LMR51450: only 0.7 V at the output

Hi Yves,

Have you ensured that you device is properly soldered onto the board? If yes, have you seen this issues with any other devices or only this device?

Thanks,

Dino

Hi Dino,

thanks for your quick reply!
I ordered the PCBs assembled and soldered from JLC. So far I have not had any problems with any order of any soldered ICs.
I now have 14 test boards here (two different basic routing layouts, four different inductors, four different configurations), all with the same problem.
The physical pin alignment/chip rotation of the LMR51450 is also correct. 

It's also the correct chip. I've checked the IC markings.

Can you provide screenshots of VOUT, VIN, inductor current, and SW?

Thanks,

Dino

VIN at the input capacitors = 13 V (blue)
VOUT at the output capacitors/inductor_out, about 0,7 V (yellow)

SW at the inductor_in (yellow)

Power Supply (1,2W = two LMR51450 test-stations in parallel without load)

One of the test-PCBs

However, I don't know how to measure the inductor current in this Setup. 

Hey Yves,

Can you populate the suspected unit onto the EVM for this device to ensure the device isn't damaged?

Thanks,

Dino

Hey Dino,

today I've ordered an EVM and a few LMR51450SDRRR ICs. So I can cross-check/resolder the LMRs due to EVM and my Test-PCBs.
I'll update this thread with the results.

What's strange is that not a single LMR circuit on my test PCBs works. They all behave the same way. That points to a systematic error in the circuit diagram or routing, right?

Thanks,

Yves

Hi Yves,

One thing that does pop out is the lack of a ground pour on the top layer of the device as is stated in the datasheet, page 26-29. This is something that should be on the PCB, but is not done on yours.

Thanks,

Dino

Hi Dino,

thanks for the hint! I'll keep it in mind for my next routing of a Test-PCB.

My two different layouts/routings are a 4-layer and a 6-layer PCB. The second, fourth, and sixth layers are each ground. This means that ground is directly beneath the entire circuit.

This is for example the second layer.


Thanks,
Yves

Hi Yves, 

Can you send us your layout files so we can take a more detailed look? Will get back to you on it early next week, the 26th is a holiday for us here in the US.

Thanks,

Dino

Hi Dino,

I've attached the KiCad 9.0 files here. They're reduced to the circuit for the 6V buck converter in a 4-layer design.

v3.0_LMR-Test_TI.zip

The EMV and the LMRs ICs should arrive on the 27th or 28th.

Thanks a lot!

Yves

Hi Yves,

Having some trouble opening the files you sent. Could you screenshot each layer and put them in a pdf file? Apologies for the difficulties.

Thanks,

Dino

Hi Dino,

yes of course. Please have a look at the attached files.

LMR-Test-PWR Supply.pdf

LMR-Test__Assembly.pdf

Thanks,

Yves

Yves,

Thank you. I will looks at this with my team and get back to you soon.

-Dino

Hi Dino,

the EMV board and the individual LMRs have arrived this week. I've tested the EMV. Without a load, it consumes no power, and with a load (1A), it doesn't get particularly warm either. With a load, I can also see much better 500kHz edges with the oscilloscpe on the SW. The EMV works as it should.

It's really difficult to re-solder an LMR by hand, but I managed to solder one of the new LMRs onto one of my test PCBs. It behaves the same way as before. The frequency on the SW is around 2.2 MHz with no clear edges, and it gets about 55 degrees Celsius in idle without a load. I think it's a design flaw, but I just can't find it yet.

Have you made any progress with my layouts?

Thanks

Yves

Hi Yves,

Our team has been very busy this week unfortunately, will get you some feedback as soon as I can this week.

Thanks,

Dino