LM5121: LM5121 - Output Voltage Dip Observed for switching load

SADHAM HUSEN S

Part Number: LM5121
Other Parts Discussed in Thread: TINA-TI

Hi,

We are using LM5121 - Synchronous buck converter in one of our application. The converter was designed to generate an output of 10V/6A from an input of 4 - 5V.

The load connected to the converter was a switching load, While testing we have noticed an Output voltage dip of about 1V.

The voltage dip is whether due to any Control loop instability.

Details On converter design are given below.

Input Voltage : 4 - 5V (Capable to source Maximum of 60A)

Output voltage : 10V / 6A

Switching Frequency : 450KHz.

Rcomp : 8.25K Ohm.

Ccomp : 0.018u Farad.

CHF : 470p Farad.

Please help us to resolve this issue.

A Ticket had been filed already. Please refer the link.

ticsc.service-now.com/csm

over 5 years ago

0 Youhao Xi over 5 years ago

TI__Guru* 78400 points

Hi Sadham,

Thank you for designing with the LM5121. We cannot open the service link. Could you please us your schematic and some key waveforms? Could you check if your load transient causes the circuit to trigger the input current limit? We would like to see VIN, DS, DG, SW voltages vs VOUT and your load transients.

Thanks,

Youhao Xi, Applications Engineering

0 SADHAM HUSEN S over 5 years ago in reply to Youhao Xi

Prodigy 80 points

Hi Youhao,

Again sorry for the late response.

Schematic and other waveforms are attached please verify.

You had proposed to check the input current limit of the converter, but the input 5V is fed from an 300W Isolated DC-DC converter. [Capable to source 60A amps]

I would like to share the analysis I had done.

I had simulated the circuit in LTspice, and noticed the voltage dip for switching load, I suggest you to simulate the same using TINA-TI for your verification. [During the switching condition no gate pulse is generated]

I suspect the loop stability of the converter will be an issue, so i increased the open loop cross over frequency by reducing the inductor value from 5.6uH to 1uH.

As an impact of inductor change Rcomp value is updated to 47K and Ccomp - 3000pF and CHF - 1000pF. By updating this change my voltage is reduced from 2V to 1V.

Please share your comments and give your best solution.

If any problem retains with the attachment please verify the case id : CS0250045 [Link : https://ticsc.service-now.com/csm?id=csm_ticket&table=sn_customerservice_case&sys_id=0384a465dbaa141456e4d7b2f39619cb&view=csp&_ticdt=MTU5ODM2ODI2OXxsaXRjNzI5NjQxMGM3Mjk2NDEwYzcyOTY0MTA1ZWNjZDU2ZDNjZDJlZXxHQTEuMi4xNzk2Njk2OTM0LjE1OTQ3NDMzNDA]

LM5121.pdf

0 Youhao Xi over 5 years ago in reply to SADHAM HUSEN S

TI__Guru* 78400 points

Our engineer will take a look. What are these two waveforms? What is the horizontal scale? The pictures are hard to read: 10ms/div or 10us/div? What about the second picture?

Thanks,

Youhao

0 Yinsong Cai1 over 5 years ago in reply to SADHAM HUSEN S

TI__Expert 7450 points

Hi Sadham,

I think Youhao is talking about the maximum inrush current limit for the disconnection switch (controlled by DG and DS) instead of the power supply.

Besides the load transient, do you have waveforms for VIN, DS, DG, SW voltages vs VOUT?

Thanks,

Yinsong

Power management

Power management forum

LM5121: LM5121 - Output Voltage Dip Observed for switching load