Compton gamma stability - Status of the drift source investigation - Nobuhiro Terunuma, KEK March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop 1

Contents Jitter and drift of the Compton signal Beam? Laser? Utilities? March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop2

Drift of the Compton signal After December 2013, the beam size measurement by IPBSM has been difficult. Sometimes the unacceptable level of drift (50% or more) on the Compton signal was happened. It shows a periodic behavior about several tens of seconds. It accidentally fakes the modulation 0.2 or less. We can eliminate such measurements by paying attentions for the consistency of the modulation profile in a scan, and the repeatability in the several measurements. But it waste the time especially for the measurement of the bigger beam size in a crossing mode (lower modulation) and the Z-scan. Cherenkov Signal (ADC raw data) measuring the signals by keeping untouched Fringe Scan Fringe Scan March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop

Drift of the Compton signal – continued (1) - March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop The survey of the jitter/drift source has been continued. Stability of the electron beam There was no significant difference by ON/OFF the ATF2 orbit FB, 1000  x optics, QD0 current The long periodic drift was measured by the FF BPMs. No significant effect on the Compton signal was detected. The BPM’s drift is due to the temperature change of the DR cooling water. The extraction kicker used this water. The new cooling unit was installed to stabilize the kicker. It seems the Compton drift was somewhat improved but not so significant. – Even if so, we do not eliminate the possible drift caused by a beam. – The evaluation of the IP beam position is essential to understand the beam stability. Stability of the laser – show later Expanding the beam size at IP by QD0 reduces the amplitude of the drift. (Updated on 3/17/2014.)

Compton signal vs. FF BPM March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop5 y-position of the FF BPM (MFB2FF) Cherenkov signal by a laser wire

Contents Need to stabilize the cooling water unit for the EXT- kicker. As a temporal approach to stabilize, we adjusted three air condfitioners in the kicker hut to relax the cooling water unit. These air conditioners had different settings such as 25-deg cooling, 20-deg heating and maximum flow with unknown setting. The temperature is still drifting after a few hours but the beam dift at MFB2FF is weaken. March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop6 y-position of the FF BPM (MFB2FF) Temperature of the extraction kicker using the cooling water for DR magnet

Drift of the Compton signal - continued (2)- March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop Alignment of the Compton line is OK – If the Compton photons passes near the collimator, clipping of the signals will be enhanced by a beam jitter and drift. – The alignment of this line was already done during the new year shutdown by using the alignment laser. – Even if so, the x-y position scans for the collimators was done by using the Compton signals. No improvement was obtained. Compton gamma detector seems OK. Both CsI and Cherenkov detectors show the drift of the signal simultaneously. These detectors are controlled independently. – HV power supply, Charged ADC, gate and trigger The HV power supplies for these detector are connected to the same AC line. The specification of the HV stability is 5mV(p-p)/2kV or less and does not introduce the bigger signal drift.

Alignment of the Compton line March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop8 Removed #4, Laser ON Removed #4, Laser OFF with #4, Laser ON #1 #2 #3(x-y movable) #4 Gamma detector e- Collimator scan (x-y)

Cherenkov vs. CsI March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop 9 Cherenkov signal in the fringe scan CsI signal in the fringe scan

Drift of the Compton signal - continued (3)- March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop AC power stability for laser AC 200V, 50 Hz Plot the maximum voltage of 20 ms sampling Laser off 2% fluctuation due to the automatic TAP of KEK main transformer Laser turned ON additional and steady 2% fluctuation due to the laser operation 6 Hz 6 Hz laser

Stabilization of Laser March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop11

What is a source of the drift? Feb. 12 th 2014, KEK, 17 th ATF2 project meeting We suspected the temperature related sources because of the drift interval, several tens of seconds. – Cooling water; both external and internal – Clean booth Air conditioner No significant improvement was obtained.

Survey of the laser condition March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop The survey of the laser condition with the company engineer was done on Feb/10 th and 28 th. – Setting of the laser oscillator is fine as expected. – The cooling water unit is also working well and stable. – The signal drift was somewhat reduced by tuning the temperature of the Harmonic Generator (1064  532nm). – The flash-lamp needs to be exchanged about 30 M-shots (~3 months). Does it introduce the instability of laser? It was exchanged in October Next exchange will be soon, before April or May run. The optical-flat viewports were installed to decrease the air turbulence in the laser transport. The outer part of the laser image at the laser hut was jittered more than others. Clipping it makes profile better.

IPBSM Laser March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop Harmonic Generator 1064  532nm Rear Mirror (Oscillator Tuning) BeamLok (pointing stabilization) Water cooled N2 purged

Monitors for the IPBSM laser March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop Laser time structure (PIN-photo Diode) Laser image Injection (CCD camera) IP screen can be used for an enlarged image. (focal lens) Laser hut Laser image (CCD camera) Laser time structure (PIN-photo Diode) Laser image 174L (CCD camera)

Stability monitoring of the laser profile March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop hutV-table IP screen or 174L hut V-table IP screen or 174L Laser buildup, water temp No significant correlation between the drift of the weighted center of the CCD image and the Compton signal was observed.

Clipping the unstable part on the laser image March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop17 The profile was getting worse during the transport to IP. It was found that the outer ring at the laser hut fluctuated much compare to the inner rings. Clipping the outer ring makes the better LW profile than before. It has been applied since Mar 3 rd. Outer ring is jittered so much. Clip the outer ring by Iris in the laser hut Not saved sorry! It seems better but still looks triangle.

Which detector is best? Use both! CsI better stability (20%) than present Cherenkov (30%) need to maintain the non standard PMTs – Noisy or dead PMTs Cherenkov better S/N better linearity fast response – two bunch separation but poor stability especially for the lower signals March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop 18 We can use both detectors simultaneously.

500 ns/div Laser Beam ADC Gate (80 ns) Cherenkov ADC Gate (4 us) 50 ns/div 2000 ns/div Compton DAQ condition (2014/3/7); 4e9/bunch, but under IPBPM study CsI System checkout to understand the sensitivity of the recent measurements by Cherenkov and CsI. N.Terunuma CsI: 6 mV (peak) Cherenkov: 150 mV (peak)

Fringe scans by CsI March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop20 waiting the next scan but signal drift much

Signal fluctuation by detectors Jitter and drift of the Compton signal Beam? Laser? Utilities? March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop21 Cherenkov signals CsI signals Signal fluctuation is dominated by a statistical reason. Lack of the number of detected Compton photons.

CsI signals -Mar/12/2014- March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop 22 Noisy Dead Noisy OK? Dead

Shut the air flow in the laser transport by optical-flat windows March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop Laser IP transport on the shield block

Air turbulence in the transport March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop Just put a window on the hole to IP area on Feb.10 th (Monday) evening. No tools to store the data was prepared at that time. Monitoring only. Comparing the displayed history, It seems the CCD image was somewhat stabilized after putting a window as follows. Before After (30min later) It looks small and stable (flat). 20 min History of the laser image center on the CCD (at vertical table) Another side of the transport was also closed by a viewport. Therefore no air flow is expected in the laser transport since Feb. 12 th.

Summary March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop The drift of the Compton signal is undesirable on the beam size measurement. A lot of surveys to reduce the jitter and drift of the Compton signal were done. The situation has been improved somewhat but the source of the drift is not identified yet. The effort should be continued.

backup March 17th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop 26

Stability at the exit of laser head March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop PIN Photo Diode located 0.5m from the exit of laser head for timing adjustment for tuning of the laser oscillator Note: It measures a part of laser profile because a laser is 10 mm but a sensor is only 0.5 mm. Results Timing jitter – typically < 2 ns – sometimes < 4 ns Power jitter – resulting by the timing jitter – for beam (if sit on peak) typical 0.5/8div ~ 6% sometime1.5/8div ~ 20% Persistence 10 sec The waveform drifts sometime but it is fast and small. It does not explain the amount of a drift measured on the Cherenkov signal.

Stability at the exit of laser head –continued - March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop PIN Photo Diode 0.5 m from the exit of laser head No significant difference to the 10 sec measurement. It looks same as before; i.e. 65 nm measurement.  not sensitive to find the source of drift. Tuning of the laser oscillator improves/recovers the jitter level. Persistence ~ 600 sec

Stability at the Vertical Table - March 17 th, 2014, LAPP, FJPPL-FKPPL ATF2 Workshop PIN Photo Diode 20 m from the exit of laser head Amplitude of PIN-PD signal jitters about 30% at IP but a few Note: It measures a part of laser profile because a laser is 10 mm but a sensor is only 0.5 mm. It suggests the enhanced jitter after the transport. – pointing jitter – laser profile jitter Persistence ~ 60 sec Stripline