Materials compatibility is a major issue for the liquid metal spoiler. The initial liquid metal options are described in materials.pdf.
ESH (Safety) issues associated with the liquid metal tests are described in:
The most successful technique was found to be to to "wet" the sample at high temperature (850C, 24 hours), and then operate at low (near melting point) temperatures. This appears to produced repeatable adherence without corrosion of the substrate. Experimental Results
Some photos of Liquid metal compatibility test are posted at: Liquid_test_photos
A prototype liquid metal spoiler has been constructed and was first tested in April 2001. Full Autocad drawings are available in DWG format (one huge drawing - a draft work in progress) as drawing.dwg. Note: this link will only work if you have access to Autocad, or a .DWG format reader. We are working on exporting the drawings to a more portable format.
Niobium (activated with Tin at 850C) was use for the main roller. Molybdenum was used for the smoothing roller. Heating was performed with 2 500W quartz halogen light bulbs. Cooling was performed with a low melting point eutectic (Ga 62%, In 22%, Sn 16%) with a 10.7C melting point. The roller OD was approximately 30mm, and was rotated at 6rpm, or about 1cm/second edge speed - similar to the value for the real spoiler system. A high torque (~100 in-lb) magnetic coupled drive, operated by a DC motor was used.
Roller Image: Molybdenum roller finished (foreground), Niobium roller ready for Tin coating before final machining.
Parts: Before Assembly, Assembly on table image 1, Assembly on table image 2,
Vacuum Chamber: System is all metal. Pump is a Varian V70LP macro-torr Turbopump, 70l/s nominal, backed with a Varian MDP12 diaphram pump. System pressure was ~3x10-8 after a 3 week pump down. Pressure measured with the heaters on, and the Tin at ~290C. Vacuum chamber interior photo (with chamber lid removed).
With the Tin bath fairly hot (290-300C temperature for the Molybdenum container), a thin (estimated by eye to be ~100 microns) coating of Tin was deposited on the Niobium roller. This coating was smoothed to a surface finish comparable to that of the finishing roller (near-mirror). Operation of the system was stable for several hours (maximum test duration). Image 1, Image 2.
With the Bath at lower temperature (280 - 290C), a much thicker (several mm) coating of Tin was deposited on the Niobium roller. This coating was rolled to a good surface finish by the Molybdenum roller. The coating thickness was unstable, and within a minute more Tin would freeze onto the roller, eventually binding the system. Thick_tin.
The system operated for approximately 2 days before one of the bearings on the Niobium roller failed. Steel needle bearings, coated with Tungsten Disulfide lubricant were used rolling directly on the Niobium. We believe that the Niobium was too soft for this application, and are planning to install pre-assembled (Tungsten disulphide coated) steel bearings.
Talk on liquid metal collimator work as of 1/11/2000. Talk.pdf
