WEBENCH® Tools/LM5176-Q1: Extra current draw when Vin close to Vout - instability?

Bart Klusek

Prodigy 200 points

Part Number: LM5176-Q1
Other Parts Discussed in Thread: LM5176

Tool/software: WEBENCH® Design Tools

I used Webench to design a power supply based on LM5176-Q1.

The design parameters were:

Output voltage: 24V
Output current: 8A
Input voltage: 11-50V
Switching frequency: 600 kHz

I have tested the circuit with a 4A load, varying the input voltage between 15 and 36V.

The power supply appears to be working - with one exception. When the input voltage is close to 24V, something weird happens. Current draw increases by approximately 1A (regardless of the load) and the inductor starts emitting noises that I learned to associate with instability. The only component that appears to get warmer is the inductor.

Any idea what may be causing this behavior?

Thanks!
Bart

M404939-00-004_02a.pdf

over 6 years ago

0 Youhao Xi over 6 years ago

TI__Guru* 78240 points

Hi Bart,

Thank you for designing with the Lm5176. It sounds like you have stability issues in the buck-boost transition mode. You may reduce the SLOPE capacitor to 100pF from its current value of 300pF, to improve noise immunity. Please try this and let us know if the issue is resolved.

Thanks,
Youhao Xi, Applications Engineering

0 Bart Klusek over 6 years ago in reply to Youhao Xi

Prodigy 200 points

Hi Youhao Xi,

I am happy to report improvement. With a 100 pF slope capacitor, I see no instability with no load. However, under load, it is still there. It seems to be more pronounced when transitioning from buck to boost. When moving in the other direction, it is either milder or not there at all.

Should I keep on decreasing the value of the slope capacitor? Or is there something else I could try?

Thank you!
Bart

0 Youhao Xi over 6 years ago in reply to Bart Klusek

TI__Guru* 78240 points

Hi Bart,

Thank you for updating. You can try to further reduce it but don't use smaller than 47pF.

Please also consider to change the current sense resistor to a quality vendor's. The sense resistor is very important in this circuit, and it should be non-inductive, but some vendor's resistor has significant parasitic inductance.

Thanks,
Youhao

0 Bart Klusek over 6 years ago in reply to Youhao Xi

Prodigy 200 points

Hi Youhao,

This seems to have helped even more. I will test the circuit with a wider range of loads and input voltages and get back to you.

Thank you!
Bart

0 Youhao Xi over 6 years ago in reply to Bart Klusek

TI__Guru* 78240 points

Thank you for the updating. Good luck.

0 Youhao Xi over 6 years ago in reply to Youhao Xi

TI__Guru* 78240 points

Hi Bart,

Can I assume the issue is resolved and close this thread now? You can always re-open this by following a new post.

Thanks,
Youhao

0 Bart Klusek over 6 years ago in reply to Youhao Xi

Prodigy 200 points

Youhao,

The circuit is still not behaving as it should, but there is a significant improvement. Some of my problems might be caused by routing issues, so I will be making another version of the PCB.

Thanks!

Bart

Power management

Power management forum

WEBENCH® Tools/LM5176-Q1: Extra current draw when Vin close to Vout - instability?