XFEL Tuner

22.04.23 Lutz Lilje DESY -MPY-

XFEL Tuner

• Lorentz Force Detuning System Setup• New Saclay design


General Remarks• Available today

– Old Saclay Tuner– Simple Single Piezo fixture

• 150-200 Hz compensation • Feedforward works well

– Piezo stacks• Desirable

– Two piezo fixture• Sensor-Actuator configuration

– Future Option: Feedback on sensor piezo?

– Bipolar operation to gain more stroke from Piezo• At cold stroke of piezo is smaller by up to a factor of 10…• … but the material is more forgiving

– New Saclay Tuner


General remarks II• Need engineering

– Detailed study on forces at the piezo position• Cold force sensor built by H1• Test in CHECHIA• Design of fixture/frame for 2 piezos

– Selection of best Piezo supplier• Material• INFN works on that

– Control system• Driver Amplifier modification for bipolar operation• Need automated procedures to setup Piezo

– Collaboration with INFN• Detuning measurement

– Preferably in one pulse– Warsaw colleagues are working on this


Lorentz Force Detuning System Setup

• Setup description

• Single pulse compensation

• Resonant excitation

• Problems


Lorentz Force Detuning System Setup

Function Generator

ADC

PZD Amplifier

Piezo actuator

Piezo sensor

PZM Amplifier

DO

OC

S se

rver

AmplifierLock-in or

Logarithmic

Chechia

MatlabFgen.m

Matlabchread.m

LP filter


Piezo Driver Amplifier (PZD)

-45.00

-40.00

-35.00

-30.00

-25.00

-20.00

-15.00

-4.7

0

-4.4

0

-4.1

0

-3.8

0

-3.5

0

-3.2

0

-2.9

0

-2.6

0

-2.3

0

-2.0

0

-1.7

0

-1.4

0

-1.1

0

-0.8

0

-0.5

0

-0.2

0

0.10

0.40

0.70

1.00

1.30

1.60

1.90

2.20

2.50

2.80

input voltage

gain

gain w ith offset

gain w ithout offset

maximum output voltage = -160.9 V

Specification:• Gain of the unloaded amplifier: –38 V/V (0.15),• Output voltage range from -160.9 V to +70 V, (voltage limiter)• Voltage offset: 0,11V• Output voltage rise and fall time: 200s (-160V +60V)• Maximal pulse current 2A (1ms)• Capacitive load 1,5F


Piezo Measurement Amplifier (PZM)

Specification:• Gain : 0.5 V/V, (adjustable),• Bandwidth: 100kHz,• Input impedance: 1k,• Output voltage range: ±5V, • Analog output 50 .


Piezoelectric tuner IM. Liepe, S. Simrock, W.D.-Moeller

Piezo

He-Tank

& Cavity

Tuning mechanism

Piezo-Actuator:

l=39mm

Umax=150V

l=3m at 2K

fmax,static=500Hz


Piezo Tuner setup II• Sensor-Actuator configuration• To compensate for Lorentz force detuning during the 1 ms RF pulse

Feed-Forward• To counteract mechanical noise, “microphonics”

Feed-Back


Drawing of current setup (H.-B. Peters)


RF signals at 35 MV/m

Blue: With piezo

Red: Without piezo


Piezo excitation of the cavity for frequency compensation (operation for 700 hours)

RF pulse (500us fill, 800us flat-top)


Single pulse compensation

Pulse Parameters: frequency = 219 Hz time delay = 0.84 ms amplitude = 95V


Single pulse compensation

MICROPHONICS


Damping of the ringing between pulses (5Hz operation)RF pulse RF pulse RF pulse


Frequency stabilization during RF pulse using a

piezoelectric tuner

Blue: With piezo

Red: Without piezo

Frequency detuning of 500 Hz compensated voltage pulse (~100 V) on the piezo. No resonant compensation


Option: Resonant excitation of a mechanical cavity resonance

• if one excites a mechanical resonance of a cavity with the piezo, one can use the cavity as an mechanical amplifier, so that a small stroke of the active element can compensate large detuning

• we have shown that with the excitation of three periods of the mechanical resonance frequency, about 1000 Hz could be compensated


Resonant excitation (stable for 200 hours)

Pulse Parameters: frequency = 219 Hz time shift = -9.5 ms amplitude = 24V offset = 24V


Resonant excitation


Frequency stabilization at 35 MV/m

Blue: With piezo

Red: Without piezo

Frequency detuning of ~1000 Hz compensated with resonant excitation of a mechanical cavity resonance at 230 Hz.

NOTE: This is rather an demonstration of the capability of active tuning. Application in a real machine is probably difficult/ impossible. Needs investigation.


Module Measurements


Single Piezo - Single Pulse Compensation • inside the module • f=200Hz, only 100Hz compensation


Single Piezo - Resonant


Problems with the active tuner

• Fundamental problem:– Preload at operating temperature not defined– Large tuning needed for both cavities tested in CHECHIA

(AC72, AC73)• ´natural´ frequency after tank welding is 780 kHz above 1,3 GHz• Normally this is more like 200-300 kHz

– This results in a very large force tearing on the piezo fixture

• Fixtures open up and piezos become loose


Force Measurement at the Piezo Position

• Design by Karsten Gadow (H1)– Available within 2-3 weeks

• Detailed measurement of the force in the cold environment– Calibration will be first done to liquid

nitrogen temperatures• Then design a fixture with sufficient

stiffness


Problems with active tuning• Single Piezo fixture

– So far only 100-200 Hz compensated (no resonant excitation of the cavity)

– Achieved compensation at 1,3 GHz – Alternative: Resonant mechanical excitation of the

cavity• Double Piezo fixture

– Has only been operated at 1,3 GHz + 600kHz– Needs a stiffer design– Alternative: Put 2 single Piezo fixtures at different posts of

the tuner• Bipolar operation of Piezos helps in any case

– Check in CHECHIA and INFN


Fixture Twist Problem

Fixture twistsduring operation


Problems with active tuning

• Automation is needed for operation of the piezos in the machine– Determination of optimum pulse shape– Interconnection to LLRF system


Available piezos (1/2)

• EPCOS

• NOLIAC

• Piezo Mechanik

• Physical Instruments


Available piezos (2/2)

Properties Units

EPC

OS

NO

LIA

C

Piez

o M

echa

nik

Phys

ical

In

stru

men

ts

Material name PZT-Nd34 PZT pz27 PZT 5H PZT 25 Young modulus: kN/mm2 51 45 55 35 Cross Section: mm2 7x7 10x10 7,5x7,5 10x10 Length: mm 30 30 55 36 Stroke (RT) µm 40 42 60 30 Resonance frequency kHz 66 14 70

Initial preload No No 400 No Calculated properties:

Stiffness: N/µm 83 150 56 97

Mec

hani

cal

Blocking force kN 3,2 6,3 4,0 3,0 Max voltage V 160 200 150 120 Control speed V/ms 1,6 Charge current A 20 Capacitance (unload) µF 2,1 5,7 13,4 12 (?) Capacitance (850N) µF 3,4 El

ectr

ical

Capacitance @10K µF 1,43

Special features displacement cryogenic

Material parameters yes yes no no

Mec

hani

cal

Elec

trica

l


New Saclay Tuner

• Design by P. Bosland• More compact design possible• Cavity should be pre-tuned so that the

tuner is not pushing• Piezo integrated• Available by the end of the year for first

tests


New Saclay Tuner 2


New Saclay Tuner 3


New Saclay Tuner 4


Conclusion• Proof-of-principle

– Single-piezo» Non-resonant:100Hz in the module» Resonant: 200 Hz in the module» (We can use this in module 6!)

– Double-piezo» Non-resonant: 400 Hz (not at 1.3GHz exact)» Resonant: 1kHz

– Test of bipolar operation

• Engeneering is needed– Choice of Piezo (-> together with INFN, IPN Orsay))– Stiffer Piezo fixture/frame for 2 Piezos

» Force measurement is underway (finished end 2004?)– LLRF integration

• New Saclay tuner tests needed– Available end 2004 / beginning 2005– Change of cavity pre-tuning desirable

XFEL Tuner

Documents

sensor piezo

voltage limiter voltage

output voltage range

piezoat cold stroke

unloaded amplifier

setup piezocollaboration

hz compensation feedforward

maximal pulse current