AT-MAS-MA field quality crisis unit

Sections:

Divers:

The Field Quality Crisis Unit of the LHC main dipoles production

Systematic differences in main field between firms

Early January 2003, CERN. Warm magnetic measurements of latest collared coils from Firm 3 show a main field of 20 units higher in Firm 3 with respect to Firm 1 and 2 (see file). These data are confirmed by three cold mass measurements at 300 K and one at 1.9 K. This difference is giving a random integrated main field of around 8 units, i.e. the maximum specified value by beam dynamics (see LHC Project Report 501).

10th January 2003, CERN. Meeting of analysts and project engineers. Present: A. Devred, P. Fessia, C. Lopez, M. Modena, A. Musso, J. Miles, F. Savary, W. Scandale, E. Todesco, C. Vollinger, I. Vanenkov, E. Wildner. Scope of the meeting is to inform project engineers of the measurements and to start to prepare a list of solutions with cost estimate.

Brainstorming on proposals for curing the problem: 1) substituting 300 cm of iron laminations with nested laminations; this is the safer solution from a technical point of view, but has some impact on costs. 2) Decreasing the stacking factor of 1%; this would have no impact on cost, but could cause variation in the longitudinal mechanical behaviour of the cold mass and also influence the dipole geometry. 3) Decreasing the coil length of 30 mm; impact on the tooling has to be evaluated. 4) Increasing the coil radius of 50 microns, acting on winding mandrels or on ground insulation. This could have some influence on field harmonics.

Actions to be taken: Firm 3 management will be asked to build lamination packs with a stacking factor lower of 0.3 %, that is within what is foreseen in the specification. This could lead, in case of positive results, to the assembly of a cold mass with a lower stacking factor to partially compensate the effect. All the Firms will be informed of the problem and of the necessity of having flexibility in the assembly of laminations as foreseen in the specification.

Origins of the problem: this could be due to a coil radius 50 micron shorter than in Firm 1 and 2. Collar manufacturer is shown to be not responsible of this effect. One should investigate dimension of cables, coil protection sheet, interlayer. No effect of the change of cross-section is observed.

14th January 2003, CERN. Present: L. Bottura, S. Fartoukh, S. Sanfilippo, W. Scandale, E. Todesco. Meeting with groups in charge of measurements at 1.9 K and of beam dynamics specifications.

Data analysis: it is agreed to review all available data concerning magnetic length and main field to have a complete picture of the correlations between collared coil, cold mass and cryomagnets.

Schedule of cold tests: it is proposed to ask to give priority to Firm 3 dipoles that feature the high BdL in the collared coil.

Origins of the problem: other topics are proposed, i.e., the collaring procedure and the magnetic properties of collared coil components.

Cures of the problem: it is pointed out that the solution of acting on stacking factor of ferromagnetic laminations will work only at high energy, when the iron is saturated. A warning is given on the idea of correcting the low main field with a longer magnetic length, since it could give a non-negligible effect on the closed orbit; an estimate of this effect will be given.

18th May 2006, CERN.
Warm magnetic measurements of the integrated transfer function (TF) in Firm1 show a difference between the collared coil and cold mass data of about 15 units. Goal of this meeting is to discuss the possible source of this effect. Present: P.Galbraith, J.Garcia Perez, P.Hagen, J.P.Koutchouk, E. Todesco, C. Vollinger.

           The difference in the transfer function showed up after exchange of the moles (see plot), in addition, the CC data did not change. Consequently, the most
           probable source of the problem is the use of wrong calibration values of the moles in the CM system. The observed difference in TF between CC and CM
           data is about 15 units.
           Currently, moles #1, #7 are in use in Firm1. Peter confirmed that it has already been detected that these moles measure a field value which is systematically
           smaller by about 6-7 units, compared to other moles.

We agreed on the following measures to be taken:

One cold mass should be measured twice: first with moles #1, #7 (currently operational on the system), and afterwards with moles #9, #21 (spare moles, already available in the company). This should confirm Peter's observation and allow us to rescale the present values.
Remark: the spare moles #9, #21 were sent to Firm1 after re-calibration and should be OK. (Gijs should coordinate this with the company).

Since the main field data of the CMs is used by the MEB for the sorting of the magnets, a warning of possibly incorrect main field values of the CMs has to be given. The MEB should use the CC data instead (Ezio will do this).

In this framework, also the calibration values of moles #23, #24 that were used in Firm1 before, should be checked in order to explain the difference in main field which was observed between the apertures. Peter will use these moles on magnet 1041, re-calibrate them and measure magnet 1041 again for comparison.

(... to be continued ...)

Content- Ezio Todesco / C. Vollinger - October 2002