SLWSTK6061A Datasheet, PDF (20/32 Pages) Silicon Laboratories – UG182: EFR32 Flex Gecko 2400/868 MHz Wireless Starter Kit

8. Advanced Energy Monitor
UG182: EFR32 Flex Gecko 2400/868 MHz Wireless Starter Kit
Advanced Energy Monitor
8.1 Introduction
Any embedded developer seeking to make his embedded code spend as little energy as the underlying architecture supports, needs
tools to easily and quickly discover inefficiencies in the running application.
This is what the Simplicity Energy Profiler is designed to do. It will in real-time graph and log current as a function of time while correlat-
ing this to the actual target application code running on the EFR32. There are multiple features in the profiler software that allows for
easy analysis, such as markers and statistics on selected regions of the current graph or aggregate energy usage by different parts of
the application.
8.2 Theory of Operation
The Advanced Energy Monitor (AEM) circuitry on the board is capable of measuring current signals in the range of 0.1 µA to 95 mA,
which is a dynamic range of alomst 120 dB. It can do this while maintaining approximately 10 kHz of current signal bandwidth. This is
accomplished through a combination of a highly capable current sense amplifier, multiple gain stages and signal processing within the
kit's board controller before the current sense signal is read by a host computer for display and/or storage.
The current sense amplifier measures the voltage drop over a small series resistor, and the gain stage further amplifies this voltage with
two different gain settings to obtain two current ranges. The transition between these two ranges occurs around 250 µA.
The current signal is combined with the target processor's Program Counter (PC) sampling by utilizing a feature of the ARM CoreSight
debug architecture. The ITM (Instrumentation Trace Macrocell) block can be programmed to sample the MCU's PC at periodic intervals
(50 kHz) and output these over SWO pin ARM devices. When these two data streams are fused and correlated with the running appli-
cation's memory map, an accurate statistical profile can be built, that shows the energy profile of the running application in real-time.
At kit power-up or on a power-cycle, and automatic AEM calibration is performed. This calibration compensates for any offset errors in
the current sense amplifiers.
LDO
AEM
Processing
EFR32
Peripherals
Figure 8.1. Advanced Energy Monitor
silabs.com | Smart. Connected. Energy-friendly.
Rev. 1.00 | 19