The reliability of a product which uses a wireless network to communicate can be a concern for many consumers. Inquiring minds sometimes ask us what happens to the 75F solution if the network is interrupted. Here are the answers to those very valid questions!
The Wi-Fi goes down, now what? Will my entire building automation system cease to work? Of course not! If a connection to the cloud is disrupted, the Central Control Unit (CCU) maintains a local cache of the past 24 hours of data, which it will use as a fallback for the next day’s plan. Typically speaking, every 60 seconds, data is streamed from the CCU to the cloud and is analyzed to create a control strategy for the next day. This is what enables our system to be predictive and proactive. We also keep a CCU monitoring service running on the server, which sends an alert if the CCU communication isn’t received every 15 minutes. This ensures the appropriate people are notified and in general, corrective action can be taken almost as soon as the connection is lost.
CCU to cloud server communication happens over a standard HTTPS protocol. Hypertext Transfer Protocol Secure (HTTPS) comes with the built-in reliability of being the mainstream protocol used for almost all protected communication happening over the internet. This reliability is enhanced by using a parallel real-time messaging system, which uses WebSockets to signal when data has changed. This ensures that there are two parallel channels running at all time – allowing for interactive, real-time communication to ensure a fast response time on apps.
Our CCU communicates with the WRMs via a wireless mesh network. In a mesh network, each node in the network can relay data to any other node in the network. All nodes cooperate, and if a node breaks, the remaining nodes will reconfigure themselves around broken paths using self-healing algorithms, finding a way to bridge the shortest path. This means that if one WRM goes down, it won’t stop information from flowing between the other WRMs to the CCU. The functioning WRMs will cooperate to reconfigure themselves around the broken path and will find an alternate route to send data to the CCU.
The CCU is unable to communicate with the WRM: Should all communication go down between the WRMs and the CCU, the system will still function in standalone mode. Essentially this means it will work like a regular, programmable thermostat. On their own, WRMs are still able to control temperature by modulating their attached dampers. They simply use a strategy which factors in the current temperature, set point, and airflow temperatures to determine whether the system is in heating or cooling mode. They are also still able to be manually controlled.
This failsafe also ensures that a WRM can never shut the damper below 50%. Even if all WRMs went into failsafe mode, there could never be a situation where more than 50% of the overall airflow was restricted, preventing back pressure build-up at the fan.
Meanwhile, on the other side of the network, the CCU is aware of when WRMs have become disconnected. It will store a localized copy of the data until a connection is restored.
The CM is unable to communicate with the CCU: The Control Module is connected to the CCU via a USB hardwired connection. This type of connection is extremely robust, but we have implemented an active heartbeat mechanism between the CCU and CM to detect when software processes are not working optimally. If this heartbeat were to fail, the CM would then control the HVAC system using a built-in temperature sensor and a set point, acting as a regular programmable thermostat until a connection is restored.
The CCU is unable to communicate with the WRMs or the CM: In the unlikely event that this happens, the CCU will try to reestablish communication by rebooting the WRM and CM. It can also be configured to send alerts to the facility manager, installer and support team. The CCU will follow these steps until the situation is resolved.
The CCU also has a back-up battery in case of a power failure. To preserve battery, it will follow multiple levels of power-saving modes – including dimming and then shutting down the screen, terminating less used background processes, and in a worst-case scenario, shutting itself down at 10% battery level to ensure the battery doesn’t go into deep discharge. In such a situation, the Control Module, again, would take the responsibility of controlling the HVAC system, acting as a regular thermostat.
Reliable data storage is part of our plan. Our backend as a service provider (Baas) utilizes hourly backs-up of application and configuration data, and retains these backups for up to 30 days. To be safe, we also do daily back-ups of your data in two additional places: the CCU and a separate service provider. This ensures that at any point, there are at least three copies of configuration data available – so in case the CCU configuration becomes corrupted for any reason, a restore can be done immediately from within the CCU by the press of a button, ensuring complete reliability and almost zero downtime.
75F is a predictive HVAC system and we’ve worked to ensure that our network, encryption and data management strategies are just as innovative.
Downtime costs money and energy – two things we’re all about saving. By using a sturdy network, levels of data storage, regular backups, and numerous failsafe plans, we’ve done everything in our power to keep our system working round-the-clock.