The number of particles in the beams used at CERN is measured by a family of devices called the Beam Current Transformers (BCTs). One of such devices is the DC Current Transformer (DCCT) measuring the total number of particles in an accelerator. The DCCT is built around a magnetic core made of a soft magnetic material. Currently, the DCCT cores are purchased from one of a few industrial partners. This situation might result in some signiï¬cant issues with development of the DCCTs. Firstly, the choice of the core can be made only among the already developed and manufactured cores. Secondly, the industrial partners do not normally guarantee long-term availability of
their products which could manifest itself as future maintenance problems. Thirdly, the cores available on the market have not been developed with CERNâs intended application in mind and, hence, might be non-optimal for the DCCTs. These three factors, among others, led to a decision to investigate a possibility of in-house core manufacturing. BCT cores are made up of a soft magnetic material in ribbon-form between 20 and 30 microns thick, wound into a toroidal shape. They are amorphous or nanocrystalline cobalt and iron-based alloys that show high permeabilities and low coercivities. Several commercial alloys in their amorphous as-cast state were purchased after identifying the key parameters of the materials for the application, contacting suppliers and studying the available materials to ï¬nd the most suitable ones. This choice was made by confronting the as-cast properties of the alloys with the needs of the instruments which would house the magnetic cores.
The materials were then characterised to determine how to thermally treat them in order to obtain a range of diï¬erent ï¬nal magnetic properties. This study also resulted in establishing the best candidates to fabricate the sample cores. After this, three alloys were selected (iron-based alloy Finemet FT-3 from Metglas and two-cobalt based materials, 6025 G40 from Vacuumschmelze and 2705 M from Metglas) based on their as-cast properties. The cores and ribbons of the selected materials were then annealed at diï¬erent temperatures and durations to study the eï¬ect of the treatment in both. Three treatments were done on the selected samples, below the Curie temperature, between the Curie and the crystallisation temperature and above the crystallisation temperature. Measurements of the samples were taken after the thermal treatment and compared to the results of the as-cast materials to see the eï¬ects of the annealing. This gave a range of diï¬erent ï¬nal materials with various permeability vs frequency behaviours, BH-curves and Barkhausen Noise. These new materials, in addition to the untreated as-cast ones, provide a good range of samples to choose from for applications with diï¬erent needs.
Based on the results presented in this thesis, one can suggest that the best material to build DCCT cores would be the iron-based nanocrystallised Metglas Finemet FT-3 material. It shows a high permeability up to 10 kHz, low coercivity and a rounded BH-curve. Regarding FBCTs, the cores need a very ï¬at BH-curve with high permeability and minimum coercivity. From all the materials presented, Vacuumschmelzeâs cobalt-based Vitrovac 6025 G40 is the one that would be the most suitable. It has the lowest coercivity. However, the perfect material would need magnetic thermal annealing to create a high magnetic material anisotropy for a flat curve.
EPFL2018
