DOES NOT USE FET Switching architecture

• If you have tried other current measurement devices you may have seen "artefacts" in the measurement.  All other current measurement equipment use multiple shunt resistors and FETs to select which shunt resistor to use based on the magnitude of the current.  Switching between FETs is susceptible to charge injection from the Gate-Source capacitance of the FET.  That charge injection causes artefacts in the measurement.  There is no way to fix that.   FET switching architectures must also have an algorithm that must "predict" the current in order to switch the FET(s) into the right range.  Some current profiles can cause such algorithms to "thrash" between  ranges.

Internal to the P1150 there are 10 resistive loads, 2M, 200K, 20K, 2K, 400, 200, 100, 40, 20, 10 Ohms.   During calibration these loads are used to create an expected current.  From that expected current, offset and gain correction is computed.  The resistors that are used in calibration are 0.1%.  The ones that are not are 1%.

In addition the Output Voltage is also calibrated with a dedicated high precision 16bit ADC.

Calibration is done when the device connects to the PC.

Calibration can be repeated using Menu->Calibrate.

From a cold start at room temperature the P1150 raises in ~15DegC temperature.   Although I haven't seen much temperature change, you may want to re-calibrate once the P1150 gets up to running temperature.  The P1150 temperature is displayed in the GUI.

Although the EXE is signed, it is signed with the lowest kind of certificate which was the least expensive at the time I bought it.   Windows pops up a dialogue asking the user to trust this app.  Apparently, over time, Windows observes that the P1150 GUI is trusted by a growing number of users and will trust the application.  

I need to buy the "EV" level of certificate to avoid this headache.

The P1150 is a low source resistance power supply.  If the DUT pulls high current, the P1150 will attempt to deliver that current.   A common DUT design architecture is to have an LDO power a subsection, and enable/disable that LDO for battery current saving.  That circuit subsection, powered by the LDO, may have large decoupling caps, that LDO will attempt to power up those caps.  At time zero, the current into a cap is "infinite".  This is called "in-rush" current.

When a real battery is used, it cannot deliver "infinite" current, because a battery is a chemical process, and has significant source resistance.

The P1150, being a low source resistance power supply will show you the "in-rush" current problem you didn't know you had.

In-rush current(s) are notorious for causing "random" resets in your DUT.  Especially when the battery is low in capacity, cold, or "aged", or all of the above.  With the P1150 Aux A0 input you can monitor critical power supply rails and see the brown out that in-rush currents can cause.  Many microcontrollers have a "brown-out" detector on their VCC supply and will reset.  Some microcontrollers, like the STM32 series, have a reset reason register that you can check on boot.

How to fix in-rush currents?  Ideally its a hardware change, to use an LDO (or SMPS) with a "soft-start" capability.  Often a hardware change is not available.  One thing you can try is to "PWM" the enable signal of the LDO (or SMPS).  By toggling the enable pin at ever increasing duty cycle, you are "charging" up those decoupling caps slowly.  The P1150 has helped many designers tune the toggling of the enable pin.

There are two approaches, one with the GUI and the other using the Python driver and writing a script.

Regardless of the method you choose, there will be a lot of data exported to a csv file, and plotting that data may prove to be a challenge.

• There are a number of csv tools that can plot data, one for Windows is called "Flow CSV", which some have had success with.

The GUI approach, use the mAhr plotting tab

• Enable both Rolling and Stream checkboxes, set the window to be 1min (the default).

• Choose the decimation bandwidth, the default is 1KHz.  The higher the bandwidth the more resolution you will get, as well as more data.

• Note that the P1150's sampling rate is not changed when setting this decimation bandwidth, and during decimation the min/max current values are tracked along with the average current.  The Max current will track your targets maximum current usage over a decimation window, and not be averaged out.

The P1150 Python script approach starts with the cloning the repo, the link is above.

• After getting the P1150 Python repo installed, extend one of the example scripts to create your exported data.