LM25117: Unable to get more than 7.8A before shutdown.

Hi Clayton,

Please forward the schematic for this design. The current limit setpoint of the LM25117 is based on the 120mV CS threshold and the chosen current sense resistance. Refer to equation 9 in the LM25117 data sheet.

Regards,
Tim

PSU Extract.pdfHi Timothy,

Thanks for the reply.

I have carried out the calculations for equation 9 and get the result of 24Apk, which should be way more than sufficient to cover 12Apk.

I have attached the PSU schematic for you to look at.

Getting desperate now!!!!!

Thanks for the help.

Hi Clayton,

Can you measure the CS resistor voltage to verify the correct signal amplitude? Note that additional resistance is obtained if the CS resistor is not soldered correctly.

Regards,
Tim

Hi Timothy,

I will do that.

At what current draw level do you want it measured?

I'm guessing, correct me if I am wrong, that if I set up a 6A load, I should get roughly half that which shall be present at 12A?

Clayton

Hi Clayton,

Assuming a 5mOhm shunt (current sense) resistnace, a 6A load should give you a 30mV signal (and the inductor ripple componnet is superimposed on that).

Regards,
Tim

Thanks Timothy.
I will check that out in the morning (it's 7pm here now!!)
Will let you know my findings.

Hi Timothy,

Did the measurement across Rsense. It was 30mV (avg) as you said it should be.
However, there are some very large what I would call, switching spikes (2Vp-p, very high frequency) apparent on the signal.
Would these be having a detrimental effect on the CS input?

Clayton

Hi Clayton,

Yes, the large spikes are likely the cause of the early current limit. Please send the layout for review. You may be able to add an RC filter to attenuate these spikes if they are high frequency. Also, please include the waveform.

Regards,

Tim

OK, will do.
Had my reservations about them.

6A Load Chart.pdfRsense @ 2A Expanded.pdf2A Load Chart.pdfHi Timothy,

Here are the waveforms requested.

Will need to get permission to send layout.

Hi Clayton,

Please send a scope waveform of the CS voltage at 20mV/div.

Regards,
Tim

Hi Timothy,

Will do.

In the meantime, here is a snippet of the layout used.

Btw, have tried overiding the 'hickup' mode, and this seems to have increased the output drive current to 10A!

Does this give any extra clues?

Here are the artwork/layout pics:-

Hi Timothy,

Having trouble getting a clean capture of the CS at 20mV/div.

Signal is very noisy, and it is difficult to extract the signal from the noise.

Averaging of the signal helps, but this then is not a true representation of what we are seeing.

Will upload it when successful.

In the meantime, we have a 12uH inductor fitted, and by connecting DEMB and RES to VCC, we can now get about 10.5A out, but the Vout is dropping quickly (expected, I think), and is sitting around the 10.5V level.

This has produced a much cleaner switching/stable signal (which is good), but it still has the large switching 'ringing' on the rising edge of the switching output.

We are going to try the CC/CV additions as per Page 32/33 of datasheet.

Regards,