Power consumption (AxelLite)

From DAVE Developer's Wiki
Jump to: navigation, search
Info Box
Axel-lite 02.png Applies to Axel Lite

Introduction

Providing theoretical maximum power consumption value would be useless for the majority of system designers building their application upon AXEL LITE module because, in most cases, this would lead to an oversized power supply unit.

Several configurations have been tested in order to provide figures that are measured on real-world use cases instead. Please note that AXEL LITE platform is flexible, so it is virtually impossible to test for all possible configurations and applications on the market. The use cases here presented should cover most of real-world scenarios. However actual customer application might require more power than values reported here. Generally speaking, application specific requirements have to be taken into consideration in order to size power supply unit and to implement thermal management properly.

Tests have benn performed using DXLH SOM (Quad core CPU) limited to 1 or 2 cores actives.

Set #1

The following paragraphs report the power consumption measurements for the following typical use cases, using different core #/clock speed configurations, detailed in the following sections:

  • U-Boot prompt
  • Linux prompt
  • CpuBurn
  • GPU test
  • VPU test

Testbed

Measurements have been performed on the following platform:

  • Carrier board: AxelEVB-Lite Adapter + AxelEVB-Lite + DACU
  • System software: XELK 2.0.0
  • Power monitor: a custom current/voltage/power probe has been connected to the power rail of the SOM (for more details please refer to the net 2V8-4V5_SOM of the AxelEVB schematics).

Use cases results

The following table shows the summary result of data reported by INA226 at different checkpoints.

Checkpoint 1 core (mW) 1 core (mW) 2 cores (mW) 2 cores (mW)
800MHz 1GHz 800MHz 1GHz
U-Boot prompt 2100 2100 2100 2100
Linux prompt 2300 2300 2300 2400
Cpuburn 2750 3000 3250 3850
GPU 2850 3100 3375 3950
VPU 3500/3800 3800/4000 3900/4200 4700/5000

Additional notes and reference:

  • U-Boot uses a single core @ 800MHz
  • At Linux prompt, a fixed clock frequency has been set (userspace governor)
  • The number of cpuburn instances is equals to the number of cores
  • burnCortexA9
  • GPU test: /unit_tests/gpu.sh
    • it's a Freescale script for GPU testing (video in overlay)
  • VPU test: gst-launch -vvv playbin2 uri=file:///home/root/big_buck_bunny_1080p_surround.avi
    • decoding of a H.264 1080p30 video

Standby

The following paragraphs report the power consumption measurements when the system is in standby state.

Testbed

Measurements have been performed on the following platform:

  • Carrier board: AxelEVB-Lite Adapter + AxelEVB-Lite + DACU
  • System software: XELK 2.0.0
  • Power monitor: INA226

Use case results

Voltage Average current (mA) Power absorption (mW)
3.3V SOM 51,7 170,61

Additional notes and reference:

  • The system enters standby after launching the following commands:
# set wakeup source
echo enabled > /sys/devices/soc0/soc.1/2000000.aips-bus/2000000.spba-bus/2020000.serial/tty/ttymxc0/power/wakeup
# enter standby
echo mem > /sys/power/state