This meeting was a preparatory meeting for an extraordinary restricted/extended MEB.

Measurement of the geometry axis of the short straight sections are made using two measurement devices, the AC-mole and the GEO-mole. The AC mole uses an optical system: reflected light from four leds in the CBT is used to calculate the CBT centre by a best fit to an ellipse. The GEO mole is a mechanical system using 3 supports with similar mechanical properties (spring constant) to hold the mole in the centre of the CBT.

Originally the AC mole was chosen to measure the SSS due to its possibility to measure, at the same time, also the magnetic axis of the magnets in the assembly. Initial difficulties with calibrations of this device made it necessary to simultaneously use the GEO mole for cross checks. This habit has been maintained due to some lack of confidence in the AC mole measurements. The AC mole was qualified in the LHC note 373 where results from measurements were described (In this report AC-mole 6 is used, in our statistics a difference between the two moles of ~0.3 have been detected). From this qualification, the AC mole has been used as the base line measurement for GA-calculation and for qualifying the  mechanical axis measurements of all quadrupole assemblies. Only when AC mole data is not good, the GEO-mole data are used.

We have seen from all measurements made so far that there is a statistical difference of the two moles of 0.1 mm in the vertical plane. Larger differences have been detected for several magnets.  For the last MEB batch (SSS 49, 152, 177, 249, 349) no AC mole measurements were available for 152 and 177. Seen that the GEO-axis for the other magnets in the batch was ~0.3 mm lower than the AC mole measurements, it was asked that the magnets in the batch could be investigated in more detail before shifting.

A presentation given in the extraordinary MEB meeting 1/11 2005 (see document: presentation) was prepared based on the discussions in this meeting.

The GEO mole 13 (used for the magnets in the batch) was found showing 0.12 mm too low values in the vertical plane. The mole has been repaired. The AC mole 4 (used for the batch) was found showing 0.89 mm too high. The mole has been re-calibrated.

A cross calibration bench was set up, with a calibrated tube to be used for comparative measurement. The center of the tube was determined by best fits to a circle of measurement points taken on the outer side of the tube. The two moles were used to measure  the center. The results of the measurements were not conclusive. A better bench will be set up, using a longer tube for which the centre will be determined from measurements in the metrology lab (3D measurements with 20 micrometers precision). The measurement-conditions will be also improved.

From the statistics:

In the following tables are shown the difference between measurements in our data base, many different mole pairs are used. There are 37 horizontal and 39 vertical points outside 3 sigma in the point by point comparison. Histograms are in the slides.

Difference AC-GEO, point by point, linear interpolation [mm]

Difference AC-GEO, integrated over the CBT [mm]

Most often the GEO-mole shows lower values than the AC-mole. This is not always true.

Other magnets than SSS also have the problem, for example 607, passed the MEB.

Improvements in the future:

The calibrations of the moles will be followed up more closely. AT/MAS will trace calibrations in the database and may help with a tool  to fill information after calibrations.

Cross calibrations in the new bench will be done regularly. MTM and SU will set up a scheme for this.

The operator software will be enhanced with a check if the integrated difference is larger than 0.2 mm. In that case a warning will be given and the measurement will be repeated with two other moles. The old moles will be calibrated/repaired. This may be iterative.

Jerome was asked to give the chronologically the difference between the two moles, with information about the mole pairs. The result is in the slides. page 8 and 9.

Also dipoles may be affected by the problems with calibration. Studies will be made.

Magnetic axis measurements also need to be carefully checked and calibrated.

Saw tooth effects on the GA calculation will be checked on a few magnets by Patrick.

Saw tooth effects of SSS measurements are not critical (shorter magnets). However the long SSSs may be influenced.

Saw tooth may come from calibration problems or from station relocation. To be investigated case by case by the measurement specialists.

Magnetic axis measurement instruments should also be calibrated.

Conclusion for the batch:

The measurements are within 0.2 mm , taking into account the calibration offsets and 0.1 mm (still not explained).

Magnets 152 and 177 will be re-measured since no AC-mole data available. Magnet 249 will be re-measured (the only magnet not to be immediately installed). For this magnet, a scan will be made of the ends of the magnet to have a better idea of which measurement is closest to the measurement of the flange (the flange is not measured by the moles).

Magnets 49 and 349 will be evaluated using existing data and information.

Elena