LBNF / RR Injection Kicker Angled Adapter Measurement Update Dennis - - PowerPoint PPT Presentation

▶
lbnf rr injection kicker angled adapter measurement update
SMART_READER_LITE
LIVE PREVIEW

LBNF / RR Injection Kicker Angled Adapter Measurement Update Dennis - - PowerPoint PPT Presentation

Oct 31, 2022 21 likes •275 views

LBNF / RR Injection Kicker Angled Adapter Measurement Update Dennis Barak, Chris Jensen, George Krafczyk MI/RR Department Meeting 16 August 2017 Outline Purpose/Background Test setup overview Recap Oct2016 results and what

slide-1
SLIDE 1

Dennis Barak, Chris Jensen, George Krafczyk MI/RR Department Meeting 16 August 2017

LBNF / RR Injection Kicker Angled Adapter Measurement Update

slide-2
SLIDE 2
  • Purpose/Background
  • Test setup overview
  • Recap Oct2016 results and what

remained to be measured

  • Interesting findings
  • Measurement results

Outline

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

2

slide-3
SLIDE 3
  • Proposed LBNF beamline rolled dipole

interferes with existing RR Injection kicker magnet input cables and load resistors

  • Fabricated two prototype LBNF angled

adapters (LBNFAA) for testing

  • Testing is needed to understand how

the LBNFAA affects the kicker magnet’s response

  • Initial testing completed in Oct 2016
  • Each RR Inj Kicker feeds two RKAA

magnets in parallel

  • R:KPS1B feeds

– KM103D (RKAA002) – KM103E (RKAA007)

Background

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

3

HV input connector

slide-4
SLIDE 4
  • Component capacitances:

– Regular Fluorinert feed-thru (with disc): 27.4 pF (+19.8 pF) – Modified Fluorinert feed-thru (no disc): 7.6 pF (+0 pF) – LBNE load adapter: 40.0 pF (+32.4 pF) – Whole magnet: 715-735 pF (RKA/RKAA)

Background (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

4

Fluorinert feed- thru Magnet load HV input connector

slide-5
SLIDE 5

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

5

Oct 2016 Results – Test Setup

slide-6
SLIDE 6

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

6

0.2 0.4 0.6 0.8 1 1.2 0.5 1 1.5 2

Time (us)

KM103E - Normalized Values

Oct2016 - VinE Oct2016 - VoutE Oct2016 - KM103E, L315

Oct 2016 Results (cont’d)

Int(B)

slide-7
SLIDE 7

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

7

0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1.055 1.06 1.065 1.07 1.075 1.08 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76

Time (us)

Normalized Int(B) - Rise Time

Oct2016 - KM103D, L316, LBNFAA Oct2016 - KM103E, L315

Injection Kicker Spec:

  • Rise Time: 57 ns max (3 buckets)
  • Flattop ripple: +/- 3%

57 ns

Oct 2016 Results (cont’d)

slide-8
SLIDE 8
  • Initial testing showed:

– Large increase (+4.5%) in peak kicker Int(B) overshoot – Kicker Int(B) rise time increased to >90 ns (spec is 57 ns) – Adapter removed from kicker magnet in tunnel at request of MI/RR department

  • Proposed additional testing:

– Quantify measurement certainty – LBNF angled adapter with a modified (disc-less) Fluorinert feedthrough – LBNF angled adapter on magnet input connection – Perform these tests on spare kicker magnet in A0 kicker test cage

Oct 2016 Results (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

8

HV input connector

slide-9
SLIDE 9
  • Testing with spare RKA magnet in A0 test

kicker system not successful

– Unable to replicate Oct 2016 results – RKA vs. RKAA

  • RKAA is revised RKA
  • Testing resumed in tunnel at beginning of

Shutdown 2017

  • Noteworthy findings:

– Failed kicker load on KM103D

  • Measurement-to-measurement Int(B)
  • vershoot variation
  • May have contributed to large overshoot seen

in Oct 2016 results

  • Replaced with spare load
  • Also replaced kicker load on KM103E to

match new load on KM103D

– KM103D is an early RKAA magnet

  • Pulse response similar to RKA not RKAA
  • Contributes to large overshoot seen in Oct

2016 results

Since Last Update

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

9

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1.055 1.06 1.065 1.07 1.075 1.08 1.085 1.09 1.095 1.1 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized B Field Flattop

Oct2016 - KM103D with adapter 2017M1 - KM103D no adapter 2017M2 - KM103D no adapter 2017M4 - KM103D no adapter 2017M5 - KM103D no adapter

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1.055 1.06 1.065 1.07 1.075 1.08 1.085 1.09 1.095 1.1 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized B Field Flattop

Oct2016 - KM103D, L316, LBNFAA 2017M1 - KM103D, L316 2017M2 - KM103D, L316 2017M4 - KM103D, L316 2017M5 - KM103D, L316

0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1.055 1.06 1.065 1.07 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized B Field Flattop

RKA007, L318 2017M1 - KM103D, L316 2017M6 - KM103D, L318 2017M7 - KM103D, L318 2017M8 - KM103E, L309

0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 1.045 1.05 1.055 1.06 1.065 1.07 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized B Field Flattop

RKA007, L318 2017M1 - KM103D, L316 2017M6 - KM103D, L318 2017M7 - KM103D, L318 2017M8 - KM103E, L309

slide-10
SLIDE 10

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

10

Magnet Pulse Response Test Setup

/ 318 / 309

slide-11
SLIDE 11

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

11

Magnet Pulse Response Test Setup (cont’d)

slide-12
SLIDE 12

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

12

0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized Int(B) - Flattop

2017M1 - KM103E, L315 2017M2 - KM103E, L315, LBNFAA 2017M4 - KM103E, L315, LBNFAA, mod FLFT

Results – Effect of LBNFAA / LBNFAA + mod FLFT

1.37% peak 2.54% peak 1.90% peak

Int(B) – Normalized at “Mean” Flattop Value

slide-13
SLIDE 13

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

13

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized Int(B) - Flattop

2017M1 - KM103E, L315 2017M2 - KM103E, L315, LBNFAA 2017M4 - KM103E, L315, LBNFAA, mod FLFT

Results – Effect of LBNFAA / LBNFAA + mod FLFT

+/- 0.97% +/- 1.5% +/- 1.25%

Injection Kicker Spec:

  • Rise Time: 57 ns max (3 buckets)
  • Flattop ripple: +/- 3%

Int(B) – Normalized at Median Flattop Value

slide-14
SLIDE 14

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

14

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76

Time (us)

Normalized Int(B) - Rise Time

2017M1 - KM103E, L315 2017M2 - KM103E, L315, LBNFAA 2017M4 - KM103E, L315, LBNFAA, mod FLFT

Injection Kicker Spec:

  • Rise Time: 57 ns max (3 buckets)
  • Flattop ripple: +/- 3%

57 ns

Results – Effect of LBNFAA / LBNFAA + mod FLFT

+/- 0.97% +/- 1.5% +/- 1.25%

Int(B) – Normalized at Median Flattop Value

slide-15
SLIDE 15

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

15

0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized Int(B) - Flattop

2017M7 - KM103E, L309, LBNFAA, mod FLFT 2017M8 - KM103E, L309

Results – Effect of LBNFAA + modified FLFT after changing kicker load

1.67% peak 1.18% peak

Int(B) – Normalized at “Mean” Flattop Value

slide-16
SLIDE 16

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

16

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized Int(B) - Flattop

2017M7 - KM103E, L309, LBNFAA, mod FLFT 2017M8 - KM103E, L309

Results – Effect of LBNFAA + modified FLFT after changing kicker load

+/- 1.1% +/- 0.85%

Injection Kicker Spec:

  • Rise Time: 57 ns max (3 buckets)
  • Flattop ripple: +/- 3%

Int(B) – Normalized at Median Flattop Value

slide-17
SLIDE 17

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

17

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76

Time (us)

Normalized Int(B) - Rise Time

2017M7 - KM103E, L309, LBNFAA, mod FLFT 2017M8 - KM103E, L309

Injection Kicker Spec:

  • Rise Time: 57 ns max (3 buckets)
  • Flattop ripple: +/- 3%

57 ns

Results – Effect of LBNFAA + modified FLFT after changing kicker load

+/- 1.1% +/- 0.85%

Int(B) – Normalized at Median Flattop Value

slide-18
SLIDE 18

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

18

0.96 0.965 0.97 0.975 0.98 0.985 0.99 0.995 1 1.005 1.01 1.015 1.02 1.025 1.03 1.035 1.04 0.095 0.19 0.285 0.38 0.475 0.57 0.665 0.76 0.855 0.95 1.045 1.14 1.235 1.33 1.425 1.52 1.615

Time (us)

Normalized Int(B) - Flattop

2017M8A - KM103E, L309 2017M9 - KM103E, L309, LBNFAA (input)

Results – Effect of adding LBNFAA on magnet input

Int(B) – Normalized at Median Flattop Value

slide-19
SLIDE 19
  • KM103D pulse response differs from the other five RR Inj Kicker magnets

– 𝚬Int(B)KM103D,flattop = 4.4% (or +/- 2.2%) – 𝚬Int(B)RKAAxxx,flattop is typically 2% (or +/- 1%)

  • LBNFAA / LBNFAA + mod FLFT on load side
  • LBNFAA on input side

– Virtually no effect on input voltage rise time (13.2 ns vs 13.0 ns)

  • Within typical measurement-to-measurement variation

– Int(B)peak changed less than 0.1% on KM103E

  • Within typical measurement-to-measurement variation

Final Results

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

19

Normal LBNFAA LBNFAA + mod FLFT 𝚬Int(B)flattop,KM103E,L315 +/- 0.97% +/- 1.54% +/- 1.25% 𝚬Int(B)flattop,KM103E,L309 +/- 0.85%

  • +/- 1.08%

% Increase

  • + 0.57%

+ <0.3%

slide-20
SLIDE 20

Thank you

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

20

slide-21
SLIDE 21
  • Voltage probes:

– 200:1 HV probe -> input voltage (Vin) – Load current monitor -> output voltage (Vout) – All signals 50 Ohm terminated, scope inputs at high impedance

  • Measurement configurations:
  • Average the two measurements
  • Modify gain of output voltage

– Vin-Vout = 0 towards the end of the pulse – Account for propagation delay in calculating Vin-Vout

  • Calculate integrated B field

– Total flux ϕ = Li = B∗A∗N – v = L(di/dt) → Li = ∫vdt

(inductor)

– ∫vdt = B∗A = B∗w∗l – B∗l = (1/w) ∫vdt →

Int(B) ∝ ∫vdt – Assumptions

  • If skin depth is small during

pulse compared to w, then w is constant

  • If all flux is perpendicular to the

beam, then B*l is same as kick field

  • Normalize all quantities for plotting

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

21

Backup Slides

slide-22
SLIDE 22

Backup Slides (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

22 KM103D KM103E Measurement m Load/Changes Adapter/mod.FLFT Load/Changes Adapter/mod.FLFT Oct 2016 1,2 Load 316 Load side Load 315 2017M1 3,4 Load 316 Load 315 2017M2 5,6 Load 316 Load 315 Load side 2017M4 9,10 Load 316 Load 315 Load side + mod FLFT 2017M5 11,12 Load 316 Load 315 Load side + mod FLFT 2017M5A 13,14 Load 316 Load 315 Load side + mod FLFT 2017M6 15,16 Load 318 Load 315 Load side + mod FLFT 2017M6A 17,18 Load 318 Load 315 Load side + mod FLFT 2017M7 19,20 Load 318 Load 309 Load side + mod FLFT 2017M8 21,22 Load 318 Load 309 2017M8A 23,24 Load 318 Load 309 2017M9 25,26 Load 318 Input side Load 309 Input side

  • Additional measurements:
  • RKA007 in kicker test cage
  • HV probe connected to load box input
slide-23
SLIDE 23

Estimate measurement and calculation errors:

  • Multiple measurements with identical configurations (1 day to 1 week apart)
  • Probe comparison (HV probe vs. Load current monitor voltage divider)
  • Found bad load by looking at measurement-to-measurement repeatability
  • RKA vs RKAA

– KM103D is RKAA002 – KM103E is RKAA007

  • M8 vs M8A

– Peak Int(B) overshoot increased 0.5% from M8 to M8A

  • Int(B) tail dips below zero about 6%
  • Propagation delay changes when adding LBNFAA and modified FLFT

Backup Slides (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

23

slide-24
SLIDE 24

Conclusions:

  • Found bad load by looking at measurement-to-measurement

repeatability

– L316 caused +/- 1% variation in peak Int(B) overshoot – With L318, variation is less than +/- 0.4%

  • RKA vs RKAA

– KM103D is RKAA002 – KM103E is RKAA007 – RKAA has higher Z0 than RKA – RKAA002 was RKA010 with some modifications

  • Peak Int(B) overshoot increased 0.5% from M8 to M8A

– M8A completed one week after M8 – Power outage on day before M8A

  • Vin rise time decreased to 12.8-13.0 ns rather than 13.2-13.4 ns
  • Difference in magnet case temperature / magnet capacitance?

Backup Slides (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

24

slide-25
SLIDE 25

Conclusions (cont’d):

  • Int(B) tail dips below zero about 6%

– Occurs in load box measurement as well (no magnet) – Cause not understood, but it only affects the B field waveshape on/after falling edge – Int(B) rise, overshoot, and flattop are believable

  • Propagation delay changes when adding LBNFAA and modified FLFT on

magnet load-side

– Output voltage pulse delayed by 1.4 ns with LBNFAA only – Output voltage pulse delayed by 0.8 ns with LBNFAA + mod FLFT – Due to added stray capacitance – Very difficult to measure output voltage directly and not through the load

Backup Slides (cont’d)

8/16/17

  • D. Barak | LBNF/RR Injection Kicker Load Adapter

25