MKI heating estimates Vasilis Vlachodimitropoulos Acknowledgements MBarnes H Day L Vega Cid MKIs role Take beam from SPS and put in a stable orbit of LHC SPS LHC Injected Beam Circulating Beam ID: 813007
Download The PPT/PDF document "Status and procedures for" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Status and proceduresforMKI heating estimates
Vasilis Vlachodimitropoulos
Acknowledgements:
M.Barnes
, H. Day, L. Vega Cid
Slide2MKI’s roleTake beam from SPS and put in a stable orbit of LHC
SPS
LHC
Injected Beam
Circulating Beam
Courtesy: Lorena
Courtesy: Mike
Slide3MKI in shortMKI is a transmission line kicker magnet – To meet rise time requirements
It consists of 33 blocks – Capacitors + Ferrite Yoke
Slide4Beam ScreenScreen design: 24 conductive
wires on a ceramic tube
To keep rise time low: wires are
capacitively
coupled (cc) to the ground in one end and grounded only in the other
Slide5Beam Impedance issuesPart of the TEM beam field couples to the cylinder-wires cavity
The
wave propagates in the cavity and gets radiated at the ends
The
emitted power reaches the ferrite yoke and heats it up
Power is distributed non-uniformly: Upstream blocks are affected more
Slide6What we wantKeep the yoke below its Tc
Accurate impedance model of the MKI (CST and measurements)
Power loss distribution (
CST+Matlab
)
Slide7Our approach – in math
What we expect
What we estimateSo we
approximate
Slide8Our approach – in wordsCST does not provide a time power loss monitor for dispersive material
Place CST power loss monitors at resonant frequencies of Real Longitudinal Impedance
Integrate power loss density over transverse planes
→
Longitudinal power distribution at each frequencyNormalize each distribution to unit total power loss per frequency; to ignore normalizations from CST
Scale each distribution by a factor proportional to the power loss at the same frequencySum the distributions and normalize the final distribution to unityScale the total distribution with the total expected power loss
Slide9The results
Courtesy:
Hugo
Slide10Model Validation
1.
Magnet_Up
(Ferrite
yoke
#3)
2. Upstream Ferrite RingsPT100 measured49°C
87°CANSYS calculations
46.7°C
93°C
Courtesy: Lorena
2
1
Slide11Run2 & HL predictionsRun 2: all simulated blocks below Tc
HL-LHC: all seven blocks above Tc
Courtesy: Lorena
Slide12Discrepancies with previous predictions
preLS1
postLS1
HL-LHC 25ns
HL-LHC 50ns
CR01_T05
28
/
14
45
/
25.5
167
/
93.5
206
/
116.1
CR02_T10
30
/
15.2
48
/
27.2
177
/
99.6
218
/
124
CR03_T01
25
/
14.5
39
/
23.5
145
/
85.85
179
/
106
CR05_T02
21
/
11.2
36
/
19.6
130
/
71.8
160
/
90
CR07_T08
27
/
14.6
45/24.8163/91199/113.4CR08_T1126/14.643/24.9158/91.2192/112.8CR09_T0326/13.844 /24.8162/90.8198/112.9CR10_T0625/13.341/23.3150/85.5184/106.22CR11_T1335/24.752/38.2191/139.8240/172.66CR12_T1220/12.734/21.7125/79.5151/98.44
Table: Estimated power loss per magnet, with measured impedance data obtained using the resonant method
Slide13Impedance – Total Power Loss
preLS1
postLS1
HL-LHC
25ns
HL-LHC
50ns
coarse
29.7
56.7
207.7
196.3fine1
28.134.5
126.2
185.85
fine2
30.2
37.8
138.5
199.3
fine3
33.4
43.3
158.5
231.3
Table: Power loss estimates from impedance data of different meshes of the MKI model
Slide14Solutions we look intoShorter overlap b/w cylinder and wires
→
Impedance peaks are upshifted where the beam is less
energetic →
Lower total loss Greater exposure of the radiated field to damping materialsAdditional ferritic collar (Hugo)
Optimization of current configuration of existing ferrite ringsWindow openings in the metallic cylinder
Slide15Summary Main goal: keep ferrite yoke below Tc
E
stimate the distribution of power deposition in the MKI
CST+Matlab
→ normalized distributionCST/measurements → total power lossCurrent estimations suggest MKI will be a major issue for HL-LHC
Tool validation – Any feedback is more than welcome!Test proposed solutions
Slide16Thank you for your attention!