TPS63901EVM: Current limiter

Hello Taner,

The expert for this part is out this week. But let me try to help to understand the issue.

Could you please send measurements of the current limit operation you are seeing and the peaks you observe when the output gets close to the expected output voltage?

Is it possible that the part is operating in continuous conduction mode while limiting and just pulsing when getting close to the 2.5V?

Best regards,
Brigitte

Hi Brigitte,

The figure shown below shows the transition from the 1mA current limiting setting to exceeding the 1mA current limiter.

Regardless of the mode I do not think the current limiter should exceed the setting that has been set utilizing the CFG pins. This would be weird. 

The average current while charging the cap is 1.14mA with a maximum peak of 1.31mA (left thick line). When the supercap reaches the 2.5V output voltage treshold of the buck/boost. the average current is approximately 500uA. But the peak currents are 5mA (right side oscillation).

Hello Taner,

I expect that the output voltage is such close to the input voltage that the device is not able to permanently switch anymore. The details need to be checked by the expert next week.

Best regards,
Brigitte

Hi  Taner,

Thanks for considering Ti's device.

TPS63900 is an averange current limit device, it will not try to limit peak current for each switching cycle(in light load), this peak current is depend on your actual input and output voltage.

How do you meansure your averange input current? by a multimeter?

Regards

Tao

Hi Tao,

As I explained earlier the limiter seems to be working according to the datasheet before the supercap reaches the same output voltage as the buck/boost (2.5V). However, after reaching the voltage the limiter seems to stop working. The measurement of this effect has been measured with a power profiler. 

Hi  Taner,

Yes, i got your point. As explained, current limit of TPS63900 in light load is an average limit but not peak limit, in light load, device works in pfm mode as below. it consist two phase, switching and pause phase.  input current only can be observed in switching phase and goes to zero in pause phase. Then, seems the second half of your screenshot is make sense. But i am confused what happed in your first half screenshot. Will do bench test to check.

Regards

Tao

Hi  Taner

I dId some bench test and saw different performance from my side.

Bench test set-up.

input: short cable+ 100uF/10V MLCC 3.6Vin

output: EC load+CC mode

Device set-up::1mA current limit   target vout:2.5V

Screeshot as below  Channel1-Vout   Channel3--Iout   Channel4-input current.(5mA/div)

You can see that the peak current increase with Vout, when peak current is higher, pause time is longer to make sure averange input current is close to 1mA.

Actual peak current has big relationship with your input cable, input cap and your load features. But anyway, the averange input current is still close to your set up.

Regards

Tao

Hi Tao, 

I understand the point that You are making. I tested in this test setup as well (with a load). And this is indeed what i saw as well. 

However, the 'problem' occurs when there is a capacity parallel to the load (like in the first picture). Is it possible for You to test this as well ? 

So with a capacity that has close to the same capacitance and the same voltage.

Hi Taner,

Yes, will try with a cap similar with your. Please allow me a delay because i have several urgent case to take care. Next week is ok for you?

Regards

Tao

Thanks for your help Tao ! 

This resolved our issue and it explains what we are observing.

With best regards.

Taner 

Hi Taner,

You are welcome, i made a mistake and waste some time, my apologize.

Regards

Tao