Begin by setting up your Firebase Realtime Database. Ensure you have a Firebase project created and your Realtime Database is configured. Make a note of your Firebase database URL and configure your rules to allow read access for testing. Always remember to secure your database for production environments.

Set up a Node.js environment to interact with Firebase. Install Node.js and create a new project directory. Run `npm init` to set up a new Node.js project. This environment will serve as the bridge between Firebase and Kafka.

Use the Firebase Admin SDK to access your Firebase Realtime Database. In your Node.js project, run the command `npm install firebase-admin` to install the Firebase Admin SDK. This SDK allows server-side access to the Firebase Realtime Database.

Create a Firebase service account and download the JSON key file. In your Node.js script, initialize the Firebase Admin SDK using this service account key. Here's a sample code snippet:

```javascript
const admin = require('firebase-admin');
const serviceAccount = require('path/to/serviceAccountKey.json');

admin.initializeApp({
credential: admin.credential.cert(serviceAccount),
databaseURL: 'https://.firebaseio.com'
});
```

Replace `` with your actual Firebase database name.

Install Apache Kafka on your local machine or server. Follow the official Kafka documentation to download and start the Kafka server and Zookeeper. Ensure Kafka is running by creating a new topic for testing purposes using the command line.

In your Node.js project, install the Kafka client library to produce messages to Kafka. Use `npm install kafka-node` to add Kafka support. This library will help you send data from Firebase to Kafka.

Write a Node.js script to read data from the Firebase Realtime Database and send it to Kafka. Use Firebase's event listeners to track data changes in real-time and produce messages to Kafka. Here's a basic example:

```javascript
const kafka = require('kafka-node');
const Producer = kafka.Producer;
const client = new kafka.KafkaClient({kafkaHost: 'localhost:9092'});
const producer = new Producer(client);

const db = admin.database();
const ref = db.ref('path/to/data');

ref.on('value', (snapshot) => {
const data = snapshot.val();
const payloads = [{ topic: 'your_topic', messages: JSON.stringify(data) }];

producer.send(payloads, (err, data) => {
if (err) console.error('Error sending to Kafka:', err);
else console.log('Data sent to Kafka:', data);
});
});
```

Update `'path/to/data'` and `'your_topic'` with your Firebase data location and Kafka topic name, respectively.

By following these steps, you can set up a direct pipeline from Firebase Realtime Database to Kafka without relying on third-party connectors. This setup will allow you to process data changes in real-time, sending them directly to your Kafka infrastructure.