Beam Loading Studies in BEPC2 - DimtelBeam Loading Studies in BEPC2 Junhui Yue1, Jianping Dai1, Yuan Zhang1, Haipeng Wang2, D. Teytelman3 1IHEP, Beijing, China 2JLAB, Newport News,

Beam Loading Studies in BEPC2

Junhui Yue1, Jianping Dai1, Yuan Zhang1, Haipeng Wang2, D.Teytelman3

1IHEP, Beijing, China2JLAB, Newport News, VA, USA

3Dimtel, Inc., San Jose, CA, USA

December 18, 2016

(IHEP,JLAB,Dimtel) Beam Loading Studies in BEPC2 2016-12-18 1 / 23

RF Transfer Functions

Cavity Transfer Functions

Start from the last measurement we made (2016-12-11);Open and closed-loop transfer functions measured using anetwork analyzer.



Cavity Transfer Functions

Start from the last measurement we made (2016-12-11);Open and closed-loop transfer functions measured using anetwork analyzer.



Measurement Goal and Conditions

The goal is to estimate direct loop gain at the nominal operatingpoint;A difficult measurement — need to detect small NWA excitation inpresence of large RF fundamental signal;

Field setpoint 222 kV — as low as possible to reduce fundamentalsignal;Cavity detuned by setting load angle offset to −40◦;Amplitude and phase loops turned off;Measurements with direct loop open and closed.

No beam, of course.






No beam, of course.






No beam, of course.



System Model

∑

Excitation

Cavity field probe

Hfb

Feedback

Klystron and cavity

Hcav(ω)

V1

V2

Cavity response: Hcav(ω) = 2iσωω2−2iσω−ω2

rGe−i(ω−ωrf)τeiφ0

Five parameters: gain G, damping rate σ, center frequency ωr ,delay τ , and phase shift φ0;Feedback response is just gain and phase shift: Hfb(ω) = Gfbeiφfb ;In open loop estimate the parameters of Hcav(ω);Two parameter fit (Gfb, φfb) to the closed-loop S21(ω).



System Model

∑

Excitation

Cavity field probe

Hfb

Feedback

Klystron and cavity

Hcav(ω)

V1

V2






System Model

∑

Excitation

Cavity field probe

Hfb

Feedback

Klystron and cavity

Hcav(ω)

V1

V2






System Model

∑

Excitation

Cavity field probe

Hfb

Feedback

Klystron and cavity

Hcav(ω)

V1

V2






Wideband Open Loop Transfer Function

−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−60

−50

−40

−30

−20

−10

0

Frequency offset (kHz)

Ga

in (

dB

)

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

200 kHz span;Points near the RF frequencyshow significant scatter;For fitting, ignore points in−0.5–12.25 kHz range aroundRF;Increasing errors at largeoffsets;Near the resonance fit seemsreasonable.




−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.064, Q = 254651, (wr − w

rf) = −1.31 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 686.862 ns, φ = 194.2 deg

Data

Fit





−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.064, Q = 254651, (wr − w

rf) = −1.31 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 686.862 ns, φ = 194.2 deg

Data

Fit





−20 −15 −10 −5 0 5 10 15 20−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.064, Q = 254651, (wr − w

rf) = −1.31 kHz

Data

Fit

−20 −15 −10 −5 0 5 10 15 20−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 686.862 ns, φ = 194.2 deg

Data

Fit




Open Loop Transfer Function, 10 kHz Span

−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)

Gain = 0.065, Q = 251414, (wr − w

rf) = −1.21 kHz

Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

τ = 656.612 ns, φ = 189.6 deg

Data

Fit

A good fit (not wideband enoughto reliably estimate delay);QL is 251414, expected 210000;Fitted detuning and QL give theloading angletan−1(2ωd QL

ωr) = −51◦

Suspect at nominal settingsmight be running with −21◦

loading angle, not −10◦.



Open Loop Transfer Function, 10 kHz Span

−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)

Gain = 0.065, Q = 251414, (wr − w

rf) = −1.21 kHz

Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

τ = 656.612 ns, φ = 189.6 deg

Data

Fit

A good fit (not wideband enoughto reliably estimate delay);QL is 251414, expected 210000;Fitted detuning and QL give theloading angletan−1(2ωd QL

ωr) = −51◦

Suspect at nominal settingsmight be running with −21◦

loading angle, not −10◦.



Correcting Systematics

−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.064, Q = 254651, (wr − w

rf) = −1.31 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 686.862 ns, φ = 194.2 deg

Data

Fit

Magnitude error is most likelydue to RF fundamentalfeedthrough: cavity responserolls off as 1/∆f 2 while NWA IFfilter rolls off as 1/∆f , so errorincreases with offset;180◦ phase shift across theresonance explains why RFfundamental subtracts belowthe resonance and adds above;Use linear (in dB) correctionfunction;Much closer fit.




−100 −50 0 50 100−6

−4

−2

0

2

4

6


Ma

gn

itu

de

err

or

(dB

)





−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.063, Q = 245832, (wr − w

rf) = −1.30 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 948.124 ns, φ = 193.3 deg

Data

Fit




Fitting and Sensitivity to Q

−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.059, Q = 210000, (wr − w

rf) = −1.30 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 1038.58 ns, φ = 194.5 deg

Data

Fit

Forcing QL = 210000 worsensthe fit to compensated data;Even if we use QL = 210000during initial fitting (used toextract linear compensation),final fit is worse;Full 5 parameter fit still comesback to higher QL, within1.6 × 10−5.




−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.066, Q = 210000, (wr − w

rf) = −1.30 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 1038.54 ns, φ = 194.5 deg

Data

Fit





−100 −80 −60 −40 −20 0 20 40 60 80 100−70

−65

−60

−55

−50

−45

−40

−35

−30

−25

−20


Ga

in (

dB

)

Gain = 0.071, Q = 245836, (wr − w

rf) = −1.30 kHz

Data

Fit

−100 −80 −60 −40 −20 0 20 40 60 80 100−200

−150

−100

−50

0

50

100

150

200


Ph

ase

(d

eg

ree

s)

τ = 948.056 ns, φ = 193.3 deg

Data

Fit




Closed Loop Transfer Functions

−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)

Gain = 0.065, Q = 251414, (wr − w

rf) = −1.21 kHz

Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

τ = 656.612 ns, φ = 189.6 deg

Data

Fit

Open-loop transfer function;Closed-loop transfer functionsmeasured at loop gain settings:

4 V;5 V;8 V;10 V.

Some saturation at higher controlvoltages;Nominal direct loop gain is 0.5(50% increase in Robinson beamloading limit).




−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)

Gain = 0.49, Phase = 12.0 deg

Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

Data

Fit


4 V;5 V;8 V;10 V.





−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)


Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

Data

Fit


4 V;5 V;8 V;10 V.





−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)


Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

Data

Fit


4 V;5 V;8 V;10 V.





−6 −4 −2 0 2 4 6−42

−40

−38

−36

−34

−32

−30

−28

−26

−24

−22


Gain

(dB

)


Data

Fit

−6 −4 −2 0 2 4 6−200

−150

−100

−50

0

50

100

150

200


Phase (

degre

es)

Data

Fit


4 V;5 V;8 V;10 V.





4 5 6 7 8 9 100.4

0.5

0.6

0.7

0.8

0.9

1

Control voltage (V)

Lo

op

ga

in

Loop gain vs. control voltage

Nominal operating point


4 V;5 V;8 V;10 V.



Beam Loss Events: Diagnostics and Analysis

Experimental Setup

Attempted to diagnose beam loss events where RF phase activityhas been observed;Set up iGp12 (demo unit) and iGp8 to generate abort triggers andcapture longitudinal bunch-by-bunch data during the abort:

iGp8 connected to a front-end channel tuned for amplitudedetection of the BPM sum signal;Bunch-by-bunch feedback filters are configured to differentiatebunch currents with 105 turn delay;iGp12 runs longitudinal feedback with a different front-end channel,configured for phase detection;External trigger for iGp12 is generated by iGp8 DAC (105 turndifferentiator), trigger threshold adjusted to detect small drop from asingle bucket.Pre-trigger acquisition feature of iGp12 is used to capture themotion both before and after the trigger.

Set up automatic abort data readout, ran overnight.



Experimental Setup






Experimental Setup






Experimental Setup






Experimental Setup






Experimental Setup






Experimental Setup






Beam Loss Event

−80 −70 −60 −50 −40 −30 −20 −10 0 10−2000

−1500

−1000

−500

0

500

1000

1500

Time (ms)

AD

C c

ounts

Converted from abort_20161210_075607.txt

−1 −0.8 −0.6 −0.4 −0.2 0 0.2−2000

−1500

−1000

−500

0

500

1000

1500

Time (ms)

AD

C c

ounts

Bunch 1

What looks like oscillation isactually phase wraparound inthe 1.5 GHz phase detector;Second negative peak is muchsmaller due to current loss —we are measuring ib × sinφb;Full 360◦ oscillation provides allthe necessary information toextract the phase signal;All bunches move together.



Beam Loss Event

−80 −70 −60 −50 −40 −30 −20 −10 0 10−100

−80

−60

−40

−20

0

20

Time (ms)

Phase (

deg@

RF

)

Converted from abort_20161210_075607.txt

−1 −0.8 −0.6 −0.4 −0.2 0 0.2−100

−80

−60

−40

−20

0

Time (ms)

Phase (

deg@

RF

)

Bunch 1

What looks like oscillation isactually phase wraparound inthe 1.5 GHz phase detector;Second negative peak is muchsmaller due to current loss —we are measuring ib × sinφb;Full 360◦ oscillation provides allthe necessary information toextract the phase signal;All bunches move together.



Beam Loss Event

−1 −0.8 −0.6 −0.4 −0.2 0−90

−80

−70

−60

−50

−40

−30

−20

−10

0

Time (µs)

Ph

ase

(d

eg

@R

F)

075607: Phases of all 119 filled bunches What looks like oscillation isactually phase wraparound inthe 1.5 GHz phase detector;Second negative peak is muchsmaller due to current loss —we are measuring ib × sinφb;Full 360◦ oscillation provides allthe necessary information toextract the phase signal;All bunches move together.



Beam Loss Event (Continued)

−80 −70 −60 −50 −40 −30 −20 −10 0−10

−8

−6

−4

−2

0

2

4

6

8

10

Time (ms)

Ph

ase

(d

eg

@R

F)

075607: Phase of bunch 1 Excitations every 20 ms;Fairly large steady-stateexcursions (5◦ peak to peak,0.6◦ RMS);Excitations seem to get biggerjust before the abort, could be acoincidence;Step excitation (HVPS SCRs?);Synchrotron oscillation after astep.




−80 −70 −60 −50 −40 −30 −20 −10 0−10

−8

−6

−4

−2

0

2

4

6

8

10

Time (ms)

Ph

ase

(d

eg

@R

F)





−8 −7 −6 −5 −4 −3 −2−10

−8

−6

−4

−2

0

2

4

6

8

10

Time (ms)

Ph

ase

(d

eg

@R

F)





−7.4 −7.2 −7 −6.8 −6.6 −6.4 −6.2 −6−10

−8

−6

−4

−2

0

2

4

6

8

10

Time (ms)

Ph

ase

(d

eg

@R

F)




Beam Loss: Analysis

−1 −0.8 −0.6 −0.4 −0.2 0−90

−80

−70

−60

−50

−40

−30

−20

−10

0

Time (µs)

Phase (

deg@

RF

)

075607: Phases of all 119 filled bunches

Exponential beam phase runaway isa typical signature of high beamloading Robinson limit;Using cavity parameters estimatedearlier, at zero loading angle andwithout direct feedback the limit is900 mA (1350 mA with directfeedback);For negative loading angles the limitincreases rapidly, for positive — dropsrapidly;



Beam Loss: Analysis

−2 0 2 4 6 8 100

500

1000

1500

2000

2500

3000

Loading angle (degrees)

Beam

curr

ent (m

A)

Robinson limit (no direct feedback)

Robinson limit (direct loop gain 0.5)

Exponential beam phase runaway isa typical signature of high beamloading Robinson limit;Using cavity parameters estimatedearlier, at zero loading angle andwithout direct feedback the limit is900 mA (1350 mA with directfeedback);For negative loading angles the limitincreases rapidly, for positive — dropsrapidly;



Beam Loss: Analysis (Continued)

−2 0 2 4 6 8 100

500

1000

1500

2000

2500

3000


Beam

curr

ent (m

A)



With loading angle of −10◦ (or even−21◦) there should be no beamloading limit;Is it possible the loading angle iswandering during operation?Small positive angle (3-4 degrees) areconsistent with loss events observed;Increasing direct loop gain to 0.96(10 V) should provide a 30% highermargin, a good test of the hypothesis;RF parameters of BEPC2 allow directloop operation at gains of 10–30.




−2 0 2 4 6 8 100

500

1000

1500

2000

2500

3000


Beam

curr

ent (m

A)







−2 0 2 4 6 8 100

500

1000

1500

2000

2500

3000


Beam

curr

ent (m

A)







−2 0 2 4 6 8 100

500

1000

1500

2000

2500

3000


Beam

curr

ent (m

A)





Synchronous Phase Transients

Introduction

The goal was to investigate synchronous phase transients andtheir compensation with fill pattern modulation;Turned out to be a very difficult task: due to the machine dynamicspeak to peak transients are small, in the 0.5–1.5◦ range;To measure absolute phase to 0.1◦ all reflections, coupling, HOMshave to be below −71 dB;After trying many different measurement approaches as well asdifferent pickups we settled on using iGp12 as a sampling scope;BPM sum signal was directly connected to the ADC input afterappropriate attenuation;Could probably get similar or better performance from a widebandoscilloscope.



Introduction




Introduction




Introduction




Introduction




Introduction




Input Signal and Attenuation

0 500 1000 1500 2000−1200

−1000

−800

−600

−400

−200

0

200

400

600

800

Time (ps)

AD

C c

ou

nts

150by2, 574 mA, 20 dB attenuation

Uniform fill of 150 bunches, 4 nsspacing;Time sweep generated byadjusting digital delay line with10 ps resolution;Jumps in the sweep correspondto binary transitions — delaystages are not perfect 10-20-40-80-160-320-640-1280 ps;Modulated pattern doublesbunch current for 24 bunches inthe beginning and 24 bunchesin the end of the train;AM-to-PM conversion.




0 500 1000 1500 2000−1200

−1000

−800

−600

−400

−200

0

200

400

600

800

Time (ps)

AD

C c

ou

nts

150by2, 574 mA, 20 dB attenuation





0 500 1000 1500 2000−2500

−2000

−1500

−1000

−500

0

500

1000

1500

2000

Time (ps)

AD

C c

ou

nts

150by2 modulated, 682 mA, 20 dB attenuation





0 500 1000 1500 2000−2500

−2000

−1500

−1000

−500

0

500

1000

1500

2000

Time (ps)

AD

C c

ou

nts

150by2 modulated, 682 mA, 20 dB attenuation




Input Signal and Attenuation (Continued)

0 500 1000 1500 2000−2500

−2000

−1500

−1000

−500

0

500

1000

1500

2000

Time (ps)

AD

C c

ou

nts

150by2 modulated, 682 mA, 20 dB attenuation Phase shift is due to the iGp12input being overdriven;BPM signal has much widerbandwidth than the iGp12 ADC;To get nearly full-scale ADCswing, input amplifier isoverdrive by a factor of 2!Phase shift with amplitudedisappears with additional 8 dBof attenuation.




0 500 1000 1500 2000−2500

−2000

−1500

−1000

−500

0

500

1000

1500

2000

Time (ps)

AD

C c

ou

nts





0 500 1000 1500 2000−2500

−2000

−1500

−1000

−500

0

500

1000

1500

2000

Time (ps)

AD

C c

ou

nts





0 500 1000 1500 2000−800

−600

−400

−200

0

200

400

600

Time (ps)

AD

C c

ou

nts




Delay Line Calibration

0 500 1000 1500 2000−500

−400

−300

−200

−100

0

100

200

300

400

500

Time (ps)

AD

C c

ou

nts

Calibration based on ameasurement of the RFreference signal;Optimize delay weights to fit apure sinewave;

Bit Nominal Fit0 10 ps 10.5 ps1 20 ps 31.2 ps2 40 ps 46.6 ps3 80 ps 89.3 ps4 160 ps 139.3 ps5 320 ps 341.9 ps6 640 ps 646.7 ps7 1280 ps 1212.5 ps




0 500 1000 1500 2000−500

−400

−300

−200

−100

0

100

200

300

400

500

Time (ps)

AD

C c

ou

nts

Data

Sinewave






0 500 1000 1500 2000−500

−400

−300

−200

−100

0

100

200

300

400

500

Time (ps)

AD

C c

ou

nts

Data

Sinewave





Bunch Phase and Amplitude Estimation

200 400 600 800 1000 1200 1400 1600 1800 2000−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

Time (ps)

Arb

. units

Start from individual bunchsignals normalized by theirpeak-to-peak amplitude;Calculate average shape signal;Fit a 21st order polynomial to theaverage;For each bunch perform a twoparameter fit: time shift andamplitude scaling;Result: bunch-by-bunchcurrents and phases.




200 400 600 800 1000 1200 1400 1600 1800 2000−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

Time (ps)

Arb

. units





200 400 600 800 1000 1200 1400 1600 1800 2000−0.8

−0.6

−0.4

−0.2

0

0.2

0.4

0.6

Time (ps)

Arb

. units

Data

Polynomial fit





0 50 100 150 2000

2

4

6

8

10

Bunch number

Cu

rre

nt

(mA

)

0 50 100 150 200−2

−1.5

−1

−0.5

0

0.5

1

Bunch number

Ph

ase

(d

eg

@R

F)

2.49° peak−to−peak





0 50 100 150 2000

2

4

6

8

10

Bunch number

Cu

rre

nt

(mA

)

0 50 100 150 200−2

−1.5

−1

−0.5

0

0.5

1

Bunch number

Ph

ase

(d

eg

@R

F)

2.49° peak−to−peak




Uniform Train: Measurement and Simulation

0 50 100 150 200 250 300 350 400−2

−1.5

−1

−0.5

0

0.5

1

1.5

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 200 250 300 350 4000

1

2

3

4

5

6

7

8

RF bucket number

Cu

rre

nt

(mA

)

BEPC2 e− at 696 mA and 1.08 MV

Measurement

Simulation

Measurement

Simulation

To maximize the transient filledhalf the ring (99 bunches in 4 nsspacing);RF voltage reduced to 1.08 MV;Calculated transient usingPedersen’s small-signal model;Feature around bucket 60 is dueto an HOM roughly 18 mdownstream.




0 50 100 150 200 250 300 350 400−2

−1.5

−1

−0.5

0

0.5

1

1.5

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 200 250 300 350 4000

1

2

3

4

5

6

7

8

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation





0 50 100 150 200−2

−1.5

−1

−0.5

0

0.5

1

1.5

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 2000

1

2

3

4

5

6

7

8

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation





0 50 100 150 200−2

−1.5

−1

−0.5

0

0.5

1

1.5

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 2000

1

2

3

4

5

6

7

8

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation




Modulated Train: Measurement and Simulation

0 50 100 150 200 250 300 350 400−1.5

−1

−0.5

0

0.5

1

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 200 250 300 350 4000

2

4

6

8

10

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation

Modulated fill: 22 bunches atthe beginning and the end ofthe train at twice the current;Expect partial transientcompensation for 55 bunches inthe middle;Reasonable agreementbetween measurements andsimulation.




0 50 100 150 200 250 300 350 400−1.5

−1

−0.5

0

0.5

1

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 200 250 300 350 4000

2

4

6

8

10

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation





0 50 100 150 200−1.5

−1

−0.5

0

0.5

1

RF bucket number

Ph

ase

(d

eg

@R

F)

0 50 100 150 2000

2

4

6

8

10

RF bucket number

Cu

rre

nt

(mA

)


Measurement

Simulation

Measurement

Simulation




Experimental Options for Studying Gap Transients

0 50 100 150 200 250 300 3500

0.5

1

1.5

2dec2216/181706: Bunch Current Monitor, Io=388.9299mA, nCM=1

mA

0 50 100 150 200 250 300 350−100

−50

0

50

100Averages of bunch signals

AD

C c

ou

nts

0 50 100 150 200 250 300 350−10

−5

0

5

10Synchronous phase (relative to reference oscillator)

bunch number

de

g@

RF

Nominal fill pattern at the ALS,reduced beam current (388 mAinstead of 500 mA);Harmonic cavities tuned in;15.8 degrees peak-to-peak;Should detune harmoniccavities to simplify the analysis;Can try both current and density(spacing) modulations.




0 50 100 150 200 250 300 3500

0.5

1

1.5


mA

0 50 100 150 200 250 300 350−100

−50

0

50


AD

C c

ou

nts

0 50 100 150 200 250 300 350−10

−5

0

5


bunch number

de

g@

RF





0 50 100 150 200 250 300 3500

0.5

1

1.5


mA

0 50 100 150 200 250 300 350−100

−50

0

50


AD

C c

ou

nts

0 50 100 150 200 250 300 350−10

−5

0

5


bunch number

de

g@

RF





0 50 100 150 200 250 300 3500

0.5

1

1.5


mA

0 50 100 150 200 250 300 350−100

−50

0

50


AD

C c

ou

nts

0 50 100 150 200 250 300 350−10

−5

0

5


bunch number

de

g@

RF



Summary

Summary

Measurements of RF system transfer functions suggest low directloop gains and unexpected loaded Q;More careful measurements are needed to better quantify RFtransfer functions and tuning angles;Phase transients in BEPC2 are small and difficult to measure;Charge/density modulation seems to work as expected.


Summary

Summary



Summary

Summary



Summary

Summary



Beam Loading Studies in BEPC2 - DimtelBeam Loading Studies in BEPC2 Junhui Yue1, Jianping Dai1, Yuan Zhang1, Haipeng Wang2, D. Teytelman3 1IHEP, Beijing, China 2JLAB, Newport News,

Documents