1. Main Coupler R&D at DESY
TESLA Technology Collaboration (TTC) Meeting, Milano, 6-9 February 2018.
Main Coupler R&D at DESY
Denis Kostin, DESY

2. Outline
E-XFEL Main Coupler R&D
Main Coupler for CW Operation

3. E-XFEL Main Coupler
E-XFEL Input RF power coupler consists of warm, cold and waveguide main parts. Coaxial coupler is made of copper and copper plated (10/30µm) stainless steel with alumina TiN coated ceramic windows. Motorized antenna tuning (10mm) allows for Qext adjustment ( ).

4. E-XFEL Main Coupler Design
Geometry
Copper Plating
part
inner [μm]
inner RRR
outer [μm]
outer RRR
cold
solid copper
10±20%
bellow: ±30%
30..80
warm
30±20%
RF Power
peak power
rep.rate
equivalent CW power
150 kW
10 Hz
1.9 kW

5. E-XFEL Main Coupler Push-Rod Bellow Problem [1]
All capacitors are replaced by coax gaskets: no push-rod leaks (RF is ON) since then
problem solved
FPC push-rod leak:
leak spot discovered
E-XFEL FPC
FPC coax-gasket:
capacitors are replaced
FPC simulation:
leak spot may have a contact-discharge
CST MWS simulation

6. E-XFEL Main Coupler Push-Rod Bellow Problem [2]
possible solution with capacitor (to use HV bias):
push-rod isolation inside the bellow to prevent the electrical contact.

7. E-XFEL Main Coupler Actuator Optimization
possible actuator optimization:
increase the moving range over 10 mm;
improve the motor axle connection;
shift the peek axle connection to the push-rod outwards for easy access and removal;
in case of not implementing the HV bias consider the design without the peek axle.

8. E-XFEL Main Coupler Parts Problems [1]
FPC test/operation problems examples #
Problem
Number
Repair Action 1
T70K over temperature caused by a loose contact between warm and cold parts inner conductors (inner screw problem) 18
13 warm parts replaced,
5 warm parts cleaned,
cold parts cleaned 2
T70K over temperature caused by warm part problem, RF discharge in some cases 10
warm parts replaced 3
not conditionable RF discharge in a warm part 11 4
T70K overheating in the test
warm parts disassembled and checked: no problem found,
re-assembled 5
warm part damaged by the discharge with not tightened bias capacitor screws
1 warm part replaced,
1 warm part cleaned 6
waveguide part damaged by a warm part RF discharge during the test (see p.3)
warm part replaced 7
actuator axle broken 8
axles replaced
push-rod leak 26
push-rods replaced 9
push-rod cold end not connected
push-rods re-assembled

9. E-XFEL Main Coupler Parts Problems [2] (XTL)
position
module
coupler part
status
comments 1
A4.M4.C4
XM72
CPI-CP-024
shorted
T70K → 200K, high e-1 signal (discharge), cold part problem. AMTF: push-rod replaced, no other problems. 2
A12.M4.C1
XM53
THRI-WP-706
T70K → 150K, discharge, warm part problem. No problem / observations in AMTF. 3
A13.M2.C8
XM51
THRI-WP-683 OK
high e-2 signal – conditioned. 4
A16.M1.C1
XM69
THRI-CP-880
high e-1 signal – conditioned. 5
A16.M2.C1
XM60
CPI-CP-028
T70K → 150K high e-1/2 signals, cold part problem. Problem seen in AMTF, T70K → 115K (some overheating). 6
A16.M2.C2
CPI-WP-030
high e-2/3 signals – conditioned. 7
A16.M4.C7
XM66
THRI-WP-323 8
A19.M3.C4
XM-2
AC3H26
testing
T70K overheating – no discharge. Observed in AMTF as well. 9
A19.M3.C6
AC3H38
conditioned (5hr) up to 150kW FT-pulse (25MV/m) 10
A20.M4.C1
XM65
THRI-WP-470
high T70K, cpl.vac. increased, high e-2 signal bursts, warm part problem. 11
A21.M2.C5
XM92
THRI-WP-155
T70K overheating – no discharge. WP was already replaced in AMTF – same problem.
5 FPCs are not conditionable and showing T70K overheating – shorted / disconnected from RF.

10. Main Coupler CW operation [1]
Qload increase is needed N
cavity
Qload.max 1
AC114
2.80×107 2
AC156
1.93×107 3
AC146
2.15×107 4
AC154
2.02×107 5
AC157
2.00×107 6
AC158
1.85×107 7
AC151
1.78×107 8
AC152
2.03×107
CMTB
CW run example
Pfor = (Eacc×L)2/(4×Rsh/Q×Qload)
limited by a motor/mechanics
Pfor ~ 1/Qload
test with 20 MV/m
simulated data
5 mm
T70K increases

11. Main Coupler CW operation [2]
Warm Part shift to achieve higher Qload
shift antenna by +5 mm outwards: Qload.max 2×107 → 5 ×107 1
8.4×107 2
6.8×107 3
1.1×107 4
9.1×107 5
7.5×107 6
6.6×107 7
2.6×107 8
achieved Qload.max
warm part / 70K window
shifted by 10 mm using
special iso-vac flange

12. Main Coupler CW operation [3] – Copper Plating
with the CW RF power in a range of kW inner conductor overheating will be an issue, with a maximum temperature on the warm inner conductor bellow;
to decrease the inner conductor temperature it must be gas cooled from inside or copper-plated with increased layer thickness (150µm) and RRR (50);
coupler tests with warm parts having 150µm copper re-plated inner conductor are in preparation.
inner conductor
outer conductor

13. Conclusions
Main coupler design for the European-XFEL linac is well established and proven;
Possible R&D could be push-rod and actuator design improvements;
Additional study should look for the warm parts overheating issue (E-XFEL 35 WPs replaced);
Main coupler CW application is another R&D direction – pushing to higher Qload and achieving better thermal properties with increased copper plating thickness are the possible applications.