TPS562201DDCR short

Thread Starter

fpgamakemyday

Joined Sep 30, 2022
3
Hello AAC-Members,

I want to preface this by saying that I'm really new to PCB-Design.
I have designed this fairly simple PCB with a Texas Instruments TPS562201DDCR step-down voltage regulator. It is supposed to step-down 12V to 3.3V for a microcontroller.
After soldering just the TPS related circuitry and testing it out, however, it blew. I tried on multiple boards with the TPS flipped. (Couldn't, and still can't, figure out where PIN 1 is on these chips, I assume it is where the bar is located, but that's besides the point)

Can someone point out to me what I'm doing wrong? I quintessentially copied and slightly modified the example circuit in the datasheet of the IC (Page 12).

tps-related.pngtps-schematic.png
(Regarding the attached pictures; I disabled the back copper-layer for visual clarity, no footprint is placed there, only traces connecting Via's)

Thanks in advance and best regards
 

Attachments

Sensacell

Joined Jun 19, 2012
3,785
Look at page 18 of the data sheet.

Although you *might* get this to work, your PCB layout violates many fundamental rules for switch mode regulators.
These chips are high performance devices, they require proper layout to work right.
Your layout adds much parasitic inductance and resistance to the circuit, these extra impedances may cause the part to be unstable.

You cannot just "connect the nets and call it a day" with switch mode regulators.
 

Thread Starter

fpgamakemyday

Joined Sep 30, 2022
3
Hi fpg,
This TPS package shows the pin #1 Ident.
On the typical circuit it shows the Enable is connected via a 10K, yours is direct to +12V

Which part blew on test?
E
Hey ericgibbs,

the TPS blew, I did notice that the TPS had a 10K resistor going to EN on the example circuit.
But after checking the characteristics, the EN pin is stated to have typically 400K resistance to ground, as such I left the 10K resistor out, thinking that it's redundant in this case. Was that wrong?

Regarding the index area; I have seen that in the datasheet, however I'm not really sure what this is supposed to correspond to on the actual IC, here is a close-up of the TPS.
tps-close.jpg
I assume the bar indicates Pin 1, but just to be sure I prepared two boards and soldered the TPS to them in both orientations, neither worked.
 

ericgibbs

Joined Jan 29, 2010
21,442
Hi,
I would say the Bar is the 1/6 pin's end.
Checked your artwork looks OK for the tracking.

Your layout could be better.:)
E

EG57_ 1125.png
 

Thread Starter

fpgamakemyday

Joined Sep 30, 2022
3
hi fpg,
Look at this PCB image for an improved layout.
E

Added PDF image,

View attachment 301964
Look at page 18 of the data sheet.

Although you *might* get this to work, your PCB layout violates many fundamental rules for switch mode regulators.
These chips are high performance devices, they require proper layout to work right.
Your layout adds much parasitic inductance and resistance to the circuit, these extra impedances may cause the part to be unstable.

You cannot just "connect the nets and call it a day" with switch mode regulators.
I see, I have updated the PCB to feature a ground plane and adhere to the TI Layout guidelines as closely as possible on just 1 layer, could this work?
I also added the 10K EN resistor just to be sure.

Best regards
 

Attachments

Top