Bunch-by-bunch Feedback at the MLS - Dimtel Bunch-by-bunch Feedback at the MLS Updates and Beam...

Setup Beam Studies Summary

Bunch-by-bunch Feedback at the MLSUpdates and Beam Studies

F. Falkenstern1, J. Feikes1, M. Ries1, A. Shaelicke1, D.Teytelman2, et. al.

1HZB, Berlin, Germany2Dimtel, Inc., San Jose, CA, USA

February 13, 2014


Outline

1 SetupSystem Updates

2 Beam StudiesBeam Transfer FunctionBunch CleaningLongitudinal Feedback on The RampLongitudinal Grow/Damp MeasurementsPost-mortem Data


Work Summary

Updated all 4 iGp12 units to a new gateware/softwarerelease:

Feedback filter lengthened from 16 to 24 taps;BRAM acquisition memory increased from 192k to 324k;Single-bunch acquisition engine added, BTF functionality;Trigger capture, polarity;Front-end phase tracking;Time-domain modulation of drive signal.

Performed timing offset calibrations, should simplify futureupdates;Found front/back-end DAC unresponsive (controls phaseshifters, fan speed). Need to replace GPIO cable, willprovide a standard short cable.


What Was Done When

Day by DayJanuary 8 Calibration of Y and SPARE units, a bit of tuning in

the longitudinal plane, some testing with beam,BESSY II work.

January 9 Updated and calibrated Z unit, timed and phasedlongitudinal and horizontal, optimizedbunch-to-bunch isolation, demonstrated bunchcleaning, single bunch transfer function.

January 10 Timed and phased vertical plane, set uppost-mortem diagnostics, Z feedback on the ramp,growth rate studies, 10 ps optics.


Outline




Longitudinal Beam Transfer Function

85 90 95 100 105 110 115 120 125 130−70

−60

−50

−40

−30

−20

−10

Frequency (kHz)

Mag

nitu

de (

dB)

Peak −16.2 dB at 105.4 kHz

85 90 95 100 105 110 115 120 125 130−50

0

50

100

150

200

250

Frequency (kHz)

Pha

se (

deg)

Center phase 90.0 deg @ 105.3 kHz

Open-loop BTF;Shift gain of 0;Shift gain of 1;Shift gain of 2;Shift gain of 3;Shift gain of 4;Shift gain of 5;Shift gain of 6;BTF peak vs. feedbackgain.



85 90 95 100 105 110 115 120 125 130−70

−60

−50

−40

−30

−20

−10

Frequency (kHz)

Mag

nitu

de (

dB)


85 90 95 100 105 110 115 120 125 130−50

0

50

100

150

200

250

Frequency (kHz)

Pha

se (

deg)





−45 −40 −35 −30 −25 −20 −15 −10 −5 0−60

−55

−50

−45

−40

−35

−30

−25

−20

−15

Feedback attenuation (dB)

BT

F p

eak

(dB

)

MeasuredEstimated



Outline




Bunch Cleaning, Vertical Plane

After re-tuning checkedbunch cleaning at injectionand at the top of the ramp;With the back-endoptimized see goodisolation bunch-to-bunch;Spelling MLS in Morsecode here.


Bunch Cleaning, Vertical Plane

15 20 25 30 35 40 45 50 550

50

100

150

200

250

300

Bunch number

Bun

ch c

urre

nt (

arb.

uni

ts)

After re-tuning checkedbunch cleaning at injectionand at the top of the ramp;With the back-endoptimized see goodisolation bunch-to-bunch;Spelling MLS in Morsecode here.


Outline




Limiting Factors in the Longitudinal Plane

Found two limiting effects:Downward frequency shift of mode 0 due to beam loading;Synchrotron frequency change from 85 to 105 kHz.

An example from the the ALS;Mode 0 shifts from 12 to4.8 kHz.


Limiting Factors in the Longitudinal Plane

Found two limiting effects:Downward frequency shift of mode 0 due to beam loading;Synchrotron frequency change from 85 to 105 kHz.

−12 −10 −8 −6 −4 −2 04

5

6

7

8

9

10

11

12

13

14

ℜ (ms−1)

ℑ (

kHz)

400 mA

0 mA

An example from the the ALS;Mode 0 shifts from 12 to4.8 kHz.


Minimum Delay Filter

0 5 10 15 20 25−1

−0.5

0

0.5

1

Tap number

0 50 100 150 200 250 300 350 400 450 500−30

−20

−10

0

10

20

30

Frequency (kHz)

Gai

n (d

B)

0 50 100 150 200 250 300 350 400 450 500100

120

140

160

180

200

Frequency (kHz)

Pha

se (

deg)

Used a 24 tap filter with nodownsampling;Roughly half synchrotronperiod long;Filter gain peaks around270 kHz, but the beam has noresponse there;Only 14◦ phase shift between85 and 105 kHz;Gain at 25 kHz is 20 dB downand 50◦ away from resistivephase.


Longitudinal Feedback and Ramping

Minimum turn-on energy is 150 MeV ...... but the lifetime suffers;Converged on feedback turn-on at 200 MeV;Use phase servo to maintain front-end phase detection;Seemed fairly stable in the short testing we haveperformed;Long term experience?


Outline




Longitudinal Growth Rates vs. Beam Current (1/3)

60 80 100 120 140 160 1801.5

2

2.5

3

3.5

4

4.5

Beam current (mA)

Gro

wth

rat

e (m

s−1 )

MLS Z: Mode 43, threshold 18 mA, zero current damping 1.8 ms

60 80 100 120 140 160 18082.9

82.95

83

83.05

83.1

83.15

Beam current (mA)

Mod

al fr

eque

ncy

(kH

z)

Mode 43 open-loopeigenvalues vs. beam current;Threshold of 18 mA, zerocurrent damping of 1.8 ms;



60 80 100 120 140 160 180

0.8

1

1.2

1.4

1.6

1.8

2

Beam current (mA)

Gro

wth

rat

e (m

s−1 )


60 80 100 120 140 160 18082.7

82.75

82.8

82.85

82.9

82.95

Beam current (mA)

Mod

al fr

eque

ncy

(kH

z)




60 80 100 120 140 160 1800.5

1

1.5

2

2.5

Beam current (mA)

Gro

wth

rat

e (m

s−1 )


60 80 100 120 140 160 18082.75

82.8

82.85

82.9

82.95

83

Beam current (mA)

Mod

al fr

eque

ncy

(kH

z)



Longitudinal Growth Rates: Historical Comparison

0 20 40 60 80 100 120 140 160 1800

0.5

1

1.5

2

2.5

Beam current (mA)

Gro

wth

rat

e (m

s−1 )

Mode 12

0 20 40 60 80 100 120 140 160 1800

1

2

3

4

5

Beam current (mA)

Gro

wth

rat

e (m

s−1 )

Mode 43

2014−01−102011−11−26

2014−01−102011−11−26

0 20 40 60 80 100 120 140 160 1800

0.5

1

1.5

2

2.5

Beam current (mA)

Gro

wth

rat

e (m

s−1 )

Mode 71

2014−01−102011−11−26

A clear reduction for modes12 and 43, mode 71unchanged.


Outline




Beam Loss: Longitudinal Instability

−3 −2.5 −2 −1.5 −1 −0.5 0−1500

−1000

−500

0

500

1000

1500

Time (ms)

AD

C c

ount

s

Longitudinal

−3 −2.5 −2 −1.5 −1 −0.5 0−100

−50

0

50

100

150

Time (ms)

AD

C c

ount

s

Horizontal

−3 −2.5 −2 −1.5 −1 −0.5 0−200

−100

0

100

200

300

400

Time (ms)

AD

C c

ount

s

Vertical

Beam loss due tolongitudinal instability;Relatively slow loss over0.5 ms;Spectrum is dominated bysynchrotron frequency andharmonics;A faint knockout line isvisible;Quadrupole/sextupoleoscillation or harmonics?


Beam Loss: Longitudinal Instability

Beam loss due tolongitudinal instability;Relatively slow loss over0.5 ms;Spectrum is dominated bysynchrotron frequency andharmonics;A faint knockout line isvisible;Quadrupole/sextupoleoscillation or harmonics?


Beam Loss: Vertical Oscillation

−12 −10 −8 −6 −4 −2 00

100

200

300

400

500

Time (ms)

AD

C c

ount

s

Longitudinal

−12 −10 −8 −6 −4 −2 0−400

−200

0

200

400

600

800

Time (ms)

AD

C c

ount

s

Horizontal

−12 −10 −8 −6 −4 −2 00

100

200

300

400

Time (ms)

AD

C c

ount

s

Vertical

During ramp down thevertical tune runs into theknockout line;Fast loss — roughly 100turns;The vertical tune shifts to1380 kHz;Fast loss produces multiplespectral features.



−4.08 −4.06 −4.04 −4.02 −4 −3.98 −3.96 −3.94 −3.92 −3.90

100

200

300

400

500

Time (ms)

AD

C c

ount

s

Longitudinal

−4.08 −4.06 −4.04 −4.02 −4 −3.98 −3.96 −3.94 −3.92 −3.9−400

−200

0

200

400

600

800

Time (ms)

AD

C c

ount

s

Horizontal

−4.08 −4.06 −4.04 −4.02 −4 −3.98 −3.96 −3.94 −3.92 −3.90

100

200

300

400

Time (ms)

AD

C c

ount

s

Vertical



Summary

Successfully updated all systems;Demonstrated longitudinal feedback operation through theenergy ramp and optics transition;Configured and tested post-mortem data acquisition, beamtransfer function measurements, bunch cleaning;Longitudinal growth rates have shifted slightly since 2011.

Bunch-by-bunch Feedback at the MLS - Dimtel Bunch-by-bunch Feedback at the MLS Updates and Beam...

Documents

Transcript of Bunch-by-bunch Feedback at the MLS - Dimtel Bunch-by-bunch Feedback at the MLS Updates and Beam...