# Prepare a custom Balena OS image

## In this learning path

- [Introduction](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/)
- [Prepare a custom Balena OS image](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/2setup/)
- [Install Balena OS on AVH](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/3install/)
- [Deploy an application to your device](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/4deploy/)
- [Next Steps](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/_next-steps/)

## Sign up for Balena Cloud
The free Balena Cloud service manages IoT devices and applications. You can deploy Balena apps without Balena Cloud using the open source [OpenBalena project](https://github.com/balena-io/open-balena), but for simplicity this Learning Path uses the hosted service.

Navigate to the [Balena Cloud Sign up](https://dashboard.balena-cloud.com/signup) form and create a new account. The free account lets you manage up to 10 devices with no cost. This Learning Path requires only one device.

After you’ve created your account, you can create a fleet. A fleet is a collection of devices serving as a single target for deploying applications.

Create a new fleet and name it `AVH-Testing`.

## Download the Balena OS image
Add a device to your fleet. Navigate to the `Devices` section of your fleet and click on the `Add Device` button. This opens a dialog to create a custom device image.

![Add Device dialog window](https://learn.arm.com/learning-paths/embedded-and-microcontrollers/avh_balena/add_device.webp)  
_Add Device dialog window_

You can use the Raspberry Pi 4 device on Arm Virtual Hardware (AVH) as the new device.

Select `Raspberry Pi 4 (using 64-bit OS)` as the device type.

Select `Development` for the edition.

Leave the default values for everything else.

Download the image by clicking on the drop-down arrow on the right side of the `Flash` button and select `Download balenaOS`.

> **Note**  
> There’s no need to install Balena Etcher, the tool used to write SD cards.

The file is large so the download may take some time.

## Convert the download to an AVH custom image
When the download finishes, you can convert the Balena OS device image into a custom AVH image that you can upload to the AVH service and run on a Raspberry Pi 4 device.

The conversion can be done using a Linux shell script.

Download the script and save it as `mkbalenaosimg.sh` on your computer.

> **Note**  
> The conversion script requires a Linux environment with root access. If needed, copy the script and the Balena OS image .zip file to a Linux machine and log in to the machine. The `zip` and `unzip` commands are also required. Install them using your Linux package manager.

On a Linux machine, execute the script as root with the downloaded Balena OS image file as the only parameter:

```bash
bash ./mkbalenaosimg.sh balena-cloud-AVH-Testing-raspberrypi4-64-2.115.7+rev4-v14.11.2.img.zip
```

If successful, the conversion script creates a new file named `balenaos_rpi4b.zip` in the same directory.

> **Note**  
> If needed, copy the `balenaos_rpi4b.zip` file back to your original machine.

You are now ready to install Balena OS on a Raspberry Pi 4 in AVH.