1. FBCT experience at Desy Presenter: Matthias Werner, DESY At CERN Workshop: „Improving the accuracy of the BCT measurements in the LHC“ January 2011

2. About this presentation: zThis presentation consists mainly of information from our BCT expert Reinhard Neumann and shows developments of our former BCT expert Willi Radloff. zI just present their results.

3. FBCTs used at DESY zBergoz FBCTs e.g. for Petra accelerator for high bandwidth zInhouse development (by W.Radloff) e.g. for FLASH and XFEL accelerator for low beam position dependence.

4. FBCTs with reduced beam position dependence zThe monitors are equipped with 4 symmetric coils to minimize beam position dependence. zIn the original design, each coil has 10 turns. zIn the actual design for FLASH, each coil has 1 turn and an RF transformer (w1/w2 = 1/10), resulting in increased bandwidth (in the order of 100 MHz) for optimized design, but decreased amplitude. zThe currents of the 4 coils are added up by a summing network combined with an impedance matching transformer (improved version under development). zUnconfirmed tests showed an amplitude change of ±0.3% for position changes of ±6 mm – this could be further improved by trimming.

5. Pickup and Electronics

6. Scheme of Tunnel-Hardware TTF2

7. Halfcore in PEEK-Carrier

8. Both Halfcores Fully equipped beam transformer halfcores

9. Split Pickup Electrode Transformer halfcores embedded in shielding halfcovers

10. Toroid TTF2- Dump Complete, with cables, combiner and filter

11. Toroid in TTF2 Linac LP-Filter Combiner

12. Specification zToroids Toroid coreVitrovac 6030F, Vacuumschmelze Germany Dimension 1116 x 80 x 17,5 (mm) AL11,4 µH... 1,6 µH Dimension 2177 x 140 x 26,6 (mm) AL21,56 µH Constructionhalf-cut for clamp over mounting, 4 single-turn windings with integrated impedance matching transformation, 2 single-turn windings for calibration signal input Toroid core carrierPEEK(Polyaryletherketon), Type NW34/ NW44, 50 x 20 mm, DESY Hamburg Toroid housingAlMg3, Type NW34/ NW44, 190 x 160 mm, DESY Hamburg Bandwidth12,5 kHz...150 MHz 12 kHz... 200 MHz (typical) Output Impedance50 Ohms Sensitivity for 1nC Pulse8 mV-pp Output signal at 50 Ohms in configuration with 6dB Feedthrough attenuation and 15MHz LP-Gaussfilter zLP-Gaussian-Filter ConstructionSMD components, DESY TypeZ-matched LP-Filter, 12 th order -3dB Cut-off Frequency15 MHz (designed for 9 MHz Linac operation) Impedance50 Ohms Overshoot0,5 % Suppression40 dB at 70 MHz 60 dB at 115 MHz zPassive Power Combiner Constructionbrass, gold plated, DESY Connectors4x SMA Impedance50 Ohms z100 MHz Amplifier ConstructionSMD components, DESY Bandwidth1,8 kHz... 100 MHz Gain 40 dB (non-inverting) 36,5 dB setting for TTF2 Toroids Adjustable Gain setting  10% Max. Output Power  2 Vpp for 4 single Outputs Impedance In/Out50 Ohms Connectors In/OutSMA Supply Voltage  10V... 15V

13. Precision of BCTs at DESY Info from Reinhard Neumann: zUp to now, a precision of ≈1% (reproducibility, not absolute accuracy) of full range was achieved, this was enough for the Desy accelerators, so no further development was done to improve the precision.

14. What can affect precision and resolution of a BCT? zCalibration method zAmplitude stability of signal processing chain zTemperature dependence zBeam position dependence zBaseline suppression: Take zero reference from bunch gap or before every bunch? zLowpass filter directly after BCT or processing a high bandwidth signal? zInterference: symmetric or asymmetric data transmission? zSignal processing principle: integrating over a window or sampling a filtered signal? zNoise in the signal processing chain: Amplifier, Integrator (if any), ADC zDigital processing algorithm

15. Thank you!

16. Transparencies beyond the presentation at the workshop

17. Hardware Schematics of the PC based Bunch Current Monitor

18. 100 MHz Amplifier

19. Types of BCTs at DESY Info from Reinhard Neumann: zAt Petra (light source, max. planned bunch frequency = 500 MHz): yBCT: Bergoz type (Spec: BW=2GHz, rise time 175 ps) – only one coil yConsiderable position dependence observed yAcquisition: Oscilloscope (8 bit resolution) zFLASH accelerator (max. bunch frequency = 9 MHz): yBCT: „Radloff“-Type (selfmade with 4 coils) yAcquisition: VME ADC modules (14 bit, 2 samples per bunch) zOther accelerators and transport lines: yBCT: „Radloff“-Type (selfmade with 4 coils) yAcquisition: PC ADC boards

20. Impedance matching at BCT coil output zNumber of turns on the toroid and Impedance ratio of successive RF transformers determines impedance matching. zWhat is the source impedance of the coils? zFor some termination, we get optimum amplitude, for maximum bandwith we need a lower termination resistance. zWhat is the optimum for our case? (Especially important if resolution is limited by noise.)