High availability starts with strong design
The term availability is frequently interpreted following the narrow definition of the time between in-service faults. But a much wider definition around all the efforts to keep the transmitter on air is more appropriate.
That starts with the transmitter design itself. Every component from resistors up should be selected for performance, but also according to the very specific demands of a transmitter and, very important, for the predicted operational lifetime of that transmitter.
Secondly, the thermal design needs to consider the component’s specifications and to eliminate hotspots and thereby thermal stress to the components mentioned before. The cooling system itself must also be considered for stability and security: for instance, the TH1 is fully liquid cooled, which means 95% of all generated heat is dissipated directly to the outside rather than requiring expensive and potentially maintenance-dependent air conditioning.
Thirdly, intelligent design, meaning the avoidance of single points of failure, further contributes to an excellent availability. So, a single hardware fault should not lead to functionality failure. Better to have resilient transmitters, which can continue operating and delivering acceptable output even in case of a major failure.