MYRRHA Superconducting linac & Fault Tolerance Concept

(1)

HAL Id: in2p3-03062748

http://hal.in2p3.fr/in2p3-03062748

Submitted on 18 Dec 2020

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of

sci-entific research documents, whether they are

pub-lished or not. The documents may come from

teaching and research institutions in France or

abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est

destinée au dépôt et à la diffusion de documents

scientifiques de niveau recherche, publiés ou non,

émanant des établissements d’enseignement et de

recherche français ou étrangers, des laboratoires

publics ou privés.

MYRRHA Superconducting linac & Fault Tolerance

Concept

Frédéric Bouly

To cite this version:

(2)

MYRRHA Superconducting linac &

Fault Tolerance Concept

F. Bouly (LPSC- CNRS)

30th International Linear Accelerator

Conference 2020

(3)

MYRRHA Project

2

1 Sept. 2020 Fault tolerance & MYRRHA Linac - F. Bouly - LINAC2020 virtual conference

MYRRHA

M

ulti-purpose h

Y

brid

R

esearch

R

eactor for

H

igh-tech

A

pplications at Mol (Belgium)



Objectives:



Demonstrate the ADS concept at pre-industrial scale



Study nuclear waste transmutation/incineration



Multipurpose and flexible irradiation facility



Staged implementation:



Phase 1: MINERVA to 100 MeV, 4 mA CW



Phase 2: 600MeV linac



Phase 3: connecting to nuclear reactor - ADS



Funding for Phase 1 approved & implementation started at

SCK●CEN (Belgium, 2018

)

(4)

MYRRHA Project

3 MYRRHA

MINERVA

Proton energy 600 MeV 100 MeV

Peak beam current 0.1 to 4.0 mA

Repetition Rate 1 to 250 Hz

Beam duty cycle

2.10-4_{to 1} 2.10-4to 0.125 (on PTF)

2.10-4_{to 0.875 (on FPBD)}

Beam power stability < ± 2% on a time scale of 100 ms

Beam current stability < ± 2% over macropulse duration (To be confirmed ) Beam footprint on reactor window / or PTF Circular ∅ 85 mm _{To be specified} Beam footprint stability < ± 10% on a time scale of 1 sec.

# of allowed beam trips on reactor longer than 3 sec

10 maximum per 3-month operation period

# of allowed beam trips on reactor longer than 0.1 sec

100 maximum per day

# of allowed beam trips on reactor shorter than 0.1 sec

unlimited

High power CW proton beam (up to 2.4 MW)

MYRRHA and MINERVA main proton beam specifications

Panorama of high power proton/deuteron beams worldwide (non exhaustive plot).

J-L. Biarrotte, Comparison of SRF projects, SLHiPP-4, CERN, 2014

With Some updates, 2020 (F. Bouly)

Non-exhaustive list

ADS

Extreme reliability

Data from :

D. Vandeplassche et al., ”Accelerator Driven Systems”, Proc. IPAC 2012,

New Orleans Louisiana, USA, 2012

&

E. Bargallo, “ESS reliability and availability approach”, IPAC2015,

(5)

Reliability & linac layout

4 

Reliability guidelines for the ADS accelerator design:

 Robust design

i.e. robust optics, simplicity, low thermal stress, operation margins…

 Reparability

(on-line where possible) and efficient maintenance schemes

 Redundancy

(serial where possible, or parallel) to be able

to tolerate/mitigate failures

(6)

Local fault-compensation scheme in SC Linac

5  A failure is detected anywhere

→ Beam is stopped by the MPS in injector at t

₀

 The fault is localised in a SC cavity RF loop

→ Need for an efficient fault diagnostic system

 New V/φ set-points are updated in cavities

(cryomodule) adjacent to the failed one

→ Set-points determined in advance: via ‘virtual accelerator

application’ during the commissioning/tuning phase

 The failed cavity is quickly detuned (to avoid the beam loading effect)

→ Using the Cold Tuning System

 Once steady state is reached, beam is resumed at

t

₁

< t

₀

+ 3sec

(7)

Linac design & accelerating gradients

6 

Enable Fault Compensation



Independently-powered cavities with moderate E

_acc

(30% margins) to ensure a fault-tolerant capability



Ensure a longitudinal acceptance as high as possible (

low synchronous phases

)

*E_accis given at β_optnormalised to L_acc= Ngap.β.λ/2 **r/Q is given at β_optwith the “linac” definition

Section # #1 #2 #3

Einput(MeV) 16.6 101.4 172.3

Eoutput(MeV) 101.4 172.3 601.6

Focusing type Normal conducting quadrupole doublets

Cavity technology Single

Spoke Double Spoke Elliptical Cavity frequency (MHz) 352.2 704.4 Cavity optimal β 0.375 0.495 0.705 Nb. of cav. / cryomodule 2 2 4 Total nb of cavity 60 18 72 Nb. of cells / cavity 2 3 5 Bpk/Eacc* (mT/MV/m) 7.3 8.75 4.6 Epk/Eacc* 4.3 4.4 2.5 r/Q** (ohms) 217 427 315 Nominal Eacc(MV/m) * 7.0 6.8 11.0 Max. Eacc(MV/m) * 9.1 9.0 14.3

Synchronous phase (deg) -45 to -15 -35 to -15

4 mA beam load / cav (kW) 1.1 to 8.4 8.2 to 16.4 2.9 to 31.9

QL 1.5 106 2.1 106 6.9 106

Nominal Qpole grad. (T/m) 5.1 to 7.9 3.8 to 4.3 4.4 to 6.0

Section length (m) 91.2 36.3 121.0

Single Spoke

Courtesy of D. Longuevergne (IJCLab/CNRS)

”MYRTE Deliverable : MYRRHA SRF spoke R&D”, 2018

F. Bouly, et al. , “Superconducting LINAC Design Upgrade in View of the 100 MeV

(8)



The retuning feasibility have been assessed



Different scenarios (with multiple failures)



Use of TraceWin code optimisation algorithms

Fault recovery beam dynamics feasibility



Still losses (> 1W/m) can occur when tested on a non

perfect machine (statistic errors analysis)



Fault compensation may induce emittance growth and

longitudinal acceptance decrease

7 Nominal

Multi. Fault-Recovery



Example with multiple failures :

 Section #1: 1 cavity

(9)



Goal : develop an algorithm to quickly pre-calculate any cavity failure compensation by being

less aggressive on the beam.



Example with the compensation of one failed cavity



Based on several criteria

 1

st

_{criterion: recover the same transfer matrix (longitudinal plane) of the retuned area than in nominal condition}



keep smooth variation of phase advance from on period to another.



Simplified model (transfer function) of cavity (assuming ΔW

_i

<< W

_i

)

𝛿𝜙

𝛿𝜙′

_𝑂𝑈𝑇

=

𝐶𝑜𝑠(𝑘. 𝐿

_𝑐

)

1 𝑘

𝑆𝑖𝑛(𝑘. 𝐿

𝑐

)

− 𝑘 𝑆𝑖𝑛(𝑘. 𝐿

_𝑐

)

𝐶𝑜𝑠(𝑘. 𝐿

_𝑐

)

_{𝐶𝑎𝑣𝑖𝑡𝑦}

.

𝛿𝜙

𝛿𝜙′

_𝐼𝑁

.

𝑘 =

𝜔

_𝑅𝐹

𝑚

₀

𝑐

3 _𝛽

3 _𝛾

3 𝑞 𝐸𝑎𝑐𝑐 sin(ϕ

𝑆

)

Method development for retuning

(10)

Method development for retuning

 2

nd

_{criterion: the total Energy gain should remain}

the same than in the nominal case

Δ𝑊

₁

+ Δ𝑊

₂

+ Δ𝑊

₃

+ Δ𝑊

₄

+ Δ𝑊

₅

= Δ𝑊

_𝑇𝑜𝑡

= Δ𝑊

₁

𝑅

+ Δ𝑊

₂

𝑅

+ Δ𝑊

₃

𝑅

+ Δ𝑊

₄

𝑅

 3

rd

_{criterion: the time of flight should remain the same than in the nominal case}

𝑇

_𝑖

=

𝐿

𝑐

𝛽

_𝑖−1

+ 𝛽

_𝑖

. 𝑐

2 

Assumption on the cavity time of flight:

 Problem can be solved by optimization of β

_i

and k

_i



(ΔW

_i

,k

_i

)  (E

_acc

, ф

_s

)  (Amplitude increase, Δϕ

_RF

)



The solution with the simplified model is injected into the TraceWin model for fine (adjustments) tuning of

the solutions

9

(11)

Example : failure compensation of a Spoke

Set points

calculation with

only with

Tracewin code

0.4

0.48 Set points with

dedicated

algorithm

10

(12)



Fast cavity retuning (or detuning) must also be assessed



in less than 3 seconds

Fast failure compensation feasibility

Cold Tuning system

11 

Cryomodule prototype under development at IJCLab Orsay.

Fast Tuner

• Compensation of small frequency shifts with a high speed

• Piezoelectric actuators

Slow Tuner

• Compensation of large frequency shifts with a low speed

• Stepper motor with planetary gearbox (1:256)

Courtesy of H. Saugnac (IJCLab/CNRS)

(13)



_{A model to study the technological feasibility of retuning}

procedures and to set the requirements on the cavity

control loops, the tuning systems.

Cavity and control loops model



Parameters : Spoke cavity

Param.

value

r/Q (@ β

_opt

)

217 Q

_L

1.5 10

6

E

acc nomi

7.0 MV/m

E

_{acc retuning (+30 %)}

9.1 MV/m

P

_max

Ampli RF

16 kW

I

_b

4 mA



Matlab Simulink Model

12

M. Dominiczak, F. Bouly, N. Gandolfo, and C. Joly, “Modeling of Superconducting Spoke Cavity with its Control Loops Systems for the MYRRHA Linac Project”, SRF'19, Dresden, TUP002.

(14)

Failure compensation – Fast retuning

(15)

Failed cavity – Fast detuning



Failed cavity has to be detuned to avoid beam loading



Example : Spoke Cavity

Superconducting 2K

Quenched critical coupling : Q0 = Qi

Quenched ~100 K

Superconducting 2K

Quenched critical coupling : Q0 = Qi

Quenched ~100 K

Threashold < 0.5 % Vacc nominal

14 kHz

14 

Consequences

(16)

Conclusions

15 

Superconducting linac



Design updated to follow the phase approach of the project ( Phase 1 : MINERVA @ 100 MeV)



In nominal condition operates with de-rated cavities to enable fault compensation



Fault Recovery procedures



Dedicated algorithm under development



_{Study to be coupled with reliability model development : to evaluate the number of failure to be expected.}



Model developments carried out at CERN (J. Uythoven et al. )



Cryomodule prototype under construction to test fast retuning scenarios :



Retune a cavity in less than 1 second



Detune the failed cavity by at least 14 kHz in 1 second



1 st

beam in MYRRHA RFQ!



_{July 2020, Louvain-la-Neuve}



4 mA & 2% duty cycle



We also work on other technics (with Machine learning) to improve the linac tuning (injector).



M. Debongnie et al., “Modelization of an Injector With Machine Learning”, IPAC'19, Melbourne, WEPTS006

(17)

16

THANK YOU

* Part of this Work was supported by the European Atomic Energy Community’s (EURATOM) H2020 Programme under grant agreement n°662186:

MYRTE project.

(18)

2009

2010

2011

2012

2013

2014

2015

2016

2017

2018

2019

2020

2021

MYRRHA Accelerator, Background studies

17

31 Aug. - 4

Sept. 2020 Fault tolerance & MYRRHA Linac - F. Bouly - LINAC2020 virtual conference

Eurotrans (FP6)

MYRRHA : XT-ADS demo.

Linac 600 Mev –

R&D on

reliability-GUINEVERE & GENEPI-3C

MAX (FP7)

 Start-to-end reference design w. error study

 Prototyping : elliptical and spoke cavity, RF

ampli., RFQ mock-up, CH-DTL cavities

 Reliability model

 Design Review

CDT (FP7)

– HEBT design

MARISA (FP7)

LEBT construction – 176 MHz RF

amplifier–

PHASE 1 (100 MeV) : MINERVA

Construction & Commissioning of the

first Accelerator section -> 100 MeV

MYRTE (H2020)

 Injector construction & commissioning

 Beam Characterisation & Control

 SRF cavities (spoke Cryomodule – CH )

 Reliability & specific R&D

End 90’s : 1

st

_{accelerator projects for ADS (APT/AAA, TRASCO, …)}

(19)

2 main studied solutions

Parameters

Reference Design

Alternative Design

6

0

0 M

e

V

Co

n

fig

.

M

YR

RHA

Total nb. of cavities

150: 60 single spokes + 18 double

spokes +72 5-cell elliptical

144: 46 single spokes + 26 double

spokes +72 5-cell elliptical

Total nb. of Cryomodules

57: 30 single spokes + 9 double spokes

+ 18 5-cell elliptical

54: 23 single spokes + 13 double spokes +

18 5-cell elliptical

Total Length (m)

248.4 m

243.3 m

Max. Quad gradient (T/m)

7.9

7.6 Emittance growth (%)

ε

Longi

ε

Trans

ε

Longi

ε

Trans

-1.2 %

+2.9 %

+0.4 %

+1.8 %

Longitudinal Acceptance

~73 × ε

RMS

~ 78 × ε

RMS

Mismatch 30 % +++

ε

Longi

ε

Trans

ε

Longi

ε

Trans

+33 %

+10 %

+26 %

+13 %

Mismatch 30 % +-+

ε

Longi

ε

Trans

ε

Longi

ε

Trans

+15.5 %

+24.5 %

+14 %

+26 %

Beam current sensitivity

Emit. Growth @ 6 mA

ε

Longi

ε

Trans

ε

Longi

ε

Trans

+1.7 %

+4.6 %

+ 3.3 %

+ 2.9 %

1

0

0 M

eV

Co

nfi

g.

M

IN

ER

V

A

Total nb. of cavities

60 ( = single spokes)

54 ( = 46 single + 8 double spokes)

Total nb. of Cryomodules

30

27 Total Length (m)

91.2 m

86.0 m

Energy Capabilities

-35° < φ

s

< -10°

Eacc (@ βopt) = 9 MV/m

141.4 MeV

144.7MeV

101 MeV @ Cavity #44

31 Aug. - 4