August 2017

A vacuum leak in the north arm

On August 1, 2017, the GEO 600 team noticed a pressure increase in the vacuum system. The first suspicions were a pump failure or a defective valve. It soon became clear that the vacuum system itself must have developed an internal or external leak!

By using data from the pressure gauges the location of the suspected leak was narrowed down to somewhere in the vicinity of the central vaccum tank cluster and the north beam tube. Spraying helium and observing the composition of the gas pumped out by the vacuum pumps, all signs pointed to the section of the north beam tube directly next to the central building.

On the vacuum tube outside the central building a pile of animal droppings (probably from a marten) was found on top of the tube. When carefully removing the droppings and cleaning the tube, a sudden, dramatic pressure increase was proof of having found the leak. Only by closing the leak with his finger, GEO600 operator Marc Brinkmann, prevented an even further pressure increase, the failure of the turbo pumps, and an uncontrolled flooding of the vacuum system with humid air. This would have meant getting back to the high vacuum level of 10^-9 mbar would have taken years.

Luckily, the leak – estimated to be about 0.1 mm in size – could be (temporarily) sealed with Tra-Bond vacuum glue. The glue is the blue substance in the image above. It took about a day of running the vaccum pumps for the detector to return to observation mode. A permament solution (vacuum welding the leak) must be found at some point in the future.

As one member of the GEO600 team said: “It was a great group experience and showed how well the team works as one, but it also in an experience I'd rather not repeat.”