1: \subsection{Jolly Green Giant repair}
2:
3: The Jolly Green Giant magnet (JGG) provides the magnetic field for charge
4: and momentum measurement of particle tracks in the TPC. The aperture of
5: the JGG magnet is large enough to fit the TPC. The magnetic field of 0.7~T
6: is (except for distortions) vertical and parallel to the electric drift
7: field inside the TPC.
8:
9: \subsubsection{Present state of the JGG}
10:
11: The JGG magnet has two pairs of water cooled copper coils with a total of
12: 1024 coil turns. Turns are insulated from each other with sheets of G10 and
13: epoxy. The coils have been power cycled many times and have been operated
14: for a long time over the past four decades. A failure in one of the coils was
15: repaired before the first MIPP run. The magnet was then used in MIPP
16: for three years. During this time we have had electrical turn-to-turn coil
17: shorts and water leaks in the coils four times. Three times the magnet was
18: restored to an operational state. The failing coils were bypassed with external
19: jumpers. The operating current was then increased to obtain the same
20: magnetic field as before each failure. The last failure close to the end of
21: the first run completely destroyed the bottom two coils.
22:
23:
24: \subsubsection{New coil design}
25:
26: The need to replace the JGG coils opened up the possibility for
27: improvements. The magnetic field of the JGG is not very uniform. The
28: region of interest is the rectangular active drift volume of the TPC
29: centered in the magnet aperture. It extends 164~cm along the beam direction,
30: 104~cm horizontally perpendicular to the beam, and 90~cm vertically. Within
31: this volume the magnetic field components perpendicular to the electric
32: field of the TPC reach up to 20\% of the magnetic field component parallel
33: to the electric field (see figure \ref{fig-JGG-field}).
34: The perpendicular components introduce distortions
35: in the TPC track data of up to 5~cm\cite{MIPPNote134}. These distortions can
36: be corrected but residuals of up to 1.4~mm remain. This impacts vertex
37: reconstruction and momentum determination. A more homogeneous magnetic field
38: results in better data.
39:
40: %
41: \begin{figure}[htb!]
42: \includegraphics[width=5in]{jgg-field-x5y5.eps}
43: \caption{\label{fig-JGG-field} Components of magnetic field in the JGG
44: in tesla plotted against grid positions (two inch per grid
45: space). Black triangles show the current coils. Blue inverted
46: triangles show the model for nine inch extended coils. Squares and
47: circles show extensions of 12 inches and 18 inches, respectively. The
48: field shapes due to the 9 inch version are considerably better than
49: the original field in that the extent of the bending field $B_y$ is
50: greater and the the other components $B_z$ and $B_x$ are better
51: behaved.}
52: \end{figure}
53: %
54:
55: The width of the coils is given by the shape of the magnet yoke. It is hard
56: to change. The pole pieces are 60 inches wide. This 1.46 times the width of
57: the TPC drift volume. The length of the coils can be expanded more
58: easily with new coils expanding symmetrically upstream and downstream
59: of the yoke. Currently the pole pieces extend only 48 inches along the
60: beam. This is only 74\% of the length of the TPC drift volume. The
61: effect of larger coils on field uniformity was modelled for several sizes
62: of extensions in spring and summer of 2006\cite{MIPPNote134}. An extension
63: of the coils by 9 inches on each end gains a significantly more uniform
64: magnetic field and does not interfere with the placement of the detectors
65: downstream of the JGG. The further gains for extensions larger than 9 inches
66: are smaller.
67: With the new coils the pole pieces will have the same size along the beam
68: direction as the TPC drift volume. The resulting $\mbox{E}\times\mbox{B}$
69: distortions will be less than 3~cm throughout the drift volume of interest,
70: half of the distortion with the old coils. After correcting for the
71: new distortions remaining residuals will be at most 0.5~mm. This is
72: more than twice as good as the results with the old coil.
73:
74: Besides this work on the coil geometry a lot of work was done on the
75: detailed design of new coils. For cost reduction the new coils will be
76: made from aluminum rather than copper. The coil conductor will have a larger
77: cross section. Two new coils will have 360 turns. The heating
78: calculation for the final design has been performed and found to be
79: satisfactory.
80: The impact on the magnet power supply
81: and power bus was evaluated. The new coil specifications are listed in
82: MIPP Note 137\cite{MIPPNote137}.
83:
84: \subsubsection{Coil replacement}
85:
86: The orders for aluminum for new coils and for the coil fabrication have
87: been placed in summer 2006.
88: The old coils have to be removed. Detectors in and upstream of the
89: JGG will need to be moved out of the way. The new coils will be larger
90: than the old coils. The pole pieces have to be lengthened in the beam
91: direction to fill the resulting gap.
92:
93:
94: \subsubsection{Ziptracking the new magnetic field}
95:
96: The magnetic field of the JGG with new coils has to be mapped. The Ziptrack
97: system was used to map the magnetic fields of both MIPP analysis
98: magnets before our commissioning run.
99: We propose to upgrade the Ziptrack system with new Hall
100: probes because the cables and connectors on the old Ziptrack Hall probes
101: have become unreliable over time. Replacing cables and connectors on
102: the existing Hall probes is not cost effective because the Hall probes
103: would need to be recalibrated.
104:
105: The PC currently used to control the Ziptrack and collect the data needs
106: to be upgraded and the Ziptrack software needs to be adapted to the new
107: hardware.
108: