Choose timezone
Your profile timezone:
Review and analysis of the aborted collaring of CC11 - Part 2
→
Europe/Zurich
Minutes of Meeting
Participants:
A. Devred, A. Milanese, D. Pulikowski, J.L. Rudeiros Fernandez, F. Savary, G. Spigo
Excused:
L. Bottura, D. Tommasini
Foreword:
The second attempt of collaring of CC11, which took place on 29 June, was aborted like for the first attempt during the second alignment step, at a pressure of 45 bar. Even though corrective actions were put in place in order to fix (1) the damaged insulation, (2) the local deformation of the cold bore tube, and (3) the presence of burrs at the bottom edge of the loading plates, these did not cure the issue that is a larger residual gap between the top and bottom parts of the collaring tooling at 45 bar.
Arnaud underlines the importance of this second meeting, which is meant to agree on and follow a collegial approach to sort out the issue, and to keep traceability of the decision making. He informs us that Miguel is aware of the problem, and that he understands the difficulties and the impact on the schedule.
Contributions
Update on NCR and preparation of the 2nd collaring attempt [F. Savary]
Frédéric describes the corrections put in place prior to the 2nd collaring attempt, and the findings during this 2nd attempt. In short, the residual gap measured at 45 bar between the top and bottom parts of the collaring tooling is unchanged, same as during the 1st collaring attempt.
Following questions posed by Attilio, Frédéric confirms that the teams of GE and CERN were present with all the needed personnel, and the collaring press worked well (it had been checked by Sebastien the week preceding the 2nd collaring attempt).
The NCR level 4, EDMS document Nr 2379821, has been updated (now v0.3) with clarifications as to (1) the initial action plan ("premières dispositions") following the abort of the 1st collaring attempt, (2) the outcome of the 2nd collaring attempt, and (3) the 2nd action plan following the abort of the 2nd collaring attempt. A Root Cause Analysis and a comprehensive technical report were produced by the contractor GE following the abort of the 1st collaring attempt. These documents are attached to the NCR in EDMS.
Coil metrology - Accuracy of the measurements [D. Pulikowski]
Dariusz recalls the metrology process, the statistics available since the beginning of the production, and more specifically the metrology results for the coils GE24 and GE25.
It is noted that coil GE25 features a large asymmetry of the mid-plane between the left and right branches, comprised between 0.3 and 0.4 mm (Slide 9). There is a general consensus on the fact that this does not correlate with the inside diameter of the coil, and therefore with the observed local deformation of the cold bore tube, which is clearly due to the presence and misplacement of the centring supports.
It is recalled that the Go-NoGo check of the inside geometry of the cold bore tube is not part of the standard QC plan. It is done as a systematic check after collaring, and it has been done at an earlier stage of the fabrication in the present case of CC11 because of the anomalies observed during the alignment steps.
As to the deformation of the cold bore tube after collaring, it is noted that this happened on CC09, as described in the NCR EDMS Nr 2263672. The gauge of 49.70 mm diameter passed through the tube, however with resistance at 800 mm from the connection side. That location corresponds also to the position of a cold bore tube centring support.
The accuracy of the measurements carried out with the FARO Arm is of the order of 29 µm. This is the repeatability guaranteed be the manufacturer of the equipment. It is consistent with our experience accumulated to date as to the execution of the measurements and data analysis. We observe an average spread on the mid-plane excess of the order of 20-30 µm (that is the error bar) over production.
CC11 - Analysis of the collaring from a geometrical perspective [J.L. Rudeiros Fernandez]
José Luis provides an analysis of the collaring press closure process based on the evolution of the gap as a function of the hydraulic pressure developed by the press. It appears clearly on slide 4 that the residual gap measured at 45 bar for the two collaring attempts of CC11 is almost the same. These two cases are also standing out from the other ones (CC01 to CC10), even if by only 72 µm.
The variation of the initial gap (measured at 25 bar) between the first and the tenth collared coils is only 62 µm (slide 6). At this stage, we are already in the linear elastic part of the modulus as the assembly is already loaded at 25 bar.
The equivalent stiffness of the assembly in the collaring press can be determined from the gap measurements plotted as a function of the force applied by the collaring press during the collaring stage. It varies only a little, between 7.3 and 8.4 GPa.
As to the stresses, a reduction of the collaring shims by 50 µm would imply a reduction of the peak azimuthal stress in the mid-plane from 146 to 141 MPa (slide 8), and a reduction of the collar nose stress from 125 to 101 MPa (slide 8), all this at the collaring force.
Attilio is questioning the necessity of adjusting the collaring shims down to the level of 25 µm thickness.
Giancarlo reports that the parts look very good, and the assembly work very well done (he visited the winding house on 17 June). He suggests that the only elements that could possibly affect the behavior of the assembly in the press are the coils and their equivalent stiffness.
Proposed plan for next collaring attempt [F. Savary]
Frédéric presents the new action plan after the abort of the 2nd collaring attempt (slide 10). It is proposed:
- To reduce the collaring shims by 50 µm;
- To secure the collaring shims by means of Mylar adhesive tapes;
- To verify, and correct if needed, the fit between the cold bore tube centring supports and the pole pieces.
As to the suggestion to do a pre-assembly and test with Fuji paper, the opinion of the experts is as follows:
- Davide (not present during the meeting but he had given his opinion to Frédéric): he would do it in order to make sure we know the level of stress seen by the coils in these conditions (tooling with 0.2 mm thick shims, and pressure of 45 bar);
- Attilio: he underlines the risk associated to the additional work and manipulation of the coils. He finds the test interesting but rather academic. Also we did not see signs of local issues on the coils, rather a global effect in the collaring press. Therefore, he would refrain to do it;
- Giancarlo: although he finds the test interesting, he also said it is not useful. In addition, it would imply extra delay. Therefore, he would refrain to do it.
It is agreed that we do not add a pre-assembly with Fuji paper. We could possibly do it with another assembly and other coils, if really needed and for additional purposes.
The cold bore tube will be reused. Attilio told that it could be turned between the NCS and the CS in order to compensate, if need be, the local deformation observed at one end. As this deformation was purely elastic, it is not necessary.
Giancarlo asks to clarify the possible correlation between the coil azimuthal excess and the collaring force. In response to this, José Luis has produced a plot, which shows that there is no correlation between these two parameters. This plot has been added to his contribution (this Indico page).
Conclusion
The experts agree on the proposed plan for the reconstruction and collaring of CC11 that consists in reducing the collaring shims by 50 µm, securing the collaring shims by means of Mylar adhesive tapes, and correcting if needed the fit between the cold bore tube centring supports and the pole pieces.
There are minutes attached to this event.
Show them.