Need help to identify this IC

[1cristidbr]

I am struggling for about three days now to find the part number for U118. Tried every online database, but I still cannot figure it out. Marking is 14YN/H 891 A981.

Any ideas about which manufacturer uses this format?

The chip in question is a buck regulator in a DFN-8 2x2mm package, pins 1-2 are VIN, and 7-8 are SW. VIN is 5V, VOUT not measured as U118 is now toasted. The board is manufactured by a South Korean supplier, and the part in question sits on the cheaper spectrum. They previously used a dual buck regulator (ADP5034) and changed to this obscure one for the mass-produced version.

Found one part from Renesas which matches the package but the pinout is different.

[wraper]

VOUT not measured as U118 is now toasted

It's also possible there is no enable signal.

[1cristidbr]

The regulator just heats up now and the board is no longer functional. Thankfully, it broke by shorting its SW/PHASE pins to GND so a replacement would fix it. But I am struggling to determine which device it is and what would be its part number

[fzabkar]

FWIW, "14YH" is a load switch with the same marking style and same package ...

https://www.ti.com/lit/ds/symlink/tps22975.pdf

Pins 1-2 are VIN, and 7-8 are VOUT.

"14YN" doesn't lead to anything at TI's web site.

https://www.ti.com/packaging/docs/partlookup.tsp

[DrGeoff]

Is it a NR891 PWM regulator chip?

[1cristidbr]

Thanks a lot. It would seem weird to me that a load switch would be placed within such an architected layout that I've not even considered it. Pinout seems to match. Here is the picture after de-soldering U118.

[fzabkar]

Has the short between Vout and Ground now been cleared? If so, could you bypass the switch by manually switching or permanently bridging the Vin and Vout pins, if only for testing purposes? You could use tweezers or a fuse ...

[1cristidbr]

The short has been cleared after removing the U118 chip. I can test this out to check if board still works. Thanks for the advice 

[1cristidbr]

Pin #5 is GND, so it fits the load switch hypothesis and the TPS22975 pinout. But why would someone place a diode and an inductor in near proximity? 

[fzabkar]

This is one of my search methods:

https://ipv4.google.com/search?q=filetype:doc+OR+filetype:pdf+"absolute+maximum"+"14yh"

I use "absolute maximum" to filter out spammy part suppliers.

[fzabkar]

Pin #5 is GND, so it fits the load switch hypothesis and the TPS22975 pinout. But why would someone place a diode and an inductor in near proximity?

The inductor could be the output of the previous DC-DC converter which is now supplying this load switch via pin #4 (Vbias).

[1cristidbr]

Turns out a transformer is used in the circuit as it is named T101 on the PCB. Why is that?

[fzabkar]

Turns out a transformer is used in the circuit as it is named T101 on the PCB. Why is that?

I suspect you have a battery powered device, and the transformer is part of a boost converter circuit. The diode would be required for rectification of the boosted output.

Edit: I'm not so sure about this, now.

[1cristidbr]

I am more confused now. The circuit has no need to boost the voltage. Why would someone use such a weird setup for a load switch or a buck regulator? :|

[fzabkar]

It might be worth tracing the circuit in that area before you start experimenting.

[1cristidbr]

I finally figured it out. Thanks fzabkar for discovering that the part is TPS22975NDSGT.

The transformer in the circuit is actually a common mode choke, likely DLW5BTM102SQ2. The diode is a protection TVS diode and they are connected as in the reference design below, having the load switch chip placed after these two.

Re-tested the board by replacing the chip with an 0 ohm resistor and it now works as intended, confirming that the TPS22975 failed by shorting itself to GND. Thanks to all who have helped out.