Residual Magnetic Field in VTA Dewar 7

JLAB-TN-08-059

September 26, 2008

Residual Magnetic Field in VTA Dewar 7

Improvement by Adjusting Compensation Coil Excitation Current andEffect on Cavity Quality Factor

R.L. Geng, P. Kushnick

An observation of lower values of the unloaded quality factor (Q0) of 1.3 GHz 9-cell cavities tested in VTA Dewar 7 was made recently [1]. Similar observation was made during a recent test of 1.5 GHz 7-cell CEBAF upgrade cavity (HG006) tested in the same Dewar. These triggered an effort for measuring the residual magnetic field because the frozen flux effect on Q0 is a well known phenomenon [2].

The measurements were done using fluxgate magnetometer (MEDA µMAG). Two sensors were used, each for the vertical and transverse component of the residual magnetic field. Sensors, attached to a long wooden marked measure stick, were lowered down to the bottom of Dewar 7 and then data was taken as sensors were raised upward stepwise (0.5 ft step size). The arrow mark on the vertical sensor points toward the bottom of Dewar. It turns out (as expected) the vertical component dominates.Therefore we ignore the transverse components and focus on the changes in the vertical component as the measurement conditions are varied. The values along the Dewar central axis are identical within 4 mGas compared to that along a line 5 inch off center.The position of the movable radiation shielding block was found to have some measurable but un-important effect.

Measurement results are shown in Fig. 1. There is an obvious bump with a peak amplitude value of about 30 mG in the “as found” condition. This bump is below the top flange level for both 9-cell ILC cavity (170 cm above Dewar bottom) and 7-cell CEBAF upgrade prototype cavity HG006 (150 cm above). Therefore, it is possible to cause lower Q0 values in these cavities due to frozen flux effect.

After a few experiments with the excitation current of compensation coils controlled by power supply 7A, it was realized an improvement is possible. An adjustment of 60 mA results in such a satisfactory improvement that the axial residual magnetic field stays within ± 5 mG up to 170 cm distance from the Dewar bottom. The excitation current of the other power supply 7B remained un-changed over the course of measurements. Finally, we decided to set the excitation current of power supply 7A & 7B at 1.60 A & 3.13 A, respectively.

Subsequent re-test of HG006 in Dewar 7 saw a 10% increase in low field Q0 values, confirming the benefit of the improvement. During further testing of HG006 following 120C bake, it was possible to discern clearly the BCS resistance difference between 1.8 & 2.0 K, further corroborating the benefit [3]. Finally, the first test of a 1.3 GHz 9-cell cavity sincesaw a low field Q0 value of 2.3×1010, which again favorably supports the improvement.These positive results triggered a decision to re-test a 9-cell cavity which exceeded ILC gradient spec (35 MV/m) but with a Q0 value lower than the spec of 8×109.

Incidentally, the residual magnetic field in some other VTA Dewars was also found excessively high and improvements were warranted[4]. It is possible that an optimal residual field achieved at one time becomes invalid due to, for example, changes in the ambient field. (The power supply output current stability, as argued by some, is another possible source.)It is advisable to repeat the described measurements routinely in all Dewars used for SRF cavity testing. Improvement must be sought when a residual magnetic field value of more than 15 mG is measured in the space occupied by SRF cavities.

References

[1] A.C. Crawford, private communication. Besides the effect discussed in this note, magnetized hardware was found in some 9-cell testing, also contributing factor to the observed low Q0 values.

[2]H. Padamsee, J. Knobloch, T. Hays, RF Superconductivity for Accelerators, P. 173, Ch. 9.3, John Wiley & Sons, Inc, (1998), ISBN 0-471-15432-6.

[3] R.L. Geng et al., to be published.

[4] P. Kneisel, private communication. A value of 300 mG was even measured in interested space.

Figure 1: Vertical component ofaxial DC magnetic field in VTA Dewar 7. A 30 mG bump is apparent in the “as found” curve. This bump is below the top flange level for both 9-cell ILC cavity and 7-cell CEBAF upgrade prototype cavity. Hence it is possible to cause lower values of unloaded quality factor of these cavities due to frozen flux effect. After a few experiments with the excitation current ofcompensation coils controlled by power supply 7A, it was found that an improvement is possible. The vertical component of the residual magnetic field along the Dewar axis was satisfactorily improved to within ± 5 mG up to 170 cm distance from the bottom of Dewar 7.