Such a horrible design from Lenovo

[terrance]

I replaced it with IHLP2525CZER8R2M11

What I meant is, they should not design the operating point at 9A ripple current. It is unreasonable even if the best part can handle 13A. For buck converters, very high ripple current also puts lot of stress on input capacitors. The low inductance also drive the switching frequency so low, that it was buzzing.

[terrance]

it is not about if the inductor can handle that ripple current, it's about the unreasonably high ripple current.

[wraper]

I replaced it with IHLP2525CZER8R2M11

What I meant is, they should not design the operating point at 9A ripple current. It is unreasonable even if the best part can handle 13A. For buck converters, very high ripple current also puts lot of stress on input capacitors. The low inductance also drive the switching frequency so low, that it was buzzing.

And then you'll find out it cannot handle the full load.

[terrance]

so the real problem is user uses too much performance for too long time.

[terrance]

a compromise against heat dispassion performance for a hot cpu IS an awful design.

or it's the user's fault using too much performance.

[wraper]

a compromise against heat dispassion performance for a hot cpu IS an awful design.

or it's the user's fault using too much performance.

Any laptop is a compromise of size/thermals and (especially sustained) performance, no way around it. Increasing inductance 8x with no consideration of how much current that rail may pull is absolutely dumb. In the end you may find out that nearly everything connected to that voltage rail is dead if TPS51285B fails short due to inductor saturation/failure.

[thm_w]

For anyone wondering: https://www.digikey.com/en/products/detail/vishay-dale/IHLP2525CZER8R2M11/1220684
8.2uH 4A

Feel free to post some waveforms.
Was the 600ns measured at full load or partial load? Is that a 24% duty cycle then?

[mtwieg]

If you increased the inductance that much I would be concerned about operation at max load. Do you know what the max load current will be?

Also I'm assuming you verified the stability margins..?

[f4eru]

perhaps the frequency increases at high loads ?

[thm_w]

perhaps the frequency increases at high loads ?

Datasheet implies fixed frequency 400 or 500kHz.
Though pulse skipping at low loads.

[terrance]

mow much load does 6 usb ports with current limiters take? 30A?

[wraper]

Up to 1.5A per port if port supports rapid charging. Not to say some phones will try to pull such current from any port and this converter may power more things than just USB. Using a 5A inductor for powering 6 USB ports is nuts.