Coherent THz radiation at NewSUBARU NewSUBARU, LASTI, Y. Shoji - - PowerPoint PPT Presentation

Coherent THz radiation at NewSUBARU

NewSUBARU, LASTI, University of Hyogo

• Y. Shoji

Introduction to NewSUBARU

Circumference 118.7 m Injection Energy 1.0 GeV Electron Energy 0.5 - 1.5 GeV Type of Bending cell DBA with Inv.B RF Frequency 499.956 MHz Natural Emittance 38 nm (1GeV) Natural Energy Spread 0.047% (1GeV)

1 GeV linac NewSUBARU

R & D (1997) Photo-chemical(1998) New Material(1998) Inteferometry(1998) LIGA(1997) Microscope(2001) Optical Klystron Long Undulator Short Undulator EUVL(1997) LIGA(1997) LIGA(2004)

SPring-8 SR

Special design: Invert Bend

34o -8o +34o =60o

• 0.5

0.5 1 1.5 2

• 5
• 4
• 3
• 2
• 1

1 2 3 4 5 dispersion function (m) s (m) α1=0 α1=0.0014

BEND Q3 Q4 INV.BEND BEND Q3 Q4

Invert Bend Normal Bend Change η in the invert bends

• -> change α1

keeping achromatic condition Momentum compaction factor

∆L/L0 ≡ α1δ +α2δ2 + α3δ3 + . . . . δ≡ (Ε−Ε0)/Ε0

Control of momentum compaction factor

Momentum compaction factor

∆L/L0 ≡ α1δ +α2δ2 + α3δ3 + . . . . δ≡ (Ε−Ε0)/Ε0

α1 =1.3×10-3 → ≈ 0 Q3 & Q4 α2 = 0 SF, SD, (SB) α3 ≈ 0.5 no control knob α4 ≈ -20 no control knob

Types of CSR

Normal radiation CSR burst normal α1, high peak current

• -> WIRMS2007

Steady state CSR small α1 (quasi-isochronous ring) Beam physics ---> Instability threshold

• Acc. technology ---> stability problem

Linac Pulse CSR --> morning session No laser induced CSR NewSUBARU has bunch shortening limit What limit the bunch shortenning?

10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 10-4 10-3 10-2 10-1 100 101 102

multi-bunch; normal a1 single-bunch; normal a1 multi-bunch; low a1, high Vrf

relative radiation power / bunch stored beam current per bunch (mA)

Bunch shortenning at NewSUBARU

2 4 6 8 10 0.005 0.01 0.015 0.02

0.24 uA/bunch; 120kV 0.24 uA/bunch; 300kV 1.8 uA/bunch; 120kV 1.8 uA/bunch; 300kV

bunch length (σ; ps) /α1

VRF=120kV VRF=300kV

Measured by a streak camera

Deviation from the scaling law

σT ∝ α1 is valid for α1 > 2 ×10−5 σT −MIN ≈1.4ps at α1 ≈1×10−5

Instability ? No no IB dependence at IB<2µA Problem of monitor ? No α1 does not reduce σT VRF does reduce σT BESSY & ANKA do not have that problem!

Comparison of NewSUBARU with BESSY

BESSY II NewSUBARU

Natural Energy Spread 0.08% 0.047% Natural Emittance (nm rad) 30π 30π α1

• 1.4x10-6

5x10-6 α3

• 0.01

0.5 Damping time 8ms 12ms Lattice non-achromatic DB DBA+IB

Ripple & stability of magnets

* α1=4X10-6 --> σT=0.6 ps (E0=1GeV, VRF=300 keV)

∆τ = 0.1 ps < 0.02 deg RF with FB ∆α2 = 2E-4 ∆α2 = 2.6E-3 * < 2E-5 2.4E-4 (12bit) Sext-F ∆α1 = 6E-7 ∆α1= 4E-6* < 2E-6 1.5E-5 (16bit) Q4 ∆B/B= 1E-6 * < 1E-5 Inv Bend ∆E/E= 1E-6 < 1E-6 B ripple resolution ripple resolution Effect Power supply Magnet * stability condition; α1 +2α2δ + 3α3δ2>0

−−> α1> 4.5 X 10-6

270A 2.36A 2.36A main power supply Western arc IB Eastern arc IB

* Imbalance between B and IB

Effect of IB field ripple -- path-length ripple

Dipole field error produces longitudinal oscillation Deflection θS ∆x ∆L

• ->∆E

∆x(s) = βSβ(s) 2sinπν θS cos ψ(s) −ψS − πν

[ ]

∆L0 = x(s) ρ(s) ds

L0

∫

= ηSθ S

Evidence of path-length ripple (at normal operation)

x(s) = βSβ(s) 2sin πν cos ψ(s)− ψS − πν

[ ]− ηSη(s)

αL0 ⎧ ⎨ ⎪ ⎩ ⎪ ⎫ ⎬ ⎪ ⎭ ⎪ θ S

• 1.5
• 1
• 0.5

0.5 1 1.5 4 6 8 10 12

horizontal displacement ( µm ) BPM number

cosine component sine component

COD drift with harmonic freq. (60Hz, 120Hz, 180Hz, . . . .)

60Hz

agreed with expected COD for IB ripple ; ∆B/B <4X10-7

∆E E = − ηS αL0 θS

Pathlength ripple & RF ripple

Forced Oscillation

ε = (ωS

2

α1 ) ∆C + jω∆P ωS

2 −ω 2 + 2 jωαE

e jωt

dτ dt = −α1ε + ∆Ce jωt dε dt = ωS

2

α1 (τ + ∆Pe jωt) − 2αEε

τ = (2αE + jω)∆

C −ωS 2∆P

ωS

2 −ω 2 + 2 jωαE

e

jωt

pathlength ripple RF ripple

Pathlength ripple & RF ripple

10-8 10-7 10-6 10-5 10-4 10-3 10-2 10-1 10-6 10-5 10-4 10-3

relative energy spread momentum compaction factor

natural spread (σ ) pathlength fluctuation ( l ε l ) rf phase fluctuation ( l ε l )

(a)

10-3 10-2 10-1 100 101 102 10-6 10-5 10-4 10-3

bunch length (ps) momentum compaction factor

natural spread (σ ) pathlength fluctuation ( l

τ l )

rf phase fluctuation ( l τ l )

(b) (180 Hz)

Measured energy fluctuation (1)

• 90
• 80
• 70
• 60
• 50
• 40
• 30

0.5 1

20Hz (dBm) 60Hz (dBm) 120Hz (dBm) 180Hz (dBm) 240Hz (dBm) 360Hz (dBm) 20 60 120 180 240 360

∆X at BPM8 (dBm)

fs (kHz) 2 3 4 5

The oscillation is harmful BESSY & ANKA have no Invert Bend !

Feed-back control

FB works at normal α1 but not at small α1 because of non-linearity?

b i

• p
• l

a r A C p

• w

e r s u p p l y x y S i n g l e

• P

a s s B P M C i r c u i t F e e d

• B

a c k C

• n

t r

• l

l e r S t

• r

a g e R i n g B P M D i p

• l

e M a g n e t

Energy feed-back

Feed-back OFF ON (below; with LPF; x2)

Non-symmetry for +∆E and -∆E

The bunch length was not always shortest at where the fS was the

• smallest. It strongly depended on δ (or ∆fRF) at small α1 (α1 = 1×10-5).

10 20 30 40 499955.480 kHz 499955.500 kHz 499955.520 kHz 1 2 3 4 5

relative strenghth time (ps)

bunch shape. FFT spectrum of the beam signal

Synchrotron Oscillation is enhanced at δ >0

300 400 500 600 700 800 900

• 30 -20-10 0 10 20 30

fS (kHz) ∆fRF (Hz)

+15Hz

Non-symmetry for +∆E and -∆E

Non-symmetry has stored current dependence

Current dependence of fs side-band. fRF peak is normalized to 0dB. White noise is subtracted from the data at 0.02mA

• 100
• 80
• 60
• 40
• 20
• 1000
• 500

500 1000 30Hz 50Hz

FFT power (dB) f - fRF (Hz)

0.47mA 0.02mA

• 100
• 80
• 60
• 40
• 20
• 1000
• 500

500 1000 50Hz 70Hz

FFT power (dB) f - f

RF (Hz)

0.47mA 0.02mA

499955530 Hz 499955550 Hz

Conclusion

・Some problems in QI operation are enhanced at NewSUBARU probably because of the Invert Bend. ・Ring stability is a main technical point of QI operation ・Higher order α would be the other problem ・Still there is a phenomenon not understood.