Skip navigation
> RoadTest Reviews

AVNET Azure Sphere MT3620 Starter Kit - Review


Product Performed to Expectations: 10
Specifications were sufficient to design with: 10
Demo Software was of good quality: 10
Product was easy to use: 10
Support materials were available: 10
The price to performance ratio was good: 10
TotalScore: 60 / 60
  • RoadTest: AVNET Azure Sphere MT3620 Starter Kit
  • Buy Now
  • Evaluation Type: Development Boards & Tools
  • Was everything in the box required?: Yes
  • Comparable Products/Other parts you considered: Regarding the Microcontroller itself probably a Beagle Bone or a Raspberry Pi can do the same type of events handler. But regarding security this integration with the Azure Sphere makes this device a very secure system.
  • What were the biggest problems encountered?: The learning curve was a bit long, there are a lot of steps to get used to this device. But after some tutorials it works like a charm. The mikroBUS interface was something new too. The easiest way to use are the click boards available, otherwise some deep study of the protocol will be needed. It will be helpful for people more familiar with high level languages the release of a .Net support for this kit.

  • Detailed Review:

    Goal of this read test

    The major goal of this review was the starter kit test itself, using the tools and the hardware it provides, but the biggest curiosity was around the Azure Sphere and the Cloud communication functionality itself.


    Kit Contents

    The kit provides us the board itself and the usb cable to connect it to the computer


    Starter kit image


    The board itself

    This is very complete board with some out-of-the-box sensors and buttons to play with like the user buttons and leds and the integrated sensors (3D accelerometer, 3D Gyro, temperature sensor, and an ambient light sensor). And for 3rd party integrations we have the Micro BUS, UART and the  I2C Grove connector.

    More details can be found at the product page or here at the community read test page.

    Board details

    The SDK, software tools and tutorials

    The Azure Sphere SDK installation is quite easy and the initial steps to start playing with it can be easily executed following the documentation provided by Microsoft at this page the sdk can be used with Visual Studio 2017 or higher.


    Running locally

    Everything runs smoothly and you will easily get the initial application prototype running in your systemising the Azure Sphere Blink template, as you can see below with this sample template is possible to start interacting with the board

    Code screenshot Blink app

    I found some very good tutorials here in the community blog too, so there is another more complex local running sample in this tutorial with the result below where is possible to watch the status of some of the board sensors

    Local app demo

    Running at the cloud

    Still using the tutorials in the community blog mentioned above is possible to interact with the board using Azure IoT resources like IoT Hub and Time Series Insights (I was not allowed to test Time Series Insights since my Azure account free credits were already expired) and the IoT Central


    Testing the IoT Hub

    The steps to use the IoT Hub can be found in this tutorial, after all the steps are done is possible to use the "Device twin" Azure functionality where eis possible to send instructions to interact with the board like below

    Azure IoT Hub

    As you can see above changing the JSON code inside the desired actions at the IoT Hub I possible to interact with the board leds, click boards, etc.


    Testing the IoT Central

    The steps to use the IoT Central can be found in this tutorial, with the IoT Central the interaction is way better since it supports measurements and properties output data from the board into the web interface and is possible to manage the board settings via the web interface too, as shown below:

    MeasurementsSettingsRed led on like requested in Settings


    This is a very abstract starter kit like I mentioned before, it has a lot of functionalities to explore, which makes it very powerful but requires some learning curve either. More time will be needed to explore other capabilities like using 3rd party sensors and actuators or even other systems. I'll try to keep the community posted as far as I can advance on that field.


Also Enrolling

Enrollment Closes: Jul 10 
Enrollment Closes: Jul 28 
Enrollment Closes: Jul 20 
Enrollment Closes: Jul 17 
Enrollment Closes: Jul 14 
Enrollment Closes: Jul 10