LM5119: DCM with interleaved (single output) operation

Hi Andrew, 

As I know, there is no known issue of part failure at DCM mode. 

What's the failure symptom? Can you measure the pin impedance to find out which pin is damaged? 

It is abnormal that the power dissipation at DCM is higher than CCM. Can you check if Vout is oscillating at DCM?

Also, please check the Vds of high-side FET and low-side FET at DCM.

Theoretically we can use EN2 to enable/disable channel 2, but we did not test that. 

You can try.

Thanks

Qian 

Hi Qian,

Thanks for responding.  I am reassured to hear that there are no known issues.

Failure can be associated with high-side driver failure, although this may not be the only thing that has failed.  HO output seems to be short to ground, along with associated VCC1 or VCC2 input.  This (I presume) is what also kills the upper FET, exceeding the gate voltage in a negative way.

All failures have been at no load.

Vout is stable when present, I think.  Or if there was oscillation I was not quick enough to see it.

Vds of high and low side FETs are as expected.  In fact I have not found any abnormal waveforms prior to IC failure.

I am going to try forced CCM instead, as initial tests seem to indicate that this is less prone to failure with my board and design.  I will probably not use EN2 as I do not want to do anything to adventurous; the increased light-load power consumption should not be an issue.

Will let you know if forced CCM this solves my problem.  Waveforms are certainly a lot cleaner with CCM.

Kind regards,
Andrew

Hi Andrew, 

Can you provide schematic in pdf file for review?

Thanks

Qian

Hi Qian,

Forcing CCM appears to have solved my problem.  At this stage I have seen no further failures.

I am happy to run in CCM; the efficiency at light load is fine.

Do you still wish to review the schematic?

Kind regards,
Andrew