TPS54040: TPS54040 input current at light load

Please supply a schematic of your design and how you are measuring your input current.

Hello Frank-san

Thank you for response.

Please see below schematic.

Regards

Tomohiro Nagasawa

TPS54040 Scematic.pptx

Hello Frank-san

I would appriciate if you could response.

Regards

Tomohiro Nagasawa

Tomohiro-san,

Sorry for the delay. It may be your multimeter is getting confused at the pulse-skipping current waveform. Can you check to make sure the circuit is operating properly at 10mA? Check that the output voltage is stable, check VOUT during a load step. Or you could take a bode plot since you have R5 already in place.

-Sam

Hello Samuel-san

Thank you for response.

I checked output current and waveform.

When  Iout=10mA setting Iout is slightly unstable.(9.76mA～9.79mA)

I attached Vout and input current waveform.

I guess that pulse skipping is mixed in DCM, but what do you think?

Vout_waveform.xlsx

Iin=Low(1.9mA) CH1:Vout CH4:Iin

TPS54040_Iin=1.9mA4.tif

TPS54040-_Iin_1.84mA時波形2.tif

Iin=high(2.5mA) CH1:Vout CH4:Iin

TPS54040_Iin2.54mA4.tif

TPS54040_Iin_2.53mA時波形2.tif

Regards

Tomohiro Nagasawa

Hello Sam-san

Sorry for delay response.

The issue is that the input current increases with two values. and not meet product specification.

This behavior is due to the pulse skipping operation being mixed with DCM and My understanding is that it is normal operation.

Is it correct?

Regards
Tomohiro Nagasawa

Tomohiro-san,

I want to make sure I understand what you mean by, "the input current increases with two values." Do you mean that in DCM/Pulse Skipping, the input current is 2.5mA or 1.9mA and this is not acceptable?

Your waveforms show the current ramping up and down depending on the switch activity. This is normal.

If you need a more constant input current, you can find an IC with forced PWM which will continue switching at the set frequency for light-load conditions. This will hurt efficiency but it will get rid of this input current swing.

-Sam

Hello Sam-san

Sorry for lack of information.

1.9mA input current is acceptable, but 2.5mA input current is not acceptable. Some devices are stable at 1.9 mA.

This is all normal behavior due to variation of pulse skip threshold in my understanding.

Regards

Tomohiro Nagasawa

Tomohiro-san,

Oh, I think I understand now. You have many boards. In the same conditions, some have input current of 1.9mA and some have input current of 2.5mA, correct?

If this is the case, you're right. There should not be this big of a difference between the identical board with identical conditions. I'd recommend checking each pin of each board on an oscilloscope, comparing the voltage waveforms between boards. Something should be different between the two. My guess is this is caused by either a wrong component placed on the board, insufficient soldering, or a marginally stable converter.

-Sam

Hello Sam-san

Thank you for quick response. Your understanding is almost correct.

In addition,input current is unstably 1.9 mA or 2.5 mA depending on the measurement timing in issue device.

The good device is stable at 1.9 mA.

Since the behavior follow the issue device, I think that it is due to the individual difference of good devices.

Regards

Tomohiro Nagasawa

Tomohiro-san,

Okay, my understanding is:

You have many identical boards. Some are good and some are bad.

If you have 10mA out, the good boards measure 1.9mA in at all times. The bad boards with the issue IC show either 1.9mA in or 2.5mA in depending on the timing of when you measure. The issue boards jump between 1.9mA and 2.5mA.

If you take a good board with a good IC, then take a bad board with an issue IC, then swap the ICs, does the issue follow the IC or the board?

-Sam