This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS62671: Switching frequency during transients

Part Number: TPS62671

Champs,

So I have a customer trying to evaluate the TPS62671; as there is no EVM they chose a TPS62621EVM and swapped the parts out; which I think is okay.  I am still trying to assess if they have (or could) swap the inductor for a 470nH as the 1uH on the '621 EVM is a tad high; BUT, in lieu of that information at the moment; what they are seeing looks a tad odd and I'd like an opinion here.

The condition is, 3.7V in, 1.8V out (TPS62671 is fixed output); mode = low, 180mA step load applied while running (eg, was in PFM mode, now switching to PWM mode).  So when they do this, it 'works' and stays in regulation, but what we see is the part starts switching at 14MHz.  Now there's a section in the datasheet (10.3.1) that states the limit is 'about 10-12MHz' though in addition to having to equate about to 15% to account for this; it doesn't seem to suggest that transients is the place where such deviation from the 6MHz nominal switch frequency would occur; plus there's nowhere in the datasheet that shows switching at this rate.  We see this last for 700us or so (which is long); then it slows and ultimately does come back to 6MHz.  The output stays in regulation; so we don't specifically see a problem; but this does seem way outside the expected behavior so we are curious if something else is going on here that we should be worried about.

In parallel, I have asked them to retest this with a 470nH inductor and 2.2uF cap, which the TPS62671 calls out as being tuned for; though to be honest this seems like it would only serve to suggest switching even faster in this condition not slower; unless the issue is somehow related to some noise path that's messing up the error amp inside the device.  But here's what it looks like today; would love to hear any opinions you might have on what's going on.

  • Hi Dave,

    I don't see expect any concern as the switching frequency is > 10MHz ~ 12MHz during transient load because the it is not guaranteed to be within this range. On the other hand, it's recommended to use the maximum no. of points in the oscilloscope to make the measurement more accurate especially when  you zoom-in the waveform.

    As stated in the datasheet (10.3.1, 2nd paragraph), when the device is in high duty cycle (during load attack), the device will operate more into "constant inductor peak current" mode hence if the inductor is change from 1.0uH to 0.47uH, I'm expecting that the switching frequency will further increase to provide more energy to the output during step-load.

    Let me know if you have further query.

    Best regards,

    Excel

  • Hi Dave,

    I will close the thread if you don't have any further query.

    You can open it again anytime by replying to this thread.

    Best regards,

    Excel