LM2576HV getting damaged on battery power source

Please provide more details to help us investigate the root cause. Please attach your schematic, BOM and layout pictures if possible.

What is the maximum load current you are running at? How much current is the device start up into?

When the part is demaged, can you identify which pins are short together?

It would also help if you can capture scope shots on VIN, VOUT and pin 2 of the devices.

Regards,

Yang

Hello Chirag,

Please send us your complete design schematic with pdf images of the layout. With your max vin being so close to the max allowable voltage at the input of the IC, any sustained spiking could cause damage to the IC. You could probe the input pin to see what kind of waveform you are getting. Also open the bandwidth of the scope while you are doing that measurement so as to see the accompanying noise.

Regards,
Akshay

Dear Mr.Yang,

        Max load current for the device is 100uA. When the part gets damaged, pins 1 & 3, pins 2 & 3 gets shorted. 

Regards,

Chirag

Dear Mr.Akshay,

        Have emailed schematic and layout pdf files on your personal email-id, please check and reply.

Regards,

Chirag

Hello,

I took a look at your layout. There are a couple of changes that are required, which should help the situation. I am also attaching the presentation on layout techniques for SMPS designs to avoid EMI. Please refer that in the future.

1) For a buck topology, the input side of the regulator contains a noisy loop. The discontinuous current flowing through the input, internal FET, and diode needs to be contained in a tight loop so as to avoid any unnecessary noise/spikes. This is done by tying the ground of the IC, diode and the input bypass cap on the same plane. You have the diode and the input bypass on the different plane from the IC.

2) If your input is close to 63V which is the absolute maximum rating, you should consider designing an input filter before the IC. This would help keep the input to the IC a little more clean. The input filter design is covered in the presentation I have attached. 

Please go through the datasheet for the layout suggestions also. 

Regards,
Akshay

Dear Mr.Akshay,

         Thanks for the suggestions. Have also been suggested to have electrical isolation for power supply input i.e battery source and regulator input, so adding additional isolation circuit can solve the regulators damage problem?  

Regards,

Chirag

I am not sure that I understand what you mean. Please send a block diagram.

Is your battery being charged by the AC source? The battery ground and regulator ground can be the same, but your input bypass ground should be right next to the regulator ground. 

Regards,
Akshay

Dear Mr.Akshay,

         Have attached block diagram for your understanding. My question was will isolating the battery power supply with the device reduce the spike if any on DC bus line which in turn will prevent damaging the regulator.

Regards,

Chirag 

One more question in datasheet it is mentioned that at output 3.3VDC, load current to be 0.2A ≤ILOAD≤1A for 4.75V ≤VIN ≤60V.

What does that indicate?

Regards,

Chirag 

Hello,

A well designed input side filter from your SMPS to the buck regulator should eliminate spurious spiking. Please refer to the EMI presentation I have sent before. Layout will also be critical.

I couldn't find the statement that says that at output 3.3VDC, load current to be 0.2A ≤ILOAD≤1A for 4.75V ≤VIN ≤60V. Please send me the page number.

Regards,
Akshay

Dear Mr.Akshay,

            The statement is there on page 5 of LM2575 datasheet, from my understanding it means the load current required should be minimum 0.2A, please correct me if wrong?

So what will be the performance if load current is less than 0.2A?

Regards,

Chirag