LM5116: Excessive noise on output after following WEBENCH reference design

Hello Jadon 

Can you capture the 1Vpp noise waveform for more investigation ? Please capture ~ 5 - 10 switching cycles. 

Also, would you please send me the completed quick start ? 

-EL 

Hi Eric, when you say "completed quick start", do you mean the startup of the regulator?

Hello Jadon 

The switching frequency (Fsw) should be ~ 140kHz (switching period should be ~ 7.1us) because RT=23.2k Ohm, but the VOUT ripple frequency based on the waveform is 20us. 

I think you are observing the VOUT ripple caused by a sub-harmonic oscillation whose frequency is 1/3 of Fsw. Please download the Quick Start Calculator at  https://www.ti.com/lit/zip/snvu051 , complete this Quick Start file based on your component selections and attached the completed excel file here.

-EL 

LM5116 Quickstart.xlsx

Hi Eric

Here is the spreadsheet with the components I have used inserted as best I can

Hello Jadon 

Your excel file is corrupted.... 

-EL

Hi Eric. That is very strange. Can you tell me which boxes to fill in that you need?

The schematic I followed and attached in my first post is the one that came out of WeBENCH. If it will help, I can try and recreate the design there and post a link to it?

Hello Jadon 

Can you ZIP the excel file which you completed and post the ZIP'ed file here ? Sometimes it helps. 

-EL

LM5116 Quickstart.zip

Here is the file, zipped... Hope this one works

Hello Jadon 

Unfortunately, it doesn't work. Lots of cells are still broken.

I believe the file was not working on your side too. Right? Then, please download the new one from https://www.ti.com/lit/zip/snvu051 and try again. 

-EL 

HI Eric

The file was working fine on my side. There has to be another, better way to do this. What info are you trying to get from the spreadsheet?

Hello Jadon 

Then, would you please capture the inductor current, RAMP, SW, VOUT(AC coupled) waveforms ? Do you agree  you are observing the VOUT ripple caused by a sub-harmonic oscillation whose frequency is 1/3 of Fsw ?

-EL

Yes i agree. I will measure when i am back at the office tomorrow and send the pictures here.

Hello Jadon 

The goal of capturing the waveform is to confirm if any sub-harmonic oscillation.

If the issue is caused by the sub-harmonic oscillation, then we can try to get rid of the sub-harmonic oscillation by adjusting the amount of slope compensation.

-EL  

Hi Eric

Here are the files for the images of each waveform except inductor current, how would you like me to measure that?

https://drive.google.com/drive/folders/1Kvtaw6uJzxqXHQuLx6oR4M8uMwyAGtRk?usp=sharing

Hello Jadon 

Unfortunately, I am not allowed to access the external hard drive. Please post the waveforms here.

-EL

RAMP (DC Coupled)

SW (DC coupled)

VOUT (AC Coupled)

How should I measure the inductor current? Also, All the testing/waveforms so far have been with a 5A load.

Hello Jadon

Please let me know the input and output voltage 

-EL

Hi Eric

For this test the input voltage was 47.6V as I was running it off battery to eliminate noise caused by a power supply. 

The output voltage was 7V (6.93V) for this test.

Hello Jadon 

Your waveforms are close to the waveforms using 0.45uH, rather than 4.5uH. Please measure the inductor current.  It might be saturated.

<4.5uH>

<0.45uH>

-EL

Hi Eric.

I will see what I can do to measure inductor current, but the inductor I am using is the XAL-1580-452 from Coilcraft and was reccommended from WeBench if that helps.

Hi Eric, Here is the waveform of Inductor current through a 3mOhm Resistor. 

Hello Jadon 

The peak inductor current should be ~10A (= 10A*3mOhm =30mV) but it is ~ 3 times bigger now. Which means the inductance of your inductor is less than your expectation. Please use current probe and double check you measurement is correct. 

-EL

This is how I measured the inductor current. I measured again and got the same results.

Hello Jadon 

You have to use a differential probe if you'd like to measure the sense resistor voltage. Current probe is easier to handle and more noise immune. 

-EL

Hi Eric

I have used my scope to measure the difference between ch 1 and ch 2 and hooked them up to either side of the shunt resistor but now I about ~96mV differential. This works out to about 32A.

I do not have access to a current probe or differential probe at this stage.

Hello Jadon 

If you don't have current probe and differential probe, then just replace the inductor with new inductor (>6.8uH) and check if any improvement. 

Or you can reduce RAMP capacitor and check if any improvement. 

-EL

Hi Eric. I will try and reduce ramp capacitor.

Hello Jadon 

Please post the test results when you are ready.

-EL

Hi Eric. 

This made no difference and the design will need to be changed for any changing of inductor values. I will have to redesign the board.

Hello Jadon 

Would you please share the waveform before & after changing the inductor, and also beforea& after chaging the RAMP capacitor ? 

-EL