Power Profiler Kit 2 Reading Low and High

[edmcnicholas]

Hello I have a board that I am trying to profile with the power profiler kit 2 and the readings seem very wild.  I have a device that I know the base configuration with the sensors on is about 100uA measured with a fluke multimeter.  The issue I am seeing is that I am getting current levels between 2uA and 30mA with the BLE on and I can not get a steady state current.  This is the case with 2 separate PPK2 win Ampere Meter mode and Source Mode.  In source mode the current spikes show over 100mA.  it seems that the auto sensing is a bit haywire.  I am using the latest release of the power profiler sw.

From my measurements I should not be going below 20uA even with the sensors off and I am seeing sub 1uA for several mS even with a rate of 100k samples per second.

[Einar Thorsrud]

Hi,

It looks like you are seeing the LDO refresh mode. When the device is consuming low power, the regulator will be turned off most of the time, and only be enabled for a short while to optimizse average current consumption. You will see that as a sequence of spikes when you observe this witha  PPK or power analyzer or similar. When you measure the idle current with a multimeter you will not see this spikes, but instead see the average current.

[edmcnicholas]

But I am doing this in ampere meter mode. Would that still be the case.  I know I have an rms in one state of 150ua and it is flat since it is a sensor front end.

[Einar Thorsrud]

Is the question how you can measure the avarage current during an interval? If so, you can select the range you want to measure and look at the average number (you get one both for selection and the whole window). 

[edmcnicholas]

What i am asking is that if i know the measurement should be about 100uA steady state without much fluctuation as it is an amplifier circuits quiescent current how do i get a proper reading when I see that most of the time it is almost 0 and the spikes are not at my average current.  if it is in 100k sample per second mode i would expect it to be properly sampling at that rate without just drastic fluctuations especially in ampere meter mode.

[Einar Thorsrud]

How do you know taht it should be 100uA steady state? The multi meter will not show you short fluctuations over time? If you have an oscillioscope you could use for instance a 10 ohm resistor in series with the supply and measure the voltage drop over that to measure the current consumption over time without the PPK. Do you see the same current consumtpion pattern then?

[edmcnicholas]

I have a fluke and all i am turning on is the amp circuit that i measured the current using the multimeter. THe fluke is actually quite quick and will show fluctuations on the bar graph at the bottom even if it does not show on the screen.

I also measured the voltage drop across the feed resistor into the amplifier that is 27.5uA so there is no way i should be seeing 3uA readings.  

[edmcnicholas]

I measured the device with several resistors and it was fine.  Will a boost regulator make the power profiler unstable?

[edmcnicholas]

The boost is operating in light mode so that might have some pulsing

[Einar Thorsrud]

edmcnicholas said:

I measured the device with several resistors and it was fine.

Did you meassure the voltage drop over a series resistor? Was that located in the same place in the cuircuit that  you measure current with the PPK2? Can you share the plot of the measurmenents you did over the resistor with an oscillioscope?

edmcnicholas said:

Will a boost regulator make the power profiler unstable?

That is a possibility, particularily if the switching frequency is very high. Which boost converter do you use?

[edmcnicholas]

I was measuring the power profiler with known resistance and that gives fairly steady data.  I could try using my scope  but i have just been looking at rms for worst case.

The regulator we are using is TPS61099DRVR the frequency is 1-3mhz.  There is a pfm light load mode.  When in boost or pass through I see voltage spikes that sag the input voltage by over 0.1v

[Einar Thorsrud]

I see. Then based on the current information I do not see a good reason to doubt the measurments from the PPK. But if you want to digg deeper, measuring the voltage drop over a series resistor with an oscilliscope would be the next step. I would expect to see the same pattern as you see in the PPK plot. If not, there may be a need to look more closely at the PPK measurements.

[edmcnicholas]

These cant be right as there is no way to be at 3uA with the circuit on. I do see alignment between pulses on the input voltage and current spikes match at 5.5ms intervals