What is TSN and do we really need it?

Two of the most significant innovations in industry since the 2000s have been the consolidation of fieldbuses into just a handful of open-vendor offerings (although some could argue there are still too many!) and the almost universal acceptance of Ethernet as an industrial networking standard.

That every fieldbus now utilises Ethernet is truly significant, as automation applications demand real-time performance for time-sensitive messaging — something that standard Ethernet was never designed to provide. Fieldbus designers have therefore had to adapt Ethernet to make it operate in real time.

Their hard work has paid off as the vastly improved fieldbuses we now enjoy can be attributed to Ethernet. Devices from different vendors can now communicate to each other, in real time, over a high capacity and relatively low-cost network.

Improving Ethernet?

While real-time Ethernet has been successfully employed in C2D (controller-to-device) fieldbuses, the same could not be said for C2C (controller-to-controller) connections, where there’s no predominant standard that can provide real-time capability.

Work has however been underway since 2012 to add real-time capability to standard Ethernet. It goes by the name of ‘TSN’, meaning Time Sensitive Networking.

The work, which is still ongoing, is being undertaken by the IEC/IEEE 60802 joint working group. It centres on Ethernet’s data-link layer, the second of the seven-layer ISO communication model. Many different standards have been incorporated, mostly within the IEEE 802.1Q umbrella, to provide functionality such as time synchronisation, scheduling, traffic shaping and the selection of communication path.

Significantly though, TSN is not a new application layer protocol. Rather, it’s an underlying technology; a transportation mechanism utilised by existing higher-level protocols to give them real-time capability.

The first steps

The OPC UA FLC (Field Level Communication) initiative was founded in 2018. It’s owned by the FLC group, a subgroup of the OPC Foundation (a vendor neutral organisation). More significantly, the FLC initiative has the backing of all major industrial automation players. Both are key ingredients for a standard to be successful in the marketplace.

By late 2022, the first phase of specifications was released. Called ‘OPC UAFX’ (Field eXchange), it extended the OPC UA standard to support real-time C2C connections. Work on the second phase started this year, focusing on C2D connections, although the first standards are not due for release until 2027.

One of the stated goals of TSN is to have one monolithic, unstructured network for all industrial communications. This would mean all nodes —everything from field sensor to the cloud — would have universal access to all data. This would replace the hierarchical ‘pyramid’ structure that is strongly favoured by virtually all automation providers today. While we don’t have time to cover details, such an approach would be highly problematic, with no clear lines of responsibility, questionable performance and every node needing to manage its own security, to name but a few pitfalls.

The proof of the pudding…

Whether or not TSN becomes an industry-wide standard remains to be seen. The demand from users for an interoperable communication system between controllers that can provide real-time data is certainly there. Also, that all major vendors are behind OPC UAFX will undoubtedly aid its promotion.

Working against TSN is its complexity — the underlying standards it relies on are vast and their application difficult. TSN also requires significant new networking infrastructure and software, adding to the cost of implementation.

One thing that’s become clear is that industrial applications need ever more real-time data, while most networks are already running at or near capacity. Some vendors have responded by trying to extract greater bandwidth from their existing networks, with or without TSN. They do this by increasing Ethernet’s bit rate from the standard 100 Mb/s to 1 Gb/s and beyond.

While it’s still early days for TSN, uptake so far has been sluggish, with vendor offerings limited and fragmented. This is despite the ‘noise’ that we sometimes hear from the marketplace.

*Harry Mulder is the Principal Automation Engineer at Beckhoff Automation. He has been involved in industrial automation for over 30 years and is fascinated by how new innovations keep affecting the direction of the industry. He really enjoys the practical element of his job, where he has a chance to get his hands dirty!

Top image credit: iStock.com/Vertigo3d