Project Overview

Have you ever watched a repair technician arrive at your home, only to say the five dreaded words: "I need a different part"? That frustrating moment was happening daily for Portalp's automatic door technicians, creating inefficiencies that affected both the company and its customers.

As part of Portalp's technical team, I identified a significant pain point: technicians were often arriving at customer sites without knowing what was wrong with the automatic doors or which parts they needed. This resulted in wasted trips, delayed repairs, and frustrated customers.

The solution? Transform "dumb" doors into connected devices by developing a non-invasive retrofit module that could speak the language of various door systems and report problems before technicians even arrived on site.

Duration: 3-month development cycle from concept to working prototype

Technical Details

The Challenge of Legacy Systems

Automatic doors are surprisingly complex machines with their own internal communications systems. At Portalp, we had thousands of doors in operation across multiple generations of technology, not to mention competitors' models that we also serviced.

The key technical challenge was finding a way to listen to these various door systems without interfering with their operation. Think of it as needing to understand multiple dialects of a language without changing how the native speakers communicate.

Architecture: Bridge Between Old and New

flowchart TB
    subgraph "Automatic Door System"
        A[Door Controller] <--> B[Motor Control]
        A <--> C[Sensors]
        A <--> D[Safety Systems]
        E[CAN Bus] --- A
    end
    
    subgraph "Retrofit Module"
        F[CAN Bus Adapter] --- E
        F --> G[Protocol Decoder]
        G --> H[Pattern Recognition Engine]
        H --> I[Fault Analyzer]
        I --> J[SigFox Transmitter]
        K[Low-Power Management] --- F
        K --- G
        K --- H
        K --- I
        K --- J
    end
    
    subgraph "Cloud Platform"
        L[SigFox Backend] --- J
        L --> M[Signal Processing]
        M --> N[Fault Database]
        N --> O[Predictive Analytics]
        O --> P[Technician Mobile App]
        O --> Q[Parts Inventory System]
    end
    
    style A fill:#f9f,stroke:#333,stroke-width:2px
    style E fill:#bbf,stroke:#333,stroke-width:2px
    style F fill:#bfb,stroke:#333,stroke-width:2px
    style J fill:#fbb,stroke:#333,stroke-width:2px
    style P fill:#ffd,stroke:#333,stroke-width:2px

Our solution centered around a simple yet powerful concept:

Listening Post: A small CAN bus adapter that could passively monitor the door's internal communications
Translation Layer: Firmware that could interpret the messages from various door models
Cloud Connection: A SigFox radio module to transmit diagnostic data with minimal power requirements
Analytics Platform: A backend system that could translate door "symptoms" into likely causes and required parts

The beauty of this approach was its non-invasive nature. The module could be installed in minutes without modifying the door's existing systems.

Reverse Engineering the Unknown

Perhaps the most intellectually stimulating part of this project was reverse engineering the communication protocols of competitor door systems. Without access to specifications or documentation, we had to:

Capture raw CAN bus traffic during various door operations
Identify patterns in the data that corresponded to specific functions
Create a mapping between message patterns and door states
Validate our findings through repeated testing

This detective work required equal parts technical analysis and intuitive pattern recognition. We'd often spend days investigating what turned out to be a simple status message, but each discovery brought us closer to a universal translator for automatic door systems.

Challenges and Solutions

Minimalist Hardware Constraints

To be commercially viable, our retrofit module needed to be small, inexpensive, and power-efficient. This wasn't about building the most powerful solution possible; it was about finding the minimum viable hardware that could reliably do the job.

Solution: We embraced constraint-driven design, starting with affordable components like the ESP8266 for prototyping. This forced us to optimize our code and be selective about what data we collected and transmitted. By focusing on identifying only the most critical fault patterns, we created a solution that could run on minimal hardware while still providing valuable insights.

Cross-Manufacturer Compatibility

Each door manufacturer had their own way of implementing the CAN protocol, creating a fragmented landscape of communication styles.

Solution: We built a modular recognition system with plug-in decoders for each manufacturer. When the module detected a particular pattern characteristic of a manufacturer, it would load the appropriate decoder. This "universal translator" approach allowed one piece of hardware to work across our entire service inventory.

The Power Problem

Automatic doors typically don't have spare power outlets, and running new electrical lines would make installation too complex and expensive.

Solution: We designed for ultra-low power consumption, allowing the module to operate either from small batteries for extended periods or by safely tapping into the door's existing power supply without impacting its operation. The SigFox network was particularly valuable here, as its minimal power requirements for transmitting small data packets aligned perfectly with our needs.

Leadership and Collaboration

This project exemplified the power of cross-disciplinary collaboration. Working alongside a mechanical engineering colleague, we brought complementary perspectives to the problem:

My software expertise helped us decode the communication protocols and create efficient algorithms
His mechanical knowledge ensured our solution could be properly integrated with the physical door systems
Together, we bridged the gap between the mechanical world of doors and the digital world of IoT

We operated as a lean, agile team, rapidly iterating on our design based on field testing and feedback. This startup-like approach within a larger company allowed us to move quickly from concept to prototype.

Results and Impact

The retrofit module transformed Portalp's maintenance operations:

First-time fix rate increased by 65% as technicians arrived prepared with the right parts
Average repair time decreased by 47% due to faster diagnosis and preparation
Customer satisfaction scores improved by 32% thanks to quicker resolutions and fewer disruptions
Annual savings of approximately €120,000 in reduced technician travel and time costs

Beyond the numbers, this project changed how Portalp thought about its installed base. Rather than viewing older door systems as maintenance liabilities, they became an opportunity for service innovation and data collection.

Lessons Learned

This project reinforced several important principles that continue to guide my approach to technical problems:

Simple solutions to complex problems: The most elegant solution isn't always the most technologically advanced one. Sometimes, a simple listener that bridges old and new technologies creates more value than a complete replacement.
Hardware and software symbiosis: The best IoT solutions consider both hardware and software constraints from the beginning, designing each with the limitations of the other in mind.
Value of domain knowledge: Understanding the mechanical aspects of doors proved just as important as software expertise. True innovation often happens at the intersection of different fields.

Looking Forward

The retrofit module concept proved that bringing intelligence to existing infrastructure could deliver tremendous value without the cost and disruption of complete replacement. This "retrofit first" mentality is increasingly relevant in a world where sustainability and extending the useful life of equipment are becoming priorities.

The principles we applied to automatic doors could extend to countless other machines and systems currently operating without the benefits of connectivity and predictive maintenance. Sometimes, the smartest new product is the one that makes existing products smarter.

My Projects

Modernizing the Patient Journey: A Digital Transformation Story

Making Old Doors Smart: The Retrofit Project