Post on 03-Feb-2022
Thomas Jefferson National Accelerator FacilityOperated by the Southeastern Universities Research Association for the U.S. Department of Energy
PONDEROMOTIVE INSTABILITIES AND MICROPHONICS
A TUTORIAL
Jean Delayen
Thomas Jefferson National Accelerator Facility
SRF 2005 Cornell 10-15 July 2005
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 2
Some Definitions
• Ponderomotive effects: changes in frequency caused by the electromagnetic field (radiation pressure)—Static Lorentz detuning (cw operation)—Dynamic Lorentz detuning (pulsed operation)
• Microphonics: changes in frequency caused by connections to the external world—Vibrations—Pressure fluctuations
Note: The two are not completely independent.When phase and amplitude feedbacks are active, ponderomotive effects can change the response to external disturbances
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 3
Some History• M. M. Karliner, V. E. Shapiro, I. A. Shekhtman
Instability in the Walls of a Cavity due to Ponderomotive Forces of the Electromagnetic FieldSoviet Physics – Technical Physics, Vol. 11, No. 11, May 1967
• V. E. ShapiroPonderomotive Effects of Electromagnetic RadiationSoviet Physics JETP, Vol. 28, No. 2, February 1969
• M. M. Karliner, V. M. Petrov, I. A. ShekhtmanVibration of the Walls of a Cavity Resonator under Ponderomotive Forces in the Presence of FeedbackSoviet Physics – Technical Physics, Vol. 14, No. 8, February 1970
Stability conditions derived by comparing the rate of transfer of energy from the electromagnetic mode to the mechanical mode with the dissipation rate of the mechanical mode.Analysis valid when decay time of the electromagnetic mode much less than period of mechanical mode (τ Ωµ << 1), or when the rate of transfer of energy is very high
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 4
Some More History
• D. SchulzeMechanical Instabilities of a Superconducting Helical Structure due to Radiation PressureProc. 1970 Proton Linac Conference, Batavia
Ponderomotorishe Stabilität von Hochfrequenzresonatoren und ResonatorregelungssystemenDissertation, KFK-1493, December 1971Ponderomotive Stability of R.F. Resonators and Resonator Control Systems (ANL-TRANS-944, November 1972)
Analysis of stability of generator-driven superconducting resonator (arbitrary τ Ωµ ) using control systems methods (Laplace transforms, transfer functions) with and without phase and amplitude feedbackAnalysis of damping of mechanical vibrations by ponderomotive forces
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 5
Some More History
• J. R. DelayenPhase and Amplitude Stabilization of Superconducting ResonatorsDissertation, Caltech, 1978
Analysis of ponderomotive instabilities of resonators operated in self-excited loops, with and without feedback, using control systems methods.
Included I-Q detection
Used methods of stochastic analysis to quantify performance of feedback system in the presence of microphonics and ponderomotive effects
Later extended to include effects due to presence of beam loading and electronic damping.
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 6
Ponderomotive Effects
• Adiabatic theorem applied to harmonic oscillators (Boltzmann-Ehrenfest)
2
/
2 20 0
1If = 1, then is an adiabatic invariant to all orders
( ) (Slater)
Quantum mechanical picture: the number of photons is constant:
( ) ( )4 4
d
VV
d Udt
UU U o eU
U N
U d H r E r
ε
ωεω ω
ωω ω ω
ω
µ ε
∆ ∆∆ -Ê ˆ Ê ˆ fi =Ë ¯ Ë ¯
=
È= +ÎÚ
∼
2 20 0
(energy content)
( ) ( ) ( ) ( ) (work done by radiation pressure)4 4S
U dS n r r H r E rµ εξ∆
˘Í ˙
È ˘= - ◊ -Í ˙Î ˚Ú
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 7
Ponderomotive Effects
2 20 0
2 20 0
( ) ( ) ( ) ( )4 4
( ) ( )4 4
Expand wall displacements and forces in normal modes of vibration ( ) of the resonator
( ) ( )
( ) (
S
V
vS
V
dS n r r H r E r
d H r E r
r
dS r r
r q r
µ
µ µν
µ µµ
µ εξωµ εω
φ
φ φ δ
ξ φ
∆È ˘◊ -Í ˙Î ˚= -
È ˘+Í ˙Î ˚
=
=
Ú
Ú
Ú
 ) ( ) ( )
( ) ( ) ( ) ( )
S
S
q r r dS
F r F r F F r r dS
µ µ
µ µ µ µµ
ξ φ
φ φ
=
= =
Ú
 Ú
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 8
Ponderomotive Effects
2
2
2
Equation of motion of mechanical mode
(Euler-Lagrange)
1 (elastic potential energy) : elastic constant2
1= (kinetic energy) 2
d L L F L T Udt q q q
U c q c
qT c
µµ µ µ
µ µ µµ
µµ
µ µ
µ
Φ
ΩΩ
∂ ∂ ∂- + = = -∂ ∂ ∂
= Â
Â2
2
22
: frequency
= (power loss) : decay time
2
c q
q q q Fc
µ
µ µµ
µ µ µ
µµ µ µ µ µ
µ µ
ττ
τ
ΦΩ
ΩΩ+ + =
Â
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 9
Ponderomotive Effects
22 2 2 20
0 0
The frequency shift caused by the mechanical mode is proportional to
2
Total frequency shift: ( ) ( )
Static frequency shift:
q
FU k V
c U
t t
µ µ
µµ µ µ µ µ µ µ
µ µ
µµ
µµ
ω µ
ωω ω ωτ
ω ω
ω ω
∆
∆ ∆ Ω ∆ Ω Ω
∆ ∆
∆ ∆
Ê ˆ+ + = - = -Á ˜Ë ¯
=
=
Â2
Static Lorentz coefficient:
V k
k k
µµ
µµ
= -
=
 Â
Â
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 10
Ponderomotive Effects – Mechanical Modes
Fluctuations around steady state:0 (1 )
o
V V vµ µ µω ω δω
δD = D +
= +
Linearized equation of motion for mechanical mode:
2 2 22 2 ok V vµ µ µ µ µ µµ
δω δω δω δτ
+ +W = - W
The mechanical mode is driven by fluctuations in the electromagnetic mode amplitude.
Variations in the mechanical mode amplitude causes a variation of the electromagnetic mode frequency, which can cause a variation of its amplitude.
⇒ Closed feedback system between electromagnetic and mechanical modes, that can lead to instabilities.
( )2 2 22ok V n tµ µ µ µ µ µ
µ
ω ω ωτ
∆ ΩD + D +W = - +
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 11
Lorentz Transfer Function
TEM-class cavities ANL, single-spoke, 354 MHz, β=0.4
simple spectrum with few modes
( )
2 2 2
2 2
2 2
2 2
2( ) ( )2
o
o
k V v
k Vv
i
µ µ µ µ µ µµ
µ µµ
µµ
δω δω δω δτ
δω ω δ ωω ω
τ
+ +W = - W
- W=
W - +
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 12
Lorentz Transfer Function
SNS Med β Cryomodule 3, Cavity Position 1, Lorentz Transfer Function (5MV/m CW)
25
30
35
40
45
50
55
60
65
70
0 60 120 180 240 300 360 420
AM Drive Frequency (Hz)
Cav
ity D
etun
ing,
Res
pons
e R
atio
(dB
)
0
45
90
135
180
225
270
315
360
Rel
ativ
e Ph
ase
(deg
)
TM-class cavities (Jlab, 6-cell elliptical, 805 MHz, β=0.61)Rich frequency spectrum from low to high frequenciesLarge variations between cavities
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 13
GDR and SEL
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 14
Generator-Driven Resonator
• In a generator-driven resonator the coupling between the electromagnetic and mechanical modes can lead to two ponderomotive instabilities
• Monotonic instability : Jump phenomenon where the amplitudes of the electromagnetic and mechanical modes increase or decrease exponentially until limited by non-linear effects
• Oscillatory instability : The amplitudes of both modes oscillate and increase at an exponential rate until limited by non-linear effects
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 15
Generator-Driven Resonator
2 12oy k Vµ τ
- <
• Oscillatory:( )
( )
22 22
2 2
112
where : normalized detuning
o
g co
y k V
y
µµ
µ µ
ττ τ
τ ω ω
+ W<
W
= -
Approximate stability criteria in the absence of feedback:
The monotonic instability can occur on the low frequency side when the Lorentz detuning is of the order of an electromagnetic bandwidth.The oscillatory instability can occur on the high frequency side when the Lorentz detuning is of the order of a mechanical bandwidth.
Amplitude feedback can stabilize system with respect to ponderomotive instabilities
• Monotonic:
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 16
Self-Excited Loop
• Resonators operated in self-excited loops in the absence of feedback are free of ponderomotive instabilities. An SEL is equivalent to the ideal VCO.—Amplitude is stable —Frequency of the loop tracks the frequency of the cavity
• Phase stabilization can reintroduce instabilities, but they are easily controlled with small amount of amplitude feedback
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 17
Self-Excited Loop
where: tan no beam loading , 1, 1l ay k kϕ µµ
τθ ττ
Ω=
The stability boundary can be pushed arbitrarily far with amplitude feedback
General stability criteria without above restrictions (including beam loading) exist
Approximate stability criteria with phase feedback
• Monotonic:
• Oscillatory:
20
12aky k Vµ τ+
- <2
20 2 2
( 1)12 ( 1)
a
a
k ky k V
k kϕ
µµ µ ϕτ τ Ω
+<
+ +
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 18
Input-Output Variables
• Generator - driven cavity
• Cavity in a self-excited loop
Detuning (ω - ωc)
Field amplitude (Vo)Generator amplitude (Vg)
Cavity phase shift (θl)
Ponderomotive effects
Loop phase shift (θl)
Field amplitude (Vo)Limiter output (Vg)
Loop frequency (ω)
Pondermotive effects
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 19
Input-Output Variables Generator-Driven Resonator
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
-6 -5 -4 -3 -2 -1 0 1 2
Detuning (Norm.)
Am
plitu
de (N
orm
.)
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
-6 -5 -4 -3 -2 -1 0 1 2
Detuning (Norm.)
Cai
ty P
hase
Shi
ft
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 20
Input-Output Variables Self-Excited Loop
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
Loop Phase Shift
Am
plitu
de (N
orm
.)
-6.0
-5.0
-4.0
-3.0
-2.0
-1.0
0.0
1.0
2.0
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
Loop Phase Shift
Det
unin
g (N
orm
.)
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 21
Ponderomotive Instabilities in GDR
0 / 2 72 MHz, 9 msec, 18 Hz/ 2 89 Hz, 85 sec, / 2 500 kHzkµ µ µ
ω π τ ωπ τ π
∆Ω
= = == = =
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 22
Ponderomotive Instabilities in SEL
2 2dY dX dX Y V Vdt dt dt
ϕ δω- = =
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 23
Ponderomotive Instabilities in SEL
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 24
Microphonics
• Microphonics: changes in frequency caused by connections to the external world—Vibrations—Pressure fluctuations
When phase and amplitude feedbacks are active, ponderomotive effects can change the response to external disturbances
( )2 2 22 2 ok V v n tµ µ µ µ µ µµ
δω δω δω δτ
Ω+ +W = - +
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 25
Microphonics
Two extreme classes of driving terms:
• Deterministic, monochromatic—Constant, well defined frequency—Constant amplitude
• Stochastic—Broadband (compared to bandwidth of mechanical mode)—Will be modeled by gaussian stationary white noise
process
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 26
Microphonics (probability density)
SNS M02, CAVITY 3 BACKGROUND MICROPHONICS HISTOGRAM
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
-20 -15 -10 -5 0 5 10 15 20
Cavity Detuning (Hz)
Prob
abili
ty o
f Occ
urre
nce
Std Dev = 2.2 Hz
SNS M03, CAVITY 1 BACKGROUND MICROPHONICS HISTOGRAM
0
0.01
0.02
0.03
0.04
0.05
0.06
-20 -15 -10 -5 0 5 10 15 20Cavity Detuning (Hz)
Prob
abili
ty o
f Occ
urre
nce
Std Dev = 5.6 Hz
Single gaussian
Noise driven
Bimodal
Single-frequency driven
Multi-gaussian
Non-stationary noise
805 MHz TM 805 MHz TM 172 MHz TEM
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 27
Microphonics (frequency spectrum)
TM-class cavities (JLab, 6-cell elliptical, 805 MHz, β=0.61)Rich frequency spectrum from
low to high frequenciesLarge variations between cavities
SNS M02, Cavity 3, Bkgnd Microphonics Spectrum, 1W
1.E-03
1.E-02
1.E-01
1.E+00
1.E+01
0 60 120 180 240 300 360
Frequency (Hz)
RM
S A
mpl
itude
(Hz
rms)
TEM-class cavities (ANL, single-spoke, 354 MHz, β=0.4)Dominated by low frequency (<10 Hz) from pressure fluctuations
Few high frequency mechanical modes that contributelittle to microphonics level.
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 28
Probability Density (histogram)
0
0.5
1
1.5
-1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2
δω/δωmax
p(δω
)
0
0.1
0.2
0.3
0.4
0.5
-3 -2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5 3
δω/δωσ
p(δω
)
Harmonic oscillator (Ωµ,τµ) driven by:
Single frequency, constant amplitude White noise, gaussian
2 2max
1( )p δωπ δω δω
=-
212
1( ) exp2
pω
δωσω
δωσ π
È ˘Ê ˆÍ ˙Á ˜Í ˙Á ˜Ë ¯Í ˙Î ˚
-=
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 29
Autocorrelation Function
-1
-0.5
0
0.5
1
0 2 4 6 8 10
τ
r(τ)
-1
-0.5
0
0.5
1
0 2 4 6 8 10
τ
r(τ)
( )( ) cos( )(0) d
RrR
δωδω
δω
ττ ω τ= = /( )( ) cos( )(0)
Rr eR
µτ τδωδω µ
δω
ττ τΩ -= =
0
1( ) ( ) ( ) lim ( ) ( )T
x TR x t x t x t x t dt
Tτ τ τ
Æ•= + = +Ú
Single frequency, constant amplitude White noise, gaussianHarmonic oscillator (Ωµ,τµ) driven by:
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 30
Stationary Stochastic Processes
x(t): stationary random variable
Autocorrelation function:
Spectral Density Sx(ω): Amount of power between ω and dω
Sx(ω) and Rx(τ) are related through the Fourier Transform (Wiener-Khintchine)
Mean square value:
0
1( ) ( ) ( ) lim ( ) ( )T
x TR x t x t x t x t dt
Tτ τ τ
Æ•= + = +Ú
1( ) ( ) ( ) ( )2
i ix x x xS R e d R S e dωτ ωτω τ τ τ ω ω
π• •-
-• -•= =Ú Ú
2 (0) ( )x xx R S dω ω•
-•= = Ú
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 31
Stationary Stochastic Processes
For a stationary random process driving a linear system
( ) ( ) ( ) ( )
( ) ( ) [ ][ ] [ ]
( )
2 2
2
22
0 0
: auto correlation function of ( ) ( )
( ) ( ) : spectral density of ( ) ( )
( ) ( ) ( )
( )
y y x x
y x
y x
y x
x
y R S d x R S d
R R y t x t
S S y t x t
S S T i
y S dT i
ω ω ω ω
τ τ
ω ω
ω ω ω
ω ωω
+• +•
-• -•
+•
-•
= = = =
È ˘Î ˚
=
=
Ú Ú
Ú
( )x t ( )y t( )T iω
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 32
Performance of Control SystemResidual phase and amplitude errors caused by microphonicsCan also be done for beam current amplitude and phase
fluctuations
2
Assume a single mechanical oscillator of frequency and decay time
excited by white noise of spectral density Aµ µτΩ
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 33
Performance of Control System
( )
( ) ( )( )
( ) ( )( )
2
22 2 2 222 2 2
22
22 2 22
22
22 2 22 2
2
2
2
2
2
2
ex
a
a ex
ex
d
G i di
G i
G i G i d di
G i
G i G i d di
A A A
v A
A
µµ
µ µµ
µµ
µµµ
ϕµ
µ ϕµµ
µ
ωω ω
ω ωτ
ω
ω ω ω ωω ωτ
ω
ω ω ω ωω ω
τ
πτδω
δ δωπτ
δϕ δωπτ
Ω
Ω
Ω
Ω
Ω
Ω
+• +•
-• -•
+• +•
-• -•
+• +•
-• -•
- + +
- + +
- + +
< >= = =
< >= = < >
< >= = < >
Ú Ú
Ú Ú
Ú Ú
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 34
The Real WorldProbability Density
0.00001
0.0001
0.001
0.01
0.1
-15 -10 -5 0 5 10 15
Hz
Hz-1
Probability Density
-0.005
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
-15 -10 -5 0 5 10 15
Hz
Hz-1
Microphonics
-5
-4
-3
-2
-1
0
1
2
3
4
5
0 0.02 0.04 0.06 0.08 0.1
Time (s)
Hz
Normalized Autocorrelation Function
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 0.1 0.2 0.3 0.4
Sec
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 35
The Real WorldProbability Density
-0.005
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
-10 -5 0 5 10
Hz
Hz-1
Microphonics
-10
-8
-6
-4
-2
0
2
4
6
8
10
0 0.02 0.04 0.06 0.08 0.1
Time (sec)
Hz
Probability Density
0.00001
0.0001
0.001
0.01
0.1
-10 -5 0 5 10
Hz
Hz-1
Normalized Autocorrelation Function
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 0.1 0.2 0.3 0.4
Time (sec)
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 36
The Real WorldProbability density
0.00001
0.0001
0.001
0.01
0.1
-20 -15 -10 -5 0 5 10 15 20
Hz
Hz-1
Microphonics
-10
-8
-6
-4
-2
0
2
4
6
8
10
0 0.02 0.04 0.06 0.08 0.1
Time (sec)
Hz
Normalized Autocorrelation Function
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
0 0.2 0.4 0.6 0.8 1
Time (sec)
Probability density
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
-20 -15 -10 -5 0 5 10 15 20
Hz
Hz-1
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 37
Piezo control of microphonics
MSU, 6-cell elliptical 805 MHz, β=0.49
Adaptive feedforward compensation
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 38
Piezo Control of Microphonics
FNAL, 3-cell 3.9 GHz
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 39
Piezo Control of Microphonics
LANL 402.5 MHz scruncher cavity, QL= 109
Phase-locked with piezo only
Operated by the Southeastern Universities Research Association for the U.S. Department of Energy
Thomas Jefferson National Accelerator Facility Page 40
Final Words
• Microphonics and ponderomotive instabilities issues in high-Q SRF cavities were “hot topics” in the early days (~70s), especially in low-β applications
• They were solved and are well understood• They are being rediscovered in medium- to high-β
applications• Today’s challenges:
—Large scale (cavities and accelerators): need for optimization—Finite beam loading
• Small but non-negligible current (e.g. RIA)• Low current resulting from the not quite perfect cancellation of 2 large
currents (ERLs)