Session 5: Components Properties and Construction Processes

Session Chair: Tom Taylor, Scientific Secretary: Christine Vollinger

Francesco Bertinelli: Status and Trends of the Various Components Affecting the Field Quality

Francesco Bertinelli gave a presentation about the current status and the observed trends of the various components and their effect on field quality. The status of the copper wedges and the stainless steel collars were presented in detail.

An increase of 0.02 mm in the spread of copper wedges dimensions was first observed in September 2002. This could be traced back to tooling wear. All extrusion and drawing dyes have since been replaced. Measurements of wedge geometry carried out by manufacturer and CERN are in very good agreement. The geometrical precision of the new wedges is significantly improved in both spread and average values. Concerning the stainless steel collars, CERN has recently started a campaign to crosscheck measurements between the two suppliers. Measurements of the same collar performed by the two suppliers show differences typically up to 0.04 mm. The reasons are being investigated, but probably originate from the different probe geometry used and different positions along the collar thickness where the measurement is taken. Emphasis is therefore on the quality of the cut (angle and torn area), which may vary within the same collar. A further effect under investigation is the relaxation of the collar after fine-blanking with a possible geometric effect typically up to 0.04 mm. On the basis of the existing data, both production processes appear under control, with the collar dimensions within tolerances: evidence from one of the two suppliers is however less solid. Following recent improvements and with additional work from CERN, the coming results should be more significant.

The question is asked whether further efforts are justified to improve the confidence level of the measurements from 0.04 mm down to 0.02 mm: the Component Centre is waiting for an answer. Additional measurements to define better the geometry could be taken on a few collars with the Smartscope machine, using the 3D method as now to ensure reproducibility.

Paolo Fessia: Into the Galaxy of Winding Curing and Collaring

Paolo Fessia presented the production process of cold masses with special emphasis on feedback time, which has to be considered after any corrective action is decided on winding, curing, or collaring process. Firstly, individual components, tooling, and assembly process in the three manufacturers were presented. Then, the production time needed by the manufacturers for the assembly of the collared coils was discussed. The necessary time for the coil winding and the collaring is decreasing with the total number of manufactured coils for all three firms. The estimate of the total time for a change of a single component was presented. Clearly, this depends on various factors, such as the production rate (one or three collared coils per week), and on the specific component to be modified. Different examples were given, namely

o       A change of polar shims requires a delay of about 1-4 weeks and an estimated cost of about 1500 CHF for each set of shims that cannot be used any more.

o       A change in the mid-plane insulation requires about 1.5-2 months delay. Taking into account the cost of the ground insulation (about 9000 CHF per collared coil), it is possible to estimate an extra cost for supply between 1 % and 20  % (from 100.000 CHF till 2.000.000 CHF) Furthermore, one has to consider that 15 min. of extra work for each coil corresponds to 300.000 CHF.

o       A change of copper wedges could be realized with a delay of about 14 weeks for the extrusion of the pieces. In this case, not only the copper wedges but also the end spacers and the end spacer chips and wedge tips have to be modified. A rough estimate of costs, based on former experience, gives about 7000 CHF/dye for the new copper wedges tooling and about 150.000 CHF for the end spacer production. No reusable material has to be taken into account corresponding, presently, to the largest part of costs.

o       The costs for a correction of the magnetic length by changing the number of end laminations are estimated to be about 200.000 CHF per manufacturer if the additional nested laminations have to be produced. In this case, a cheaper solution would be to transfer nested laminations between the different cold mass manufacturers.

Iouri Vanenkov: Analysis of the Coil Shape: Trends and Correlation with Magnetic Measurements

Iouri Vanenkov gave a presentation about the trends, which can be observed in the measurements of the coil shapes and showed the correlation between the overall coil shapes and the magnetic measurements. The general objectives of coil shape measurements are the validation of the tooling and the manufacturing method as well as the control of the geometrical tolerances of the coil size and the coil components. Furthermore, the measurements can be used to minimize the random part of the field errors by sorting the poles individually.

The measured parameters to assess the coil size were explained and examples of tooling imperfections and their influence on the longitudinal profile were given. Typically, both the individual layer size and the pole size are measured. The shape of the curing mold gives a typical pattern to the longitudinal coil profile clearly seen in the measurement data. This way, large tolerances on the curing mold result in an increase of the coil waviness. The trends in the size of the assembled and measured poles show an impact on the variation of the measured pre-stress in a magnet. A significant non-systematic variation of the coil size data could be observed for each company. A reduction of this variation was successfully tried in one company by a change of the curing shim. A sorting of the coils is possible but limited due to the cable mixing restrictions. By means of sorting, the random parts of field errors, especially on the a₂ multipole, could be reduced easily.

Finally, an analysis of the geometrical measurements on the outer shape of the collared coils has been carried out which showed a good correlation with the measured multipoles, especially b₂.

Minutes by Christine Vollinger

Last update on 07-04-2003 08:30:00
MG