Can someone please help me find why the LM2596 is blowing up on my PCB ?

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
Only the LM2596 which provides 5V output is blowing up not the other one, Why ?

Here's the schematic and layout for your reference
LM2596.jpg

1632759417801.png

Thank you much in advance :)
 
Last edited by a moderator:

Irving

Joined Jan 30, 2016
2,197
How did you arrive at 470uH inductors? These appear to be much larger then the chip manufacturer's recommended value range. And what spec are your diodes and output capacitors?
 
Last edited:

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
Thermal Junction Temperature. What is your input voltage, output voltage, and desired current output? What is your current available on input?
Thank you for your comment, My input voltage is between 12 to 13.5 V, I have two output voltages one dc-dc each, 5V and 8V, the 5V one is blowing up on reverse polarity. Desired current is 250mA to 500-600mA.
 

Orson_Cart

Joined Jan 1, 2020
76
Good call on D4 & D2 - these should be low loss schottky types rated thermally for the current

But D-oh - if you reverse the input of course it will die - put a low loss shottky in the input line .....
 

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
How did you arrive at 470uH inductors? These appear to be much larger then the chip manufacturer's recommended value range. And what spec are your diodes and output capacitors?
Thank you for your comment, i took 470uH from the dc-dc module i have used and seen many times, i have never seen an LM2596 module with the recommended value as per the datasheet, so i went with the one i have used and actually seen.

I am using SK34A diode, before i was using SS34, checked for reverse polarity and it seemed to have worked just perfectly. Is there a big difference between the two diodes which might be causing the issue ? If so then why only one of the IC's not both?

Output capacitors specs: -45℃~+105℃ 35V - 2000hrs 105℃ 220uF 8mm ±20% 10mm SMD,8x10mm Aluminum Electrolytic Capacitors - SMD ROHS
 

Attachments

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
likely one of 2/3 reasons, Vin too high, too much soldering is killing it, you bought the parts from Ali-express,

Could be all 3.
Thank you for your comment, My Vin is approx. 12V never below 12V and never above 147V, nothing has been purchased from Aliexpress, everything came in assembled from JLCPCB but in my design i had added an SOD-123FL diode (SM4007PL) but the diode was heating up way too much instantly causing the IC to heat up as well, so i de-soldered it and replaced it with SS34 (package - SMA-DO-214AC) from an LM2596 DC-DC module, it worked perfecty without any heating issues at all. So i ordered some SK34A (package - SMA-DO-214AC) from RS Components since i couldn't find SS34.
 

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
Thank you for your comment, My Vin is approx. 12V never below 12V and never above 147V, nothing has been purchased from Aliexpress, everything came in assembled from JLCPCB but in my design i had added an SOD-123FL diode (SM4007PL) but the diode was heating up way too much instantly causing the IC to heat up as well, so i de-soldered it and replaced it with SS34 (package - SMA-DO-214AC) from an LM2596 DC-DC module, it worked perfecty without any heating issues at all. So i ordered some SK34A (package - SMA-DO-214AC) from RS Components since i couldn't find SS34.
Sorry not 147V just 14V
 
Are you chokes any good? if they have a fault or are too small - this may cause overheating - but the real
LM2596 has internal over temp protection - hence the possibility of fake parts.
 

Irving

Joined Jan 30, 2016
2,197
The 470uH is much too large. For 12v to 5v at 250 - 800mA, 47uH would be better. The higher inductance is generating large voltage spikes and forcing the switching speed down which is stressing the chip, despite its internal overheat limiting. I think it only luck the 8v regulator hasn't died yet. Do you actually know the current draw when failure occurs?

As you already know, the original diodes were completely wrong for this purpose. The new ones are better...

Your PCB tracks to the chip are on the thin side. What's on the back side of the PCB? There's probably not enough heatsinking.

Do you have a link to the output capacitor datasheets? Or the specific parts you purchased?

Edit: The indicators you may have seen elsewhere could have been marked 470 but actually been 47uH. Did you actually measure one with an LCR meter?
 
Last edited:

Thread Starter

nishantnidaria

Joined Sep 27, 2021
34
The 470uH is much too large. For 12v to 5v at 250 - 800mA, 47uH would be better. The higher inductance is generating large voltage spikes and forcing the switching speed down which is stressing the chip, despite its internal overheat limiting. I think it only luck the 8v regulator hasn't died yet. Do you actually know the current draw when failure occurs?

As you already know, the original diodes were completely wrong for this purpose. The new ones are better...

Your PCB tracks to the chip are on the thin side. What's on the back side of the PCB? There's probably not enough heatsinking.

Do you have a link to the output capacitor datasheets? Or the specific parts you purchased?
You're right about the 470uH i got mistaken by the printed 470 on the inductor i guess, because even for 47uH it still says 470 -_-
For sure the original diode was totally wrong, it's too bad i only realized it later on.
I have used 1mm and 2mm traces for power and 0.8mm for the Mosfet because the pin doesn't accept more than that. and on the back side of my PCB there's just Ground copper fill, no components. It's a 2 layer PCB, on the front layer there's +12V copper fill with two zone +5V and +8V copper fill.

Please find the datasheet for the O/P capacitors in the attachment.
 

Attachments

Top