TPS22929 inrush current

This device does not include current limit or over-temp shutdown features.

Hi Hidetoshi,

Thanks for the additional info.  Starting with your second question, I'm guessing that measurement point A, is the red arrow labeld "1" on the first page of your attached document and measument point B is the blue arrow labeled 2.

The difference in the Iout waveforms is that A is before the output capacitor and B is after.  So the current measured at B is just the current through the load while the current at A will be the current at B + the current flowing through the output capacitor. 

This answers question 1 and explains why you see the spike at point A and not point B.  When the switch first turns on, the voltage differential between the input and output cause current to rush through the device and start charging the output cap.  Quickly thereafter, the circuitry that controls the output rise time senses the output capacitor is being charged too fast and kicks in.   Because the device does not provide current limiting, there will be some delay before the rise time circuit can react.

Regards,

Adam Hoover

DLS Applications Team

Hi Hidetoshi,

The type of capacitor will have an effect as well as the difference in capacitances.  Is the output capacitor the only thing that is different between these two waveforms?  Has the position of the capacitor relative to the device changed at all? 

Assuming yes, it appears the second peak is also being controlled by the rise time circuitry.  The rise time is an average measurement, which takes time.  The first spike occurs because it takes time to react to the current spike charging the output cap (as previously discussed).  But look at the output voltage amplitude and slope following the spike in the two waveforms.  In the second 60uF case, the larger capacitance creates larger inrush current spike, resulting in a higher output voltage compared to the 57uF case (the type of capacitor may come into play here as well). 

Also, the slope in the 57uF case is relatively flat compared to that of the 60uF case.   The rise time circuit detected this, and the adjustment resulted in the second peak.  Again, because this device contains no current limiting, the device does not react to the current spike, just a rise in the output voltage (with inherent delay).  Notice the slope of the 57uF case also increases following the second peak.

In short, the second peak was caused by overshoot of the rise time control circuit.  Do note that at the end of both captures the output voltages are almost exactly the same, meaning the rise time control is functioning as expected in both cases.

Kind Regards,

Adam Hoover

DLS Applications Team

Hi Hidetoshi,

In general, the maximum peak current of the switch is ~50% more than the continuos current (given the pulse is <300us at 2% duty cycle), for this device that would be ~2.7A.   This would also include the inrush current.  The 57uF and 60uF scope shots we discussed before should be fine here. 

Btw, if you are able to reduce the output capacitance relative to the input capacitance, this will also reduce the amplitude of the pulse and amount of inrush current.

Best,

Adam