TPS563200: TPS563200 voltage drop

Hi MALACHI,

We cannot open Kicad file, could you please upload pictures of schematic and layout? Thanks.

Regards,

Shipeng

Hi Shipeng, thank you for the response.  Does this work better for you?ENGH-36-img-v1.zip

Hi Malachi,

The layout is not good. The FB wire should be short so R2 and R3 should be placed close to FB pin. Now the FB node may be disturbed. Could you measure SW voltage waveform to see if the device works stable? Thanks.

Hi, and thank you for the feedback.  I will look into the R2 and R3 FB distance as you suggest.  My logic analyzer is not online yet, I will have to get that going.  Would it be useful for me to give you multimeter readings on SW in the meantime?

Acknowledged

Thank you for your patience.  I am a scope newbie.  Please find SW line readings attached, and let me know if this is right or if I should take the reading differently.

Hi Malachi,

This looks not a SW waveform, you should measure by the probe from SW pin to GND. Or is there anyone in your company can help you? Thanks.

I thought that's exactly what I did.  I will double check everything.

OK. If it's real, I think the device is broken, pls change it and try again.

Shipeng Cheng said:

the device is broken, pls change it and try again

I'm going to need you to be more specific than that please.

Double checked connections and took a 2nd reading, using a much tighter time domain.  If this also looks "broken", please point me to an image with a correct appearance.

Hi Malachi,

Yes, this waveform is clear to show that instability happens. As your layout cannot change quickly, you can try to add a 10-47pF capacitor in parallel with R3 to see if any improvement. And I want to know when getting this waveform, how much load current is? Do you use an e-load? Is there extra output capacitors except you show in schematic?

Thank you for the quick reply, very helpful

There is no load on the circuit yet, that is coming next I have the load set up here.  I plan to get that measured by tomorrow.  I'll take measurements of existing circuit first, then add the cap and try it again.

The schematic describes the totality of what's on the PCB, no further output capacitors

Sure.

No load

200mA load

400mA load

Important note: Previous pictures were for Serial Number 002 part.  That died, so these and onward pictures are for S/N 001 part.

Hi Malachi,

According to your result, I have to say that your layout is not good and causing the device cannot work normally. You need to modify layout first.

If you want to do other part test on this board, a way you can try is add a 1-10nF in parallel with R2 to see if this device can work stable. And you can measure Vin to see if it's stable to check if something else make our circuit cannot work.

Thanks.

Acknowledged, and thank you for your input.

Just to restate, on this board (v1) testing with capacitor mod on R2 instead of R3 is the way to go?

I will commence layout revisions immediately.

I'm not sure if this can help, you can try it.

Thank you, I will prioritize layout first and v1 R2/R3 capacitor modding second

Please find attached a work in progress (v1.2).  Here's where I am going with the layout so far:

Please forgive the hard to read filling esp around TPS_SW.

Aside from aesthetic untidiness, any glaring issues?

I'd really like to move C2 and C3, maybe rotate U2 90deg to SW vias to like https://www.ti.com/lit/an/slva922/slva922.pdf.  Not going to, solving one problem at a time.

Hi Malachi,

This time the components placement is OK. An advice here is the wire from C4 to Pin6 can be wider. I suppose there will be a big GND plane and let me check again then. Thanks.

I'll widen that trace.  Thank you Shipeng.  Your input is very helpful and appreciated

There is a full GND plane, you are right.  Also 12V gets priority in a distinct layer, might give us more options for Pin3 vs C2, C3

Sure. Thanks.

I had to shift some things around.  Please have a look (v1.2.1).  Hopefully we're getting close!

In process of prepping board for fab

Here's the current state of the circuit (v1.2.2), overlaid on the rest of the part

Here's the updated layout v1.3 as per your suggestion.  Looking OK?

Hi,

The connection to inductor can be as wide as the inductor pad. Thanks.

How about this (v1.3.1)