The computer generated picture shows a view of the unfolded outer wall of JET through the eyes of infrared cameras. Green, turquoise and purple areas represent existing views of different quality, red earmarks the planned system.


JET is currently in shutdown to maintain and refurbish the experiment and to install the new ITER-Like Wall. The new wall requires an accompanying programme to establish operational procedures such as to avoid damaging the tiles.

That is why in October 2009 the “Protection of the ITER-Like Wall” (PIW) project was launched. It includes the development of improved temperature measurements, real time analysis and adjustments in plasma and heating system control software. The JET operator has taken on most of the tasks and a call for proposals has been sent out for additional expertise from the associations.

Paradigm shift in protection

The ITER-Like Wall being installed in JET’s vessel changes the main characteristics of the experiment. The new tiles are made of beryllium or tungsten instead of carbon, in order to avoid the drawback of tritium retention by the formation of hydrocarbons. These metal tiles are, unlike carbon, prone to melt when exposed to excessive local heat loads. In addition, most of the tungsten tiles in the divertor are in fact not bulk tungsten, but Carbon Fibre Composites (CFC) coated with tungsten, which can crack and flake off. Such damage would expose the carbon.

In JET’s former carbon age protection of the experiment was highly developed. Since April 1997 we have not had to open the vessel to recover and replace damaged or dislocated tiles. To maintain this high level of safe and efficient operation, additional protection strategies are being developed for the new quality type of wall materials. In general the protection software is based on simple, validated diagnostic data and plasma models. The protection is fairly draconian: If the software detects characteristics of plasma disruptions for example the experiment is stopped immediately. On the one hand, some erosion of the wall is accepted to protect against more severe events like disruptions and coil overheating. In these cases the plasma is extinguished deliberately on the inner wall. On the other hand, each “stop” represents a risk in itself and must be avoided as far as possible.

Advanced stop sequences are being developed to protect JET’s new metal wall. These stops involve a degree of control not previously required. In other words, there will be an additional layer of control software running on top of the old one. The new protection philosophy is key to achieve the given experimental goals and moreover to minimise down time. In an early stage of the ITER-Like Wall programme, it will be essential to avoid exposure of the bulk Carbon Fibre Composites (CFC) due to erosion or melting of the tungsten coating. Diagnostic data and plasma models to support this protection philosophy are more sophisticated and thus inherently less robust.

Detecting temperatures

In addition to the data based on observation, the protection includes surface temperature measurements. With the help of several infrared (IR) cameras a stop is initiated in real time if an area is about to reach a predefined limit. These limits are defined by beryllium melt thresholds, recrystallisation of bulk tungsten and damage to tungsten coatings.

A conservative approach with lower limits will probably be adopted initially, and raised later to 1,200o Celsius in most areas and potentially 2,200o Celsius for some of the divertor tiles much later in the programme. The IR cameras foreseen observe about half of the divertor and limiters, and all of the antennae in the vessel. This is sufficient since the surface temperature is expected to follow a repetitive symmetry. To save money and time, cameras are fitted, wherever possible, to existing periscopes. For the same reason the cameras are of a different type from the far IR cameras presently used for physics studies. Instead simple cameras are used that are only sensitive in the near IR. This choice has some consequences for operation as these cameras have a low dynamic range: As a result the images are mostly black, and the wall appears gradually, as the temperature approaches the limits.

A long-term project

JET’s first plasma with the ITER-Like Wall is expected in the spring of 2011. Not everything can and will be available on day-one. It would not even be useful! Essential for day-one are the new stop sequences required to protect the wall. The experimental programme will proceed cautiously, with wall temperatures not getting near the limits for several months. The new visible cameras will start without the IR filters for commissioning, and be converted later in the year. More sophisticated control sequences are planned taking into account the experience gained while working with the new wall. Some of these sequences will probably become available only after the next Shutdown.
Many other projects are coming to an end with the first plasma in 2011. This is not true for the project “Protection of the ITER-Like Wall”. It will continue for at least another two years to supply the control room team with all the necessary tools and to adjust the software in light of the experience gained operating with the new wall in its first year.

Klaus-Dieter Zastrow