in few days.

B R

Andy

0 Andy Chen over 6 years ago in reply to Andy Chen

TI__Mastermind 27045 points

Hi Tateo,

Let me explain it as below:

1. Normally VAB=Vboot is about 6V, the refresh switch will ON once it below 5.8V
2. Refresh circuit will only be active after recover from Sleep mode.
3. The blocking diode VF is about 1V, Vboot is about 6V, so Vin should be at least 7V higher than Vout to charge up Cboot.
4. Once refresh switch is ON, there is transient period and cause resonant between inductor and parasitic cap on SW node, you can see more resonant periods at higher Vin >20V.
5. for 5V output, you will see small ringing from Vin>13V. and ringing happen at about 18V for 10V Vout.
6. For Vin below 12V, the refresh circuit can't charge up Cboot, and no ringing happen as no charging current, but the following main switch ON/OFF cycle will cause SW node down to 0V, this will helps to charge up Cboot.

B R
Andy

Power management

Power management forum

LMR14030-Q1: Switch-node ringing at No Load