Buck regulator using LM25085 with strange waveform at SW node

Can you post your schematic?  What switching frequency do you have programmed?

Hi Mr. John Tucker,

This is the schematic that I'm using. This was provided by the WEBENCH tool.

It was supposed to work at 386KHz, as I see in the WEBENCH project. But at the practice it is working at 162Khz.

But, this unknown phenomenon is not regular. As you can see in the picture that I attach now, there are times that the controller "cuts" the signal two times, and other times it "cuts" the signal three times. I'm applying 9V at the input.

As soon as I increase the input voltage, you will see that the TOn time will decrease, and you will see that the signal has only one "cut" or any, the switching frequency is being changing.

When I apply 30V, I see an 500KHz signal in the switching node.

I hope this can help you to understand my problem.

Thank you very much,

A.Paiva

Hi Mr. John Tucker

I saw that it was very difficult to understand the schematic.

I try with a pdf file.

Please let me know if you need more information.

Best regards,

A.Paiva3386.LM25085.pdf

Hi,

Can someone help me understanding what kind of phenomenon is happing on my circuit?

Regards,

A. Paiva

There are no values for R4, C6 and C8 are they populated?  LM25085 requires a minimum of 25 mV p-p rpple at FB for stable operation.  Can you measure the AC waveform?  What is the ESR of C9?

No, they aren't populated. I want to see if they really make some difference in the ripple (if they reduce the ripple).

The ESR is 425mOhms @100KHz.

This first waveform is with the input at 30V. (For all pictures the yellow waveform is the AC waveform of the output voltage, and the pink waveform is the AC waveform of the switching node). It is strange to see that the frequency is changing with the applyied input voltage. 

This one is with the input at 15V.

And this last one is for Vin= 9V.

 I would like to reduce this spikes that appears at the output with a amplitude of 3.2Vpk-pk.

Regards,

A. Paiva

Hi Mr. John Tucker,

Did you have found an explanation with all the info's that I provided to you?

Regards,

A. Paiva

I do not directly support that device and have no special insight to its operation.  I do read most posts to this forum, and if I see that there is more information required I will sometimes ask for it or make general comments.  I'll see if  I can direct your issue to the correct persons attention.

Hi,

Thank you very much.

Regards,

A. Paiva

Hello Almerindo,

I believe at present, there is not enough ESR in the output capacitor for the required ripple injection in this circuit and the circuit enters into a kind of short burst mode. As you have mentioned earlier, since you expect a nominal switching frequency at around 390kHz, I would suggest you add an external resistor of 0.6 ohms  in series with C9.

Regards,

Sourav 

Dear Mr. Sourav,

I've no 06Ohms resistor here, but I have used a 2Ohms resistor.

This scope is for a Vi=9V

This scope is for a Vi=15V

This scope is for a Vi=30V

As you can see, the frequency is still changing with the Vin. And the spikes amplitude has increased.

Regards,

A. Paiva