Industrializing a Fleet Telematics Prototype: From 2-Layer PCB to 500+ Trucks

Client

A technology firm specializing in digital logistics and fleet data solutions, helping logistics businesses make data-driven decisions and optimize performance across their supply chains — monitoring parameters like fuel level, vehicle speed, and engine load in real time, and sending that data to the cloud for fleet management and diagnostics.

Problem Statement

The client had a functional prototype device and approached Techshlok to audit it, identify faults in the hardware and software stack, and build a production-ready system capable of large-scale deployment.

Solution

Techshlok proposed a hardware-software co-design approach, with a step-by-step plan to:

1. Redesign the hardware with a 4-layer PCB for better signal integrity and stability.
2. Implement an RTOS to handle concurrent tasks — data acquisition, queuing, and MQTT communication.
3. Add custom circuitry for over-voltage protection, reverse-polarity protection, and temperature-resilient operation.
4. Support fuel-level acquisition via direct analog wiring, integrated with CAN-based OBD data through a single-channel CAN architecture.
5. Redesign the enclosure and thermal layout to prevent SIM card failure in high-temperature environments.
6. Debug and refactor the existing firmware for modularity, efficiency, and long-term stability.

Prototype Design Review

Evaluating the existing hardware, Techshlok found the 2-layer PCB prototype suffered from runtime errors, inconsistent performance, and system failures when deployed in remote locations — increasing retrieval and maintenance costs and undermining production readiness. The prototype lacked the robustness, scalability, and thermal stability needed for real-world use across BS4, BS5, and BS6 vehicle platforms.

Hardware Redesign

The PCB was upgraded from 2-layer to 4-layer architecture, isolating power and signal layers to reduce interference and improve reliability. Automotive-grade protection components were integrated to handle input voltages of 100–120V, with reverse-polarity protection added.

Power Circuit Design

A robust DC-DC conversion system and relay-based switching architecture were designed for stable signal operation in harsh environments — letting the same port handle both OBD data and analog fuel data via separate input channels multiplexed through the CAN bus.

Software Optimization

Firmware was moved to run on RTOS with independent task management for reading sensors, managing queues, and transmitting data, with watchdogs, improved memory management, and filtered CAN messages to reduce processing load.

Communication Protocols

Real-time data transmission runs over MQTT via the in-device SIM slot, giving lightweight, low-latency communication with the cloud backend even in low-bandwidth regions.

Thermal Enclosure Redesign

SIM modules in the earlier prototype were failing at 60°C ambient temperatures. The enclosure was redesigned to incorporate a heat sink and passive airflow, mitigating overheating and protecting internal components.

Validation and Manufacturing Support

After successful internal testing, the new device was deployed across 500+ trucks operating in varied climates, with Techshlok also supporting the client’s production QA, assembly line setup, and component sourcing for scalable manufacturing.

Result

Techshlok transformed the client’s proof-of-concept into a fully industrialized, reliable, field-tested telematics solution. With real-time access to OBD and fuel data across BS4, BS5, and BS6 vehicle models, the client significantly improved its fleet data analytics capabilities:

• Stable real-time data capture and MQTT transmission.
• Increased compatibility with older and newer vehicle models.
• Full thermal protection for SIM modules in extreme temperatures.
• Reduced hardware failure and maintenance costs.
• A production-ready design with manufacturing support.

The revamped system is now a cornerstone of the client’s digital logistics offering, providing accurate vehicle insights and predictive diagnostics at scale.