LM2596: LM2596

Marina Matic

Part Number: LM2596

I am trying to figure out what caused a failure of 2 of my LM2596-5V parts where they developed low impedance between pin1 and pin 2 of ~49 ohms. They failed in two separate identical setups. Thank you

over 8 years ago

0 Frank De Stasi over 8 years ago

TI__Guru 73115 points

Please attach your schematic and layout and your input, output and load specs.

Thanks

0 Marina Matic over 8 years ago in reply to Frank De Stasi

Prodigy 100 points

Hello,

Here is my schematic and layout. The load is 1A combined to 3 ARM processors where one takes 0.5A and the other two 0.25A each.

Thank you

Marina

0 Frank De Stasi over 8 years ago in reply to Marina Matic

TI__Guru 73115 points

Your schematic and layout look good.

Based on this view, I assume those vias are going to a good solid ground plane.

You might want more ground plane on top for heat dissipation; I can not see the entire area.

I would take the FB input directly from C203 rather than the inductor. You can extend the

small trace you have now to tap into the (+) of C203.

However, these issues may or may not influence the problem you are seeing.

Does the damage occur at power up or after a while ??

Can you monitor the input voltage and switch node during a damage event??

Is there a possibility that the input voltage is spiking above the abs max of the device??

Or that some load is pulling Vout below ground??

These things could possibly damage the device.

0 Marina Matic over 8 years ago in reply to Frank De Stasi

Prodigy 100 points

Hello,

Thank you for reviewing my issue. The failure occurred well into testing - during the overnight final burn-in after a full day of setup and testing. It is not easily reproduced. The chance to get the event on the scope is small.

My power supply is well regulated and there are 4 more of these attached to the same Vin, one is 5V and 3 are adjustable +15V,-15V and +12V.

It is interesting that it has ~50R from pin2 to pin1 resulting in a DC voltage at the level of input signal presenting itself at the output (pin2 and of course far output and FB pin4. There are no oscillations. I was hoping to get a hint from that piece of information.

I keep staring at this block diagram from ON Semi.