Prologue
Hi, I am Antti from East-West Tech Solutions (EWTS). This is a review of some key phases of our IoT mesh development project:
Early 2017 in Shanghai, my friend Steven suggested that we should consider creating an air purifier product for a local market in China. Air purifiers are used to remove contaminants from the air in a room to improve indoor air quality – in our case new product’s purpose was to filtrate fine particles from air.
By mentioned time many people in China had already become aware and concerned about potential health impacts, caused by PM2.5. PM2.5 are fine particles (diameter ≤2.5 µm) in the air that reduce visibility and cause the air to appear hazy, but above all these can enter your lungs.
Earlier I had eagerly been participating and leading various international sales, product and working process development projects in other industries, but honestly – this task called for taking over totally new kind of challenges. Anyway, prospects of studying needs and expectations of a new market and its product offering, creating a new product for EWTS and ensuring good indoor air quality seemed personally and professionally inspiring and tempting.
Therefore, we gathered market data and negotiated about design and manufacturing cooperation, but later concluded that our negotiation power and resources were too limited to continue pursuit of a niche at air purifier market. However, during this process I had started to incubate an idea of shifting focus from a “purifier box hardware” on certain key electronics to measure air quality. Later this spin-off plan lead to starting a development project of an IoT mesh-system for measuring and monitoring indoor air quality.
Game of Two Halves – First Half
By summer 2019, envisioned key functions of the new system were as follows
- An IoT mesh system for measuring and viewing indoor air quality, and measuring units (nodes) to be equipped with a selection of high precision sensors
- Viewing and configuration of a single mesh system, or multiple systems enabled remotely.
- Any measuring system to be extendable to hot, cold and wet conditions, both indoors and outdoors. Some sensors set limitations to allowed temperature and relative humidity, and therefore we wanted to significantly extend operation range of our system. Natural choice for an external sensor was Ruuvi Innovations Ltd’s RuuviTag (ruuvi.com), which was introduced to market and had been inspiring me already since the early 2018.
- Effective, but low power consuming radio modules to enable smart mesh communication in public, industrial, commercial premises and homes. Nodes do not only route measuring data and your commands to other nodes across the mesh, but serve as “hosts” to external sensors and forward their measuring data also.
- Configuration options, including node hardware with a variation of selected sensors, multiple viewing options for real time, historic data and alerts, nodes’ measuring cycles and alert limits, and renaming of nodes and mesh systems
Purpose of these design targets was to make sure that our customers could have a good understanding on air quality around them, and in their desired locations. We contacted some customer prospects to introduce our project – and got positive feedback and requests, when our system could be ready for testing. We seemed to have a good plan and positive feedback from customers: what could go wrong?
Well, there were twists and turns in mobilization and implementation phases. In fact, quite many things were complicated: there were challenges to find suitable cooperation partners for design and manufacturing of hardware, development of software and firmware, printed circuit board (PCB) architecture didn’t belong to my expertise areas, lots of sourcing needed to be done to find key materials, such as suitable radio devices and sensors for the PCB – and then our system’s planned multiple functions called for lots of development, testing and modification work. I have a thermal engineering background, though, and wanted to do my best to ensure that at least a proper layout and enclosure design could minimize hot spot risks and ensure high quality measuring results.
What happened then? While nodes were being developed and tested at manufacturer’s workshop, I was able to keep an eye on development work, test configuration functions and monitor measuring results remotely. I had some ongoing concerns also, because prototypes kept on sending data reports of pretty high room temperatures.
Prototypes were finally shipped to Finland in spring 2021 after numerous software testing and modification rounds. In this phase mesh system had limited configurability, external sensor functions were excluded and node enclosure design was not finalized, but anyway we hoped to be able to start introducing an MVP version to our customers. Unfortunately my previous doubts were soon verified in new tests at home office: MCUs and radio modules created hot spots, which distorted measuring results of sensitive temperature and humidity sensors.
In earlier discussions customers had made it clear that they wanted to test and compare our sensors against their own high precision measuring devices. Sensors’ distorted measuring results could naturally have been tuned by using correction curves, but I didn’t want to compromise our design strategy: high precision measurements represented anyway a corner stone in our concept. Manufacturing partner of the completed MVP prototypes was not willing, or able to replace the MCU and create a new PCB architecture. Therefore, we had not only failed in a key feature of the prototypes, but our design was in dead end. First half of our development game had just ended in a painful face-plant.
Second Half
2nd design round started with transformation challenges. We needed to pursue major development leaps both in hardware and software: prototypes’ PCB and firmware to be rebuilt, mechanical design of enclosures prepared, and server software to be renewed. However, it was not only about technical improvements: we needed to create new partnerships to get our project successfully completed.
Pursuit of cooperation partners
- Key part of the PCB architecture of the 2nd round was a new SoC (System-on-Chip). Along with this, we needed to find a new partner company – preferably a turnkey supplier, possessing skills and strong references with the designated SoC – to design and implement the new hardware and firmware.
- Preparations for cooperation in a larger contract always require some special efforts, but based on my own experience, it’s a bit different to represent a large or a mid-size company, or a small company: in latter case both the Purchaser and the Supplier may need to focus more on weighing their strategic fit along with the project requirements. After negotiations with multiple companies – and some new twists and turns also, I finally decided to assign prototype hardware design and manufacturing, and firmware development contracts to two separate companies.
Technical challenges
- PCB’s detail and layout design were completely renewed, based on using a new SoC and some sensor upgrades.
- Mechanical design of enclosures was prepared. Along with this, we made provisions to offer enclosures made of carbon neutral and certified biomaterial in commercial phase deliveries.
- Completely new node firmware and control unit software were developed, cloud software enhanced – and as part of the software upgrading work, RuuviTag functions were finally enabled.
- Product’s brand design was complemented, and website development work finalized.
Now the Second Half of our development game is approaching a turning point: we are on the verge of introducing the beta-version of our IoT system to you!
Are you ready to sense fresh air with FresaFlo™?
Acknowledgements
Our IoT mesh system was named FresaFlo™ already in the starting phase of the project.
FresaFlo™ is a trademark, owned by EWTS. Product development has been based on Nordic cooperation: product concept was developed by EWTS, brand and website design, mechanical design of node enclosures, carbon neutral enclosure material, detail design of PCB, hardware manufacturing and server software have been developed by our partner companies in Finland.
Firmware of the nodes, as well as mesh communication software and hardware have been developed by our partner companies in Sweden.
I would like to express my warmest thanks to all our development partners. Along with this, I am most grateful for Ruuvi’s support and positive attitude towards our project!
Antti Soininen
Managing Director