- Email: [email protected]
Development of seven-gap resonators for the Heidelberg high current injector
Nuclear Instruments and Methods in Physics Research A328 (1993) 270-274 North-Holland
NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH S~(:hOI1A
Development of seven-gap resonators for the Heidelberg high current injector R. von Hahn, M. Grieser ~, D. Habs, E. Jaeschke 2, C.-M. Kleffner, J. Liebmann, S. Papureanu 3, R. Repnow, D. Schwalm ~ and M. Stampfer Max-Planck-lnstitut fiir Kernphysik, Heidelberg, Germany
A high current injector for the heavy ion storage ring TSR in Heidelberg is under construction. As a part of the injector eight seven-gap resonators with high shunt impedance are being developed. These resonators (f0 = 108.48 MHz) are designed for the synchronous velocities of/3~ = 3.7, 4.5, 5.1 and 5.7%. Low power models with scaling factors of 1:2.5 were built in order to study the characteristics of these new resonators. Following low level measurements to optimize the voltage distribution and eigenfrequency, a first power resonator was built and successfully tested at 80 kW (duty cycle of 25c/~). At this power, the resonator generated a maximum voltage (summation of all gap amplitude voltages) of 1.75 MV. This paper describes the design of the resonators and gives some details of the measurements.
1. Introduction Laser cooling e x p e r i m e n t s with ultra cold stored heavy ion b e a m s [4] of 'JBe + a n d 7Li+ are limited by the low c u r r e n t s delivered from the t a n d e m accelerator. T h e injection c u r r e n t s of 7Li+ a n d 9Be+ ions, usually less t h a n 1 ~ A , can only be a c c u m u l a t e d by m u l t i t u r n injection in the storage ring, resulting in an increase in intensity by at most a factor of 10. A new injector, consisting of a C H O R D I S ion source [5], two R F Q s [3] a n d eight seven-gap r e s o n a t o r s [1] presently u n d e r construction will result in an increase of the stored Li + a n d Be + c u r r e n t s by factors up to 1000. T h e layout of the injector is shown in fig. 1.
2. Layout of the seven-gap accelerator Based on the p r o v e n design of the n o r m a l conducting H e i d e l b e r g two-gap spiral r e s o n a t o r s [6] of the postaccelerator, o t h e r types of r e s o n a t o r s with two or six spirals, providing t h r e e or seven accelerating gaps, have b e e n developed [1,2]. T h e lack of flexibility due to the transit time factor curve is c o m p e n s a t e d by an increased s h u n t i m p e d a n c e resulting in a m u c h higher effective accelerating voltage p e r resonator. Based on m e a s u r e m e n t s for a high velocity prototype of a sevengap resonator, an effective accelerating voltage of 1.4 i Present address: University of Heidelberg, Heidelberg, Germany. 2 Present address: BESSY, Berlin, Germany. 3 Present address: IFIN, Bucharest, Romania.
M V for a low/3-resonator, o p e r a t i n g at 80 kW rf powcr with 25% duty cycle is expected. T h e design aim of the new high c u r r e n t injector is in a first p h a s e an o u t p u t energy of 13.7 M e V (/3 = 6.4%) for 7Li+ and 9.7 M e V (/3 = 4.8%) for ~Be +. A s s u m i n g an injection energy of 0.5 M e V / u from the R F Q these energies can be r e a c h e d with eight resonators for Li + or four resonators for Be + . To simplify construction, four pairs of identical resonators were designed following the velocity profile of the accelerated ions. B e a m dynamics calculations have b e e n done to m a t c h the injection of the b e a m from the R F Q and to optimize the acceptance of the linac. Based on these calculations, the resonators will be a r r a n g e d in four modules, each m o d u l e consisting of two resonators and one q u a d r u p o l e doublet (see fig. 1). The synchronous velocities are about 3.7, 4.5, 5.1 and 5.7%. A n acceptance of 40~r m m mrad was calculated with this arr a n g e m e n t . To minimize particle losses, the matching section must include a r e b u n c h c r to c o m p e n s a t e the time s p r e a d of the b e a m and two q u a d r u p o l e doublets to m a t c h the b e a m to the linac (see fig. 1). T h e calculated transmission is 98%. Adjusting thc scvcn-gap r e s o n a t o r linac to the output p a r a m e t e r s of the second R F Q , fig. 2 shows the longitudinal and transverse p h a s e space after the last seven-gap resonator.
3. Tuning of low power resonators C o n s t r u c t i o n of four low power resonators scaled 1 : 2.5 was r e q u i r e d for optimizing the field distribution
0168-9002/93/$06.00 © 1993 - Elsevier Science Publishers B.V. All rights reserved
271
R. yon Hahn et al. / Seuen-gap resonators ion-
source 0R0~s <_j
RFQaccelerator
T RFQ, I~TI[ ~ LLUl --ll
7-gap-resonator accelerator
RF[}2 DRD 7g D 7g O 7g O 7g D A rf~J~z~m~mMm ~m ]aranTtUwLuwWwl~ w I~
' A ~
Fig. 1. Scheme of the high current injector• T = quadrupole triplet, D = quadrupole doublet, 7g = seven-gap resonators, A = deflection magnet, R = rebuncher.
and shunt impedance as well as for the adjustment of the eigenfrequency to 108.48 MHz. One low power resonator (/3~ = 5.7%) is shown in fig. 3. In contrast to other multi-gap resonators, which basically combine a set of single spiral resonator arms, the seven-gap resonator has a single resonance structure. This resonance structure consists of a copper half shell with three arms attached to both sides. Drift tubes are added to the ends of the spiral arms. Driven in push pull mode, consecutive drift tubes have opposite potential. The field distribution in the gaps, the eigenfrequency and the shunt impedance have been optimized by rotating the circular part of the copper arms within the half shell. Due to this special design, tuning does not affect the alignment of the drift tubes along the beam axis. Fig. 4 shows a plot of the symmetrical field distribution at the required eigenfrequency taken by the bead perturbation method, where the phase shift A~ between resonator and generator signal is measured as a function of the location of the bead in the resonator. From this phase shift, the amplitude of the electric field and the shunt impedance of the full scale resonator can be calculated by low level measurements at the model resonator. A tuning plate (see fig. 3) opposite the arms was added to compensate for frequency shifts due to temperature changes. A frequency range of ___2.5 M H z can be achieved by moving this plate. For the full scale resonator, a range of _+ 1 M H z is expected. C o p p e r segments are fixed at both ends of the half shell to correct for frequency shifts due to small construction tolerances. These segments can compensate frequency deviations of up to _+5 MHz (_+ 2 M H z for the power resonator respectively) without changing the shunt impedance and field distribution. When tuning the resonator to the push pull mode at the required frequency, attention must be paid to other modes near the push pull mode because of mixing effects. The six arms with drift tubes correspond to six coupled oscillators and produce closely spaced modes. With the /3s = 5.1% and 5.7% model resonators, mixing effects between the required push pull mode and a push push m o d e were observed. Fig. 5A shows the
crossing of the two modes very close to the tuning frequency of 271.2 MHz. In other modes, very high currents can flow from the tuning plate to the tank and may destroy electrical contacts. To remove the unwanted mode from the operating frequency, so-called "ears" are attached at the tuning plate (see fig. 3), They change the eigenfrequency of the push push mode to lower frequencies without moving the push pull mode (see fig. 5B). Table 1 presents the measured shunt impedances of the low power model resonators and the calculated resonator voltages at a rf power of 80 kW.
4. The power resonator
A full scale power resonator (/3~ = 3.7%) was constructed based on the results of the bead perturbation measurement. Fig. 6 shows the open resonator. Segments on both sides of the half shell allow tuning of the resonator to the required eigenfrequency of 108.48 MHz. The tuning plate, extending into the resonator from below, is clearly visible.
10 .~ 05 LU 0 0 -0 5 -10
•
.
' . .
- . ,.:. • .:"~, .,~ . .i "~Alr.a .m • . , ~,]~!!~' .;.~ .~ . . . ' "':' "
. . . .
i
20
. . . .
~.0
i
. . . . ?
4_
N S
, •
. . . . .
r
.- ~ .
. . . .
10
~ 6
. . . .
0
t
20
[o~ ,
. . . .
,
.
2 o 4
s=~
_@
. . . .
-I0
6
O ~ m m m c a d i
. . . .
-5
.
,
r
~ . . . .
0 y [mrn] . . . .
,
. . . .
.... i , ' 2 " ,
5
i0
,
. . . .
4 E ×
0 -2
.
-4 -i0
..
"
£ =8 ~ - g m m m r a d 5
0 × [~m]
5
10
Fig. 2. Longitudinal and transverse phase space of the beam after the seven-gap resonators. Ill. BOOSTERS, RF CAVITIES
272
R. uon H a h n et al. / Screen-gap resonators
Fig. 3. One low power model (/3~ = 5.7%) with a scaling factor of 1 : 2.5.
R f power is coupled into the r e s o n a t o r n e a r one of the t h r e e legs which c o n n e c t the r e s o n a n c e structure to the tank, w h e r e the m a g n e t i c field has a maximum. F r o m b e a d p e r t u r b a t i o n m e a s u r e m e n t s at the power resonator, a m a x i m u m voltage of 1.75 M V at an input power of a b o u t 80 k W was calculated, which agrees very well with the expected value from the s h u n t i m p e d a n c e of the low power r e s o n a t o r (see table 1). Also a b e a m - t e s t was d o n e to m e a s u r e the accelerating voltage of the resonator. A 127j8+ dc b e a m with 81.9 M e V c o r r e s p o n d i n g to /35 = 3.7% was injected into the r e s o n a t o r a n d accelerated. W i t h a following 90 ° analyzer magnet, the m o m e n t u m distribution versus m a g n e t i c field was m e a s u r e d . Fig. 7 shows the m o m e n t u m distribution. F r o m the difference of the m a g n e t i c field b e t w e e n the d e c e l e r a t e d particles (A) a n d the a c c e l e r a t e d particles (B) the r e s o n a t o r voltage was calculated to be 1.74 MV, which c o r r e s p o n d s very
well with calculated values d e d u c e d from shunt impedance measurements. F u r t h e r power tests were carried out with up to 100 k W rf power at a duty factor of 25%. N e i t h e r p o n d e r o -
z LU.a
8 8-
d
~
s~
r~
~
1~
t~ 1¢,
d3
1,1
s [mm]
Fig. 4. Phase distribution of the push pull mode.
R. z'on Hahn et al. / Selden-gap resonators"
273
290
Table 1 The expected resonator voltages U0, calculated from the measured shunt impedances Z M of the lower power resonator models. Z L is the calculated shunt impedance of the power resonator
A
280 "~'~270
push pu[[ mo~~~
~260
/~push pushmode
~1.,-,250 240 230
J
220
15
19
23
27
31
35
39
280
,(3~
ZM [MI1/m]
ZL [M~/m]
U6 [MV] P = 80 kW
0.037 0.045 0.051 0.057
160 157 135 139
100 99 85 88
1.75 1.89 1.84 1.91
270 260
~-
"~-' 2SO
~
~
1.89kG
"1
240
~
230 220 210 200 190
10
15
'pushpushmode .~. 20
25
30
35
s [mm] Fig. 5. Mixing of the push pull and push push mode. (B) with and (A) without "ears" at the tuning plate.
I i4.029kG
I B
15.231kG
Fig. 7. Momentum distribution versus magnetic field of a t27j8+ beam after acceleration with the 3.7% power resona tor.
Fig. 6. The power resonator (/3 S = 3.7%) with segments at the half shell and the tuning plate. Ili. BOOSTERS, RF CAVITIES
R. yon H a h n et a L /
274
Table 2 Characteristic parameters of the 3.7% power resonator Frequency Quality factor Shunt impedance Maximum rf power Maximum resonator voltage Effective accelerator voltage Length Drift tube diameter
f Q Z P Uo Ueff
L r
108.48 MHz 5000 100 M I ) / m 80 kW (1:4) 1.75 MV 1.4 MV 0.4 m 20 mm
Sec, en-gap resonators
In the second phase, an E C R source can be added to the high current injector to accelerate highly charged heavy ions. The seven-gap resonators of the high current injector are designed in such a way that the output beam can be directly injected into the existing postaccelerator for further acceleration. This will allow high current acceleration to energies above the Coulomb barrier of the heaviest elements.
6. Acknowledgement motive forces nor multipactoring problems have been observed. The tuning plate was able to compensate for all eigenfrequency shifts due to variation of temperature. Table 2 summarizes the characteristic parameters of the first seven-gap resonator.
5. Outlook After the successful testing of the first full power resonator, tuning of the remaining three low power models is complete and construction of the power resonators has already begun. All resonator voltages will be greater than 1.75 MV, because with increasing resonator length the capacities between the arms decreases, which is associated with lower losses.
It is a pleasure for me to thank the technicians of the Max-Planck-Institute for their outstanding work.
References [1] M. Grieser, dissertation, Heidelberg, 1986. [2] M. Griescr, diploma thesis, Heidelberg, 1983. [3] C.-M. Kleffner, An RFQ-accelerator for the high current injector of the TSR, EPAC 92, Berlin, 1992. [4] R.W. Hasse, I. Hofmann and D. Liesen, Proc. Workshop on Crystalline Ion Beams, Wertheim, FRG, Oct 4-7, 1988, GSI-89-10, Darmstadt, 1989. [5] R. Keller, B.R. Nielsen and B. Torp, NUcl. Instr. and Meth. B37/38 (1989) 74. [6] E. Jaeschke, Rev. Phys. Appl. 12 (1977) 1605.
NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH S~(:hOI1A
Development of seven-gap resonators for the Heidelberg high current injector R. von Hahn, M. Grieser ~, D. Habs, E. Jaeschke 2, C.-M. Kleffner, J. Liebmann, S. Papureanu 3, R. Repnow, D. Schwalm ~ and M. Stampfer Max-Planck-lnstitut fiir Kernphysik, Heidelberg, Germany
A high current injector for the heavy ion storage ring TSR in Heidelberg is under construction. As a part of the injector eight seven-gap resonators with high shunt impedance are being developed. These resonators (f0 = 108.48 MHz) are designed for the synchronous velocities of/3~ = 3.7, 4.5, 5.1 and 5.7%. Low power models with scaling factors of 1:2.5 were built in order to study the characteristics of these new resonators. Following low level measurements to optimize the voltage distribution and eigenfrequency, a first power resonator was built and successfully tested at 80 kW (duty cycle of 25c/~). At this power, the resonator generated a maximum voltage (summation of all gap amplitude voltages) of 1.75 MV. This paper describes the design of the resonators and gives some details of the measurements.
1. Introduction Laser cooling e x p e r i m e n t s with ultra cold stored heavy ion b e a m s [4] of 'JBe + a n d 7Li+ are limited by the low c u r r e n t s delivered from the t a n d e m accelerator. T h e injection c u r r e n t s of 7Li+ a n d 9Be+ ions, usually less t h a n 1 ~ A , can only be a c c u m u l a t e d by m u l t i t u r n injection in the storage ring, resulting in an increase in intensity by at most a factor of 10. A new injector, consisting of a C H O R D I S ion source [5], two R F Q s [3] a n d eight seven-gap r e s o n a t o r s [1] presently u n d e r construction will result in an increase of the stored Li + a n d Be + c u r r e n t s by factors up to 1000. T h e layout of the injector is shown in fig. 1.
2. Layout of the seven-gap accelerator Based on the p r o v e n design of the n o r m a l conducting H e i d e l b e r g two-gap spiral r e s o n a t o r s [6] of the postaccelerator, o t h e r types of r e s o n a t o r s with two or six spirals, providing t h r e e or seven accelerating gaps, have b e e n developed [1,2]. T h e lack of flexibility due to the transit time factor curve is c o m p e n s a t e d by an increased s h u n t i m p e d a n c e resulting in a m u c h higher effective accelerating voltage p e r resonator. Based on m e a s u r e m e n t s for a high velocity prototype of a sevengap resonator, an effective accelerating voltage of 1.4 i Present address: University of Heidelberg, Heidelberg, Germany. 2 Present address: BESSY, Berlin, Germany. 3 Present address: IFIN, Bucharest, Romania.
M V for a low/3-resonator, o p e r a t i n g at 80 kW rf powcr with 25% duty cycle is expected. T h e design aim of the new high c u r r e n t injector is in a first p h a s e an o u t p u t energy of 13.7 M e V (/3 = 6.4%) for 7Li+ and 9.7 M e V (/3 = 4.8%) for ~Be +. A s s u m i n g an injection energy of 0.5 M e V / u from the R F Q these energies can be r e a c h e d with eight resonators for Li + or four resonators for Be + . To simplify construction, four pairs of identical resonators were designed following the velocity profile of the accelerated ions. B e a m dynamics calculations have b e e n done to m a t c h the injection of the b e a m from the R F Q and to optimize the acceptance of the linac. Based on these calculations, the resonators will be a r r a n g e d in four modules, each m o d u l e consisting of two resonators and one q u a d r u p o l e doublet (see fig. 1). The synchronous velocities are about 3.7, 4.5, 5.1 and 5.7%. A n acceptance of 40~r m m mrad was calculated with this arr a n g e m e n t . To minimize particle losses, the matching section must include a r e b u n c h c r to c o m p e n s a t e the time s p r e a d of the b e a m and two q u a d r u p o l e doublets to m a t c h the b e a m to the linac (see fig. 1). T h e calculated transmission is 98%. Adjusting thc scvcn-gap r e s o n a t o r linac to the output p a r a m e t e r s of the second R F Q , fig. 2 shows the longitudinal and transverse p h a s e space after the last seven-gap resonator.
3. Tuning of low power resonators C o n s t r u c t i o n of four low power resonators scaled 1 : 2.5 was r e q u i r e d for optimizing the field distribution
0168-9002/93/$06.00 © 1993 - Elsevier Science Publishers B.V. All rights reserved
271
R. yon Hahn et al. / Seuen-gap resonators ion-
source 0R0~s <_j
RFQaccelerator
T RFQ, I~TI[ ~ LLUl --ll
7-gap-resonator accelerator
RF[}2 DRD 7g D 7g O 7g O 7g D A rf~J~z~m~mMm ~m ]aranTtUwLuwWwl~ w I~
' A ~
Fig. 1. Scheme of the high current injector• T = quadrupole triplet, D = quadrupole doublet, 7g = seven-gap resonators, A = deflection magnet, R = rebuncher.
and shunt impedance as well as for the adjustment of the eigenfrequency to 108.48 MHz. One low power resonator (/3~ = 5.7%) is shown in fig. 3. In contrast to other multi-gap resonators, which basically combine a set of single spiral resonator arms, the seven-gap resonator has a single resonance structure. This resonance structure consists of a copper half shell with three arms attached to both sides. Drift tubes are added to the ends of the spiral arms. Driven in push pull mode, consecutive drift tubes have opposite potential. The field distribution in the gaps, the eigenfrequency and the shunt impedance have been optimized by rotating the circular part of the copper arms within the half shell. Due to this special design, tuning does not affect the alignment of the drift tubes along the beam axis. Fig. 4 shows a plot of the symmetrical field distribution at the required eigenfrequency taken by the bead perturbation method, where the phase shift A~ between resonator and generator signal is measured as a function of the location of the bead in the resonator. From this phase shift, the amplitude of the electric field and the shunt impedance of the full scale resonator can be calculated by low level measurements at the model resonator. A tuning plate (see fig. 3) opposite the arms was added to compensate for frequency shifts due to temperature changes. A frequency range of ___2.5 M H z can be achieved by moving this plate. For the full scale resonator, a range of _+ 1 M H z is expected. C o p p e r segments are fixed at both ends of the half shell to correct for frequency shifts due to small construction tolerances. These segments can compensate frequency deviations of up to _+5 MHz (_+ 2 M H z for the power resonator respectively) without changing the shunt impedance and field distribution. When tuning the resonator to the push pull mode at the required frequency, attention must be paid to other modes near the push pull mode because of mixing effects. The six arms with drift tubes correspond to six coupled oscillators and produce closely spaced modes. With the /3s = 5.1% and 5.7% model resonators, mixing effects between the required push pull mode and a push push m o d e were observed. Fig. 5A shows the
crossing of the two modes very close to the tuning frequency of 271.2 MHz. In other modes, very high currents can flow from the tuning plate to the tank and may destroy electrical contacts. To remove the unwanted mode from the operating frequency, so-called "ears" are attached at the tuning plate (see fig. 3), They change the eigenfrequency of the push push mode to lower frequencies without moving the push pull mode (see fig. 5B). Table 1 presents the measured shunt impedances of the low power model resonators and the calculated resonator voltages at a rf power of 80 kW.
4. The power resonator
A full scale power resonator (/3~ = 3.7%) was constructed based on the results of the bead perturbation measurement. Fig. 6 shows the open resonator. Segments on both sides of the half shell allow tuning of the resonator to the required eigenfrequency of 108.48 MHz. The tuning plate, extending into the resonator from below, is clearly visible.
10 .~ 05 LU 0 0 -0 5 -10
•
.
' . .
- . ,.:. • .:"~, .,~ . .i "~Alr.a .m • . , ~,]~!!~' .;.~ .~ . . . ' "':' "
. . . .
i
20
. . . .
~.0
i
. . . . ?
4_
N S
, •
. . . . .
r
.- ~ .
. . . .
10
~ 6
. . . .
0
t
20
[o~ ,
. . . .
,
.
2 o 4
s=~
_@
. . . .
-I0
6
O ~ m m m c a d i
. . . .
-5
.
,
r
~ . . . .
0 y [mrn] . . . .
,
. . . .
.... i , ' 2 " ,
5
i0
,
. . . .
4 E ×
0 -2
.
-4 -i0
..
"
£ =8 ~ - g m m m r a d 5
0 × [~m]
5
10
Fig. 2. Longitudinal and transverse phase space of the beam after the seven-gap resonators. Ill. BOOSTERS, RF CAVITIES
272
R. uon H a h n et al. / Screen-gap resonators
Fig. 3. One low power model (/3~ = 5.7%) with a scaling factor of 1 : 2.5.
R f power is coupled into the r e s o n a t o r n e a r one of the t h r e e legs which c o n n e c t the r e s o n a n c e structure to the tank, w h e r e the m a g n e t i c field has a maximum. F r o m b e a d p e r t u r b a t i o n m e a s u r e m e n t s at the power resonator, a m a x i m u m voltage of 1.75 M V at an input power of a b o u t 80 k W was calculated, which agrees very well with the expected value from the s h u n t i m p e d a n c e of the low power r e s o n a t o r (see table 1). Also a b e a m - t e s t was d o n e to m e a s u r e the accelerating voltage of the resonator. A 127j8+ dc b e a m with 81.9 M e V c o r r e s p o n d i n g to /35 = 3.7% was injected into the r e s o n a t o r a n d accelerated. W i t h a following 90 ° analyzer magnet, the m o m e n t u m distribution versus m a g n e t i c field was m e a s u r e d . Fig. 7 shows the m o m e n t u m distribution. F r o m the difference of the m a g n e t i c field b e t w e e n the d e c e l e r a t e d particles (A) a n d the a c c e l e r a t e d particles (B) the r e s o n a t o r voltage was calculated to be 1.74 MV, which c o r r e s p o n d s very
well with calculated values d e d u c e d from shunt impedance measurements. F u r t h e r power tests were carried out with up to 100 k W rf power at a duty factor of 25%. N e i t h e r p o n d e r o -
z LU.a
8 8-
d
~
s~
r~
~
1~
t~ 1¢,
d3
1,1
s [mm]
Fig. 4. Phase distribution of the push pull mode.
R. z'on Hahn et al. / Selden-gap resonators"
273
290
Table 1 The expected resonator voltages U0, calculated from the measured shunt impedances Z M of the lower power resonator models. Z L is the calculated shunt impedance of the power resonator
A
280 "~'~270
push pu[[ mo~~~
~260
/~push pushmode
~1.,-,250 240 230
J
220
15
19
23
27
31
35
39
280
,(3~
ZM [MI1/m]
ZL [M~/m]
U6 [MV] P = 80 kW
0.037 0.045 0.051 0.057
160 157 135 139
100 99 85 88
1.75 1.89 1.84 1.91
270 260
~-
"~-' 2SO
~
~
1.89kG
"1
240
~
230 220 210 200 190
10
15
'pushpushmode .~. 20
25
30
35
s [mm] Fig. 5. Mixing of the push pull and push push mode. (B) with and (A) without "ears" at the tuning plate.
I i4.029kG
I B
15.231kG
Fig. 7. Momentum distribution versus magnetic field of a t27j8+ beam after acceleration with the 3.7% power resona tor.
Fig. 6. The power resonator (/3 S = 3.7%) with segments at the half shell and the tuning plate. Ili. BOOSTERS, RF CAVITIES
R. yon H a h n et a L /
274
Table 2 Characteristic parameters of the 3.7% power resonator Frequency Quality factor Shunt impedance Maximum rf power Maximum resonator voltage Effective accelerator voltage Length Drift tube diameter
f Q Z P Uo Ueff
L r
108.48 MHz 5000 100 M I ) / m 80 kW (1:4) 1.75 MV 1.4 MV 0.4 m 20 mm
Sec, en-gap resonators
In the second phase, an E C R source can be added to the high current injector to accelerate highly charged heavy ions. The seven-gap resonators of the high current injector are designed in such a way that the output beam can be directly injected into the existing postaccelerator for further acceleration. This will allow high current acceleration to energies above the Coulomb barrier of the heaviest elements.
6. Acknowledgement motive forces nor multipactoring problems have been observed. The tuning plate was able to compensate for all eigenfrequency shifts due to variation of temperature. Table 2 summarizes the characteristic parameters of the first seven-gap resonator.
5. Outlook After the successful testing of the first full power resonator, tuning of the remaining three low power models is complete and construction of the power resonators has already begun. All resonator voltages will be greater than 1.75 MV, because with increasing resonator length the capacities between the arms decreases, which is associated with lower losses.
It is a pleasure for me to thank the technicians of the Max-Planck-Institute for their outstanding work.
References [1] M. Grieser, dissertation, Heidelberg, 1986. [2] M. Griescr, diploma thesis, Heidelberg, 1983. [3] C.-M. Kleffner, An RFQ-accelerator for the high current injector of the TSR, EPAC 92, Berlin, 1992. [4] R.W. Hasse, I. Hofmann and D. Liesen, Proc. Workshop on Crystalline Ion Beams, Wertheim, FRG, Oct 4-7, 1988, GSI-89-10, Darmstadt, 1989. [5] R. Keller, B.R. Nielsen and B. Torp, NUcl. Instr. and Meth. B37/38 (1989) 74. [6] E. Jaeschke, Rev. Phys. Appl. 12 (1977) 1605.