scieee Science in your language
[Es] (orig)

Diseño y fabricación de un banco de pruebas hidráulico para test de erosión y corrosión

Abstract

[EN] The neutron time-of-flight facility at CERN is a source of high flux of neutrons obtained by the spallation process of 20 GeV/c protons from the CERN Proton Synchrotron (PS) onto a high purity (99.99 %) lead target. A first facility, operational since 2001, was based on a non-monolithic lead target, an open cooling circuit of demineralized water and absence of control on the water chemistry. It suffered severe corrosion and extensive deformation of the central part of the target as a consequence of the local energy deposition and insufficient cooling. Excessive temperature at the beam interaction point induced creep in the lead blocks and activated severe pitting corrosion. After a few years of operation, the facility required a complete upgrade. The new facility, running since its commissioning in November 2008, consists of a monolithic lead cylindrical target with a diameter of 0.6 m, a height of 0.4 m and a total weight of 1.3 tons (Fig. 1). The target is cooled by a pressurized water cooling system, including an all-welded aluminium alloy vessel and a cooling circuit with monitored and controlled oxygen content and conductivity of the demineralized water used as coolant. The value of oxygen and conductivity is continuously monitored. Due to the presence of high electric fields, easiness of maintenance and prevention of corrosion, demineralized water is widely used at CERN in cooling circuits of particle physics detectors and accelerator magnets. During Long Shutdown (LS2) a third spallation target (target #3) will be installed in the facility, in order to substitute the present target which will have reached its end of life. The target is presently under design. Various materials are being investigated for both the core and the contact part with the cooling wáter. In the framework of the target upgrade activities, a new water erosion/corrosion test bench is required, in order to evaluate the compatibility of various materials in the n_TOF target regime. The new test bench is intended to be capable of being used to carry our erosion and corrosion tests, and check the changes in the microstructure of different materials to validate the design of the new target. This test bench will be designed in a way that will be versatile and adaptable to different types of tests and targets in the future. The vessel of the target was aluminium in the last experiments, in this case, stainless steel is going to be used to avoid chemical changes in the demineralized water. The idea is that the design can be modified variables such as flow, temperature, conductivity, oxygen content and pressure for different tests with good reading out and logging values. 3. TEST PROCEDURE. In order to simulate as close as possible the conditions prevailing in the real n_TOF experiment, a test loop circuit using demineralized water as process fluid was specially developed for the section (EN-STI-TCD). The test loop is shown in Figure 3. As in the real circuit, water passes through a mixed-bed resin cartridge in order to keep conductivity as low as possible. Water could be heated up to 80 °C in order to simulate the water temperature at the surface and the edge of the new material in order to create the worse scenario. The water circuit is split in two main branches, one adapt the water flow to the specific requirements for the test, such as temperature and conductivity, and the other allow the two different injections points in the chamber to carry out the desired test. The test bench is provided by an hydraulic pump (P1) that keeps the constant flow, three valves (V1, V2, V3) to isolate the branches and two flow regulator valves, manual (VR3) and electronic (VR2), to adjust the flow automatically depending on the needs. The duration of the test is expected to be three months for each material. The amount of erosion/corrosion will be continuously monitored. The most adapted material will be chosen after evaluation of the resul

Read accessible full text

Diseño y fabricación de un banco de pruebas hidráulico para test de erosión y corrosión

Author: Carrió Pérez, Pedro
Publisher: Universitat Politècnica de València
Year: 2017
Source: https://riunet.upv.es/bitstream/10251/89732/1/CARRI%c3%93%20-%20DISE%c3%91O%20Y%20FABRICACI%c3%93N%20DE%20UN%20BANCO%20DE%20PRUEBAS%20HIDR%c3%81ULICO%20PARA%20TEST%20DE%20EROSI%c3%93N%20Y%20CORROSI%c3%93N.pdf
UNIVERSITAT POLITÈCNICA DE VALÈNCIA
Depa amen o de Ingenie ía Mecánica y de Ma e iales
T abajo Fin de Más e en Ingenie ía Mecánica
DESIGN & DEVELOPMENT OF A HYDRAULIC TEST BENCH
FOR AN EROSION & CORROSION TEST
P esen ado po : D. Ped o Ca ió Pé ez
Di igido po : D . F ancisco Da id Denia Guzmán
Supe iso del CERN: Oli e Abe le
Valencia, Sep iemb e de 2017
ii
Ne e walk on he a elled pa h
because i only leads whe e o he s ha e been.
Alexande G aham Bell
iii
Abs ac
The neu on ime-o - ligh (n_TOF) acili y a CERN is a sou ce o high
lux o neu ons ob ained by he spalla ion p ocess o 20 GeV/c p o ons on o
a high pu i y (99.99 %) lead a ge . In 2008 he acili y, ha su e ed se e e
co osion a e ew yea s o ope a ion, was comple ely upg aded. I includes
now an aluminium alloy essel and a cooling ci cui wi h moni o ed oxygen
con en , and conduc i i y o he demine alized wa e used as coolan and
mode a o . Du ing Long Shu down (LS2) a hi d spalla ion a ge ( a ge #3)
will be ins alled in he acili y, in o de o subs i u e he p esen a ge which
will ha e eached i s end o li e. The a ge is p esen ly unde design. Va ious
ma e ials a e being in es iga ed o bo h he co e and he con ac pa wi h he
cooling wa e .
In he amewo k o he a ge upg ade ac i i ies, a new wa e
e osion/co osion es bench is equi ed, in o de o e alua e he compa ibili y
o a ious ma e ials in he n_TOF a ge egime.
The new es bench is in ended o be capable o being used o ca y ou
e osion and co osion es s, and check he changes o he ma e ials o alida e
he design o he new a ge . This es bench has been designed in a way ha
will be e sa ile and adap able o di e en ypes o es s and a ge s in he
u u e.
i
The idea is ha he design can modi y a iables such as he ci cui low,
empe a u e, luid conduc i i y and p essu e o di e en es s wi h good
eading ou and logging alues.
The p ojec o he es bench comp ises he concep ual o he de ailed
design and manu ac u e he de ice. This wo k has ca ied ou in collabo a ion
wi h se e al g oups ac oss CERN, in ol ed in he echnical speci ica ion,
analysis, design, manu ac u e and secu i y.
Key Wo ds
CERN, physics, pa icles, spalla ion, p o on synch o on, design
hyd aulic, es bench, e osion, co osion, demine alized wa e , conduc i i y,
hea , low, manu ac u ing.
Resumen
La ins alación de neu ones de iempo de uelo (n_TOF) en el CERN es
una uen e de al o lujo de neu ones ob enida po el p oceso de espalación de
20 GeV/c de p o ones sob e un obje i o de plomo de al a pu eza (99,99 %).
En 2008, la ins alación su ió una co osión se e a después de pocos años de
ope ación, és a ue comple amen e mejo ada. Incluye aho a un ecipien e de
aleación de aluminio y un ci cui o de e ige ación con con enido de oxígeno
y conduc i idad del agua desmine alizada, u ilizada como e ige an e y
mode ado , con olados. Du an e la pa ada la ga (LS2) se ins ala á un e ce
obje i o de espalación (obje i o #3) en la ins alación, pa a sus i ui al obje i o
ac ual que hab á alcanzado su in de ida ú il. El obje i o es á ac ualmen e
bajo diseño. Se es án in es igando a ios ma e iales pa a el núcleo y la pa e
en con ac o con el agua del sis ema de e ige ación.
En el ma co de las ac i idades de mejo a de obje i os, se equie e un
nue o banco de p uebas de e osión / co osión, con el in de e alua la
compa ibilidad de di e sos ma e iales en el égimen del obje i o n_TOF.
El nue o banco de p uebas es á des inado a se capaz de ealiza p uebas
de e osión y co osión, y e i ica los cambios de los ma e iales pa a alida
el diseño del nue o obje i o. Es e banco de p uebas ha sido diseñado de una
mane a que se á e sá il y adap able a di e en es ipos de p uebas y obje i os
en el u u o.

i
La idea es que el diseño puede modi ica a iables ales como el lujo del
ci cui o, la empe a u a, la conduc i idad del luido y la p esión pa a
di e en es p uebas con buenos alo es de lec u a y egis o.
El p oyec o del banco de p uebas comp ende del diseño concep ual al
diseño de allado y la ab icación del disposi i o. Es e abajo se ha lle ado a
cabo en colabo ación con a ios g upos del CERN, in oluc ados en la
especi icación écnica, análisis, diseño, ab icación y segu idad.
Palab as Cla e
CERN, ísica, pa ículas, espalación, p o on synch o on, diseño
hid áulico, banco de p uebas, e osión, co osión, agua desmine alizada,
conduc i idad, calo , caudal, ab icación.
ii
Resum
La ins al·lació de neu ons de emps de ol (n_TOF) al CERN és una on
d'al lux de neu ons ob inguda pel p océs d’espal·lació de 20 GeV/c de
p o ons sob e un objec iu de plom d'al a pu esa (99,99 %). El 2008, la
ins al·lació a pa i una co osió se e a desp és de pocs anys d'ope ació,
aques a a se comple amen millo ada. Inclou a a un ecipien d'alia ge
d'alumini i un ci cui de e ige ació amb con ingu d'oxigen i conduc i i a
de l'aigua desmine ali zada, u ili zada com a e ige an i mode ado ,
con ola s. Du an la pa ada lla ga (LS2) s'ins al·la à un e ce objec iu de
espal·lació (objec iu #3) en la ins al·lació, pe subs i ui l'objec iu ac ual que
hi hau à a iba al seu i de ida ú il. L'objec iu es à ac ualmen so a disseny.
S'es an in es igan di e sos ma e ials pe al nucli i la pa en con ac e amb
l'aigua del sis ema de e ige ació.
En el ma c de les ac i i a s de millo a d'objec ius, es eque eix un nou
banc de p o es d'e osió / co osió, pe al d'a alua la compa ibili a de
di e sos ma e ials en el ègim de l'objec iu n_TOF.
El nou banc de p o es es à des ina pe a se capaç de eali za p o es
d'e osió i co osió, i e i ica els can is dels ma e ials pe alida el disseny
del nou objec iu. Aques banc de p o es ha es a dissenya d'una mane a que
se à e sà il i adap able a di e en s ipus de p o es i objec ius en el u u .
iii
La idea és que el disseny po modi ica a iables com a a el lux del
ci cui , la empe a u a, la conduc i i a del luid i la p essió pe a di e en s
p o es amb bons alo s de lec u a i egis e.
El p ojec e del banc de p o es comp èn del disseny concep ual al disseny
de alla i la ab icació del disposi iu. Aques eball s'ha du a e me en
col·labo ació amb di e sos g ups del CERN, in oluc a s en l'especi icació
ècnica, anàlisi, disseny, ab icació i segu e a .
Pa aules Clau
CERN, ísica, pa ícules, espal·lació, p o on synch o on, disseny
hid àulic, banc de p o es, e osió, co osió, aigua desmine ali zada,
conduc i i a , calo , cabal, ab icació.
ix
Acknowledgemen s
The pe iod o his hesis b ough me aluable new expe iences in bo h a
scien i ic and pe sonal pe spec i e. This hesis would no ha e been possible
wi hou he cons an suppo and mo i a ion o many people.
Fi s ly, I would like o exp ess my since e g a i ude o my supe iso ,
Oli e Abe le. His wide knowledge and his expe ience ha e been a g ea
alue o me. His encou aging and pe sonal guidance ha e p o ided a good
backg ound o he p esen hesis.
Many hanks o he p o esso F ancisco Da id Denia Guzmán and Ma ia
Papamichali, o accep ing o e iew my wo k. Thei cons uc i e sugges ions
on he hesis a e eally app ecia ed o me. I am g a e ul ha in he mids o
hei ac i i ies, hey accep ed hese asks.
Du ing his pe iod, I ha e collabo a ed wi h many colleagues, om
di e en depa men s, he sec ion wo kshop and supplie s o whom I ha e
g ea ega d, and I wish o ex end my wa mes hanks o all hose who ha e
helped me om a p o essional and pe sonal poin o iew. Especially o
Ma co Cal iani, Damien G enie , Ch is ophe Claude Mi i io , Hube
Reymond, Domingo Gomez Dominguez, Yannic Body, Yann Lupkins,
Ramon Folch, Pie e B oussa , Edoua d G enie -Boley, Abdelhabib
Majbou , Edmundo Lopez Sola, Ra aele Esposi o, Tomasz Kazimie z
S zalka.
x i
Glossa y
CERN: Conseil Eu opéen pou la Reche che Nucléai e
LHC: La ge Had on Collide
LS: Long Shu down
PS: P o on Synch o on
PSB: P o on Synch o on Boos e
SPS: Supe P o on Synch o on
ALICE: A La ge Ion Collide Expe imen
ATLAS: A To oidal LHC Appa a uS
CMS: Compac Muon Solenoid
LHCb: La ge Had on Collide beau y expe imen
AD: An ip o on Decele a o
N_TOF: Neu on Time-O -Fligh
ECR: Elec on Cyclo on Resonance
RFQ: Radio F equency Quad upole
LEIR: Low Ene gy Ion Ring
RF: Radio F equency
HV: High Vol age

x ii
To my amily…
x iii
1
Chap e 1
1. In oduc ion
1.1. CERN
The Eu opean O ganiza ion o Nuclea Resea ch is he wo ld’s
la ges pa icle physics labo a o y [1]. I is si ua ed a Gene a ac oss
he bo de be ween F ance and Swi ze land. The o igin o he name
comes om he F ench ac onyms “Conseil Eu opéen pou la
Reche che Nucléai e”.
I was ounded in 1954 by 12 membe s a es and nowadays he
membe s a e 22, bu i ga he s physicis s and enginee s om mo e han
100 coun ies in he wo ld.
13.7 billion o yea s ago he Uni e se s a ed wi h an explosion.
The ene gy, concen a ed in an in ini ely small space, coagula ed in
equal pa o ma e and an ima e . A CERN physicis s and enginee s
wo k o push mo e and mo e in he esea ch bounda ies p obing he
undamen al s uc u e o he Uni e se.
In o de o a ain his main goal, pa icle physics s udies he ines
objec o Na u e. Looking in o he e y small and undamen al is also
looking a back in o ime, jus a ew momen s a e he Big Bang [2].
2
The in es iga ion o he undamen al pa icle, he basic cons i uen
o he ma e , is made by he collision o wo pa icle beams in he LHC
(La ge Had on Collide ), he mos powe ul accele a o in he wo ld.
The empe a u e and ene gy densi y in he collision a e simila o
hose ha exis ed a ew momen s a e he Bing Bang, in his way
physicis s hope o disco e how he Uni e se e ol ed.
The CERN ac i i y is egula ed by an al e na ion o wo pe iods:
he “Run”, when he beams ci cula e in he unnel and he expe imen s
can collec he da a, and he “LS: Long Shu down” when he beams a e
s opped and i is possible o make he main enance and he upg ades o
he de ices inside he machines.
F om he beginning o he LHC ope a ion, in 2008, CERN has had
wo pe iods: he “Run1” and he “LS1”. The ene gy eached in he
collision o Run1 was 8 TeV and, hanks o his, he expe imen s ha e
made he e y impo an disco e y o he Higgs Boson, which
imp o es ou cu en unde s anding o he Uni e se, con i ming he
S anda d Model Theo y.
Du ing he wo yea s o he LS1 he powe o he machine was
inc eased up o an ene gy o collision o 14 TeV.
3
Nowadays, we a e in he Run2, ha is going o be concluded in he
middle o 2018, he LS2 will ollow and hen, he Run3 is o eseen o
2020.
1.1.1. The CERN's accele a o complex.
Be o e hei collision, he pa icles a e accele a ed o a speed e y
close o he speed o ligh . This is made by a succession o machines
ha , s ep by s ep b ings he pa icles o highe ene gies and speed, as
he Figu e 1 shows.
Figu e 1: Schema ic accele a o complex.

4
Each machine injec s he beam in o he nex one, which akes o e
o b ing he beam o an e en highe ene gy, and so on. The LHC is he
las elemen o he chain, he e he pa icle beams a e accele a ed in
o de o each he ene gy o 13 TeV.
The LHC s ands o La ge Had on Collide . La ge is due o i s size
abou 27 km in ci cum e ence, Had on is because i accele a es p o ons
o lead ions, which a e had ons, and Collide because he pa icles
a el in o wo beams in opposi e di ec ion and in ou poin s o he
ing hey collide.
The accele a o s inside he machine a e no mally ci cula unnels
in which he pa icles a e ben o ollow a ci cula ajec o y by he
dipole magne s.
The quad upole magne s a e used o ocus he beam down o he
smalles possible size be o e he collision poin , in o de o maximize
he chance o wo p o ons o smash wi h each o he . In he LHC he
supe conduc i e magne s use a c yogenic sys em wi h a pumping o
supe luid helium ha pe mi s o ope a e a a empe a u e o -273.1 °C,
which makes i one o he coldes places on Ea h. The Figu e 2 shows
a quad upole in he LHC.
5
Figu e 2: LHC.
The accele a ion, ins ead, is made by he ans e o he beam inside
he adio equency ca i ies. The elec ic ield, c ea ed wi hin he
ca i ies, changes pola i y wi h high equency. The incoming p o ons
a e a ac ed inside he ca i y when he posi i e sign o he elec ic ield
is in he same di ec ion o he beam, while when he sign is in he
opposi e di ec ion, he pa icles a e pushed wi h highe speed.
All o he physics s a s om a li le bo le o hyd ogen, he p o ons
o which a e aken by s ipping o bi ing elec ons. These p o ons a e
accele a ed o an ene gy o 50 MeV wi h a ise o he mass o he 5 %
in he Linac2, one o he only wo linea accele a o s o he chain.
6
The p o ons a e hen injec ed in o he PS Boos e (PSB) ha
accele a es hem o 1.4 GeV and 91.6 % o he speed o ligh . A e , he
beam is going h ough he P o on Synch o on (PS) whe e i is
accele a ed o 25 GeV and 99.93 % o he speed o ligh . The Supe
P o on Synch o on (SPS) ollows, whe e he ene gy eaches 450 GeV.
Finally, a e i e accele a o s ha inc ease he mass o he p o ons
o 7000 imes and hei eloci y ill 99.9999991 % o he speed o ligh ,
he beam is ans e ed o he LHC in bo h di ec ions, clockwise and
an iclockwise. In he LHC hey a e accele a ed o 20 minu es o hei
nominal ene gy o 13 TeV. Beams will ci cula e o many hou s inside
he ul a-high acuum beam pipes unde no mal ope a ing condi ions.
The wo beams o he LHC collide only a ou speci ic poin s in
co espondence o he de ec o s. The de ec o s a e machines ha
ecognize and measu e p ope ies o he pa icles du ing he collisions,
ying o unde s and hei na u e. Each de ec o is ela ed o he one o
he main expe imen s a CERN: ALICE [3], ATLAS [4], CMS [5] and
LHCb [6].
In addi ion, mos o he accele a o s in he chain ha e hei own
expe imen al halls wi h a beam o lowe ene gy. Fo example he SPS
supplies se e al expe imen s, among hem ix a ge expe imen s a e
he No h A ea NA62 [7] o he COMPASS [8] expe imen s. SPS also
7
p o ides he pa icles o he HiRadMa [9] o expe imen s on ma e ials
hi by he beam; ins ead, he PS injec s he pa icles o he An ip o on
Decele a o (AD) [10] o o n_TOF [11]. In Figu e 3, wo pic u es o
he CMS and ATLAS colliding beam pa icle de ec o s can be seen.
Figu e 3: CMS and ATLAS expe imen a CERN
Mo eo e , he accele a o complex also accele a es lead ions. The
p oduc ion o he lead ions comes om a highly pu i ied lead sample,
he ECR (Elec on Cyclo on Resonance) [12] sou ce, which is hea ed
o a empe a u e o abou 500 °C.
The lead apou is ionized by an elec on cu en ; he sou ce
p o ides many di e en cha ge s a es, a spec um om Pb22+ o
Pb32+, wi h a maximum a ound Pb29+, which is he only o he
cha ged s a es selec ed immedia ely by a spec ome e . Now, he a el
o hese ions s a s om he linea s uc u es which accele a es he ions
o 4,2 MeV/u (ene gy pe nucleon): A Radio F equency Quad upole
(RFQ) and hen an in e -digi al Linea Accele a o , he Linac3, play
his ole.
14
Chap e 2
2. Desc ip ion o he ins alla ion
In he amewo k o he a ge upg ade ac i i ies, a new wa e
e osion/co osion es bench is equi ed, in o de o e alua e he
compa ibili y o a ious ma e ials in he n_TOF a ge egime.
The new es bench is in ended o be capable o being used o ca y
ou e osion and co osion es s, and check he changes in he
mic os uc u e o di e en ma e ials o alida e he design o he new
a ge . The es bench is designed o be as mos au oma ic as possible,
only equi ed om he ope a o he se ‐up o he ma e ials inside he
chambe , some manual al es and he manual s a o he es , i is
designed in a way ha i will be e sa ile and adap able o di e en
ypes o es s and a ge s.
The expe imen is ca ied ou inside a s ainless s eel chambe ,
whe e he a ge is ins alled. To ge all he special condi ions is needed
di e en de ices o i . The demine alized wa e is used in he eal
expe imen o a oid any ma e ial oxida ion in he a ge o in he ci cui
componen s. In his case is going o be used an Ion exchange equipmen
in s ainless s eel ha ul il all he equi emen s needed. The mixed bed
ion exchange s wo k acco ding o he ion exchange p inciple.

15
The idea o he es bench is he design can modi y a iables such
as he ci cui low, empe a u e, luid conduc i i y, and p essu e o
di e en es s wi h good eading ou and logging alues. To ge all o
hese op ions a ull ma ke esea ch has been done o ind he bes
comme cial componen s ha can achie e all he equi emen s. The
ci cui is composed by a hyd aulic pump ha gi es he low, a hea e
ha gi es he empe a u e, lowme e s and al es o con ol he low
and di e en senso s o con ol he conduc i i y and he p essu e.
E e y hing is placed on aluminium s uc u e, in o de o educe he
weigh , wi h ou wheels o make easie he placemen and o ind a
wa e ap connec ion. A ec angula design wi h wo aluminium shee s
o 10mm hickness ha e been chosen o make easie he design, he
assembly i sel , and he assembly o all he componen s, also, he space
o he elec onic de ices has been aken in o accoun .
16
The Figu e 6 p esen s he schema ic a ge and low impac inside
he chambe :
Figu e 6: Scheme o he a ge .
2.1. Tes p ocedu e
In o de o simula e as close as possible he condi ions p e ailing in
he eal n_TOF expe imen , a es loop using demine alized wa e as
p ocess luid was specially de eloped and connec ed on me al based
loop ci cui a ailable a CERN. The es loop is shown and desc ibed
in Figu e 7.
The Tes ‐Bench mus be ins alled in a loca ion wi h access o
no mal ap wa e , sewe o discha ge used wa e , 380 V iphasic
elec ic connec ion and E he ne connec ion o in e ne access.
10 mm
10 mm
17
As in he eal ci cui , wa e passes h ough a mixed-bed esin
ca idge in o de o keep conduc i i y as low as possible. Wi h a low
egula o al e, placed on he inle ca idge, is possible o con ol he
low once he conduc i i y is sui able. The wa e is hea ed up o 80 °C
o simula e he wa e empe a u e a he su ace o he es ma e ial,
whe e he beam hi s he a ge . The e a e wo empe a u e senso ha
will con ol he hea e ope a ion.
The wa e ci cui is spli in wo main b anches, one as a by-pass in
o de o adap he luid o he expe imen equi emen s and he o he
b anch has wo di e en injec ions poin s in a mock-up con aine .
Once he wa e is p epa ed wi h all equi emen s o he es
(p essu e, empe a u e, oxygen and conduc i i y), he al es V1 and
V2 will be opened manually o allow he luid o pass h ough he
second b anch o s a he p ocess. This is he mos impo an b anch
because all he expe imen depend on o main ain he low condi ions
in hese pipes. The al es VR2 and VR3 will be in cha ge o adjus ing
he p ope low h ough he es chambe . VR3 is a manual egula o
low al e ha will be se up a he beginning and VR2 will con ol he
low du ing all he expe imen as an au oma ic low egula o al e
(see Figu e 7).
18
The Tes ‐Bench was designed o send au oma ically he epo s,
wa nings/ale s o e email o he pe son in cha ge. Also, se e al
in e lock sys ems we e implemen ed o p e en and ale CERN abou
es bench inco ec use o mal unc ions.
P 1
CT 1
OT 1
ERA 1
TT 1
TT 2
Tes Chambe
Cons an T
Cons an
REA 1
PI 1
PT 1
V 4 Re illing
Expansion
Tank
5µm - 200µm
FOA 2
VR 1
200µm
DEMINERALIZED
WATER
10% o low
V 1
FT 1
FS 1
V 6
3 Ba
PI 2
V 3
FOA 1
V 2 VR 3
FT 2
VR 2
V 5
Figu e 7: Scheme o he hyd aulic sys em.
The du a ion o he es is expec ed o be h ee mon hs o each
ma e ial. The amoun o e osion/co osion will be e i ied a he end o
he es by di e en p ocesses such as dimensional e i ica ion o
mic oscopy. The mos adap ed ma e ial will be chosen a e e alua ion
o he esul s.
19

20
Chap e 3
3. Ambi o he p ojec .
The p esen p ojec includes he design and manu ac u ing o he
mechanical pa o a hyd aulic es bench o check he e osion and
co osion o di e en ma e ials and hus alida e he nex a ge o he
n_TOF expe imen a CERN.
The es samples and he ins alla ion o hem inside he chambe a e
no conside ed pa o he p ojec .
Pa s o he elec onic design, so wa e design, and design o he
wi ing, p o ec ion and con ol elemen s a e no conside ed pa o he
p ojec . Howe e , hese elemen s mus comply wi h he IEC S anda ds.
The limi s o each sec ion a e speci ied in he poin s desc ibed
below.
3.1. Tes bench links.
Only he mechanical links o he bench es o he ou side will be
conside ed as belonging o he p esen p ojec . In his case i would be
he s anda d wa e inle / ou le connec ion and he ex ac ion al e in
he chambe .
21
The design o he inpu elec ical connec ion and i s sa e y ea u es
a e no conside ed as belonging o he p ojec .
4. Design o he ins alla ion
The design o he hyd aulic es bench began in Feb ua y 2016 wi h
he objec i e o de e mina e he bes ma e ial op ion o he new a ge
in e ms o e osion and co osion. The es bench should ec ea e, as
close as possible, he condi ions in he eal n_TOF expe imen .
4.1. Limi a ions o he sys em
• Fluid o use: Demine alized wa e
• Flow a e: 1-90 L/min
• Fluid empe a u e: 20 °C / 80 °C
• Wa e speed inside he chambe : 0.1 o 5 m/s
• P essu e inside he chambe : 1-3 ba
• O2 con en : ppm (1/10 o ppm)
• Conduc i i y: 0.175 µS/cm
• Tow impac angles: 90 deg. 0 deg.
• 10 mm je diame e
• Con aine sealed capable o eaching up o 100 °C
• Tes con aine : X: 30-40 cm; Y: 30-40 cm; Z: 15-30 cm
• S uc u e: X: 150-200 cm; Y: 80-100 cm; Z: 60-110 cm
22
Based on he gene al limi a ions posed by he sys em, a concep ual
design is ca ied ou . The comme cial de ices necessa y o achie e he
equi emen s a e ag eed upon, h ough discussions wi h o he sec ions
ha will con ibu e hei expe ience and wo k.
All he equipmen is moun ed on a ec angula s uc u e, able o
mo e easily. The selec ion o he equipmen is de e mined by he
es ablished equi emen s om he beginning, he equi emen s ha he
componen s hemsel es equi e and he amoun o wa e o be mo ed.
The hyd aulic es bench consis s o a hyd aulic pump, a hea e ,
elec ical esis ances, con ol and p o ec ion elemen s (manome e s,
he mome e s, il e s, compensa o s, elec o- al e, he mos a s,
esis o s and low swi ch), and sphe ical closing al es o low
adjus men and also he pipes.
The ci cui consis s o pipes in cha ge o p o iding he equi ed
p essu e o he chambe . The use is esponsible o make he pipe
connec ion. The en ance o he es bench ci cui is made by a quick
connec ion o 1".
23
To design all he sys em, he empe a u e, p essu e and low mus
be de e mined in a gene al way, aking in o accoun ha he mos
c i ical elemen is he chambe whe e he es s will be ca y ou .
4.2. Desc ip ion o he elemen s o he ins alla ion
This chap e will desc ibe he p ocess o he design o each
componen , he calcula ions de eloped, he componen s selec ed and
he solu ions ha ha e been adop ed. In he Figu e 8 one may isualize
he componen s o he ins alla ion o be designed.
Figu e 8: Tes bench design.
30
The equi emen s o he selec ion o he hyd aulic pump a e:
• Flow a e: 1-90 L/min
• Fluid empe a u e: 20 °C / 80 °C
• P essu e inside he chambe : 1-3 ba
• Ma e ial: s ainless s eel
Fo he design we chose he e ical pump, because o he inhe en
limi ed space o he s uc u e. The empe a u e and ma e ial
equi emen s a e qui e es ic i e and lea e us ew op ions in he
ma ke . The b and DAB is chosen, because i s eliabili y, based on
p e ious expe ience, he sho deli e y ime and he p ice.
The e a e wo sui able op ions o he design o he es bench
a ailable in he DAB ca alogue; KV 3/12 T and KV 3/15 T. The chosen
pump o his p ojec is KV 3/12 T because i wo ks p ope ly o he
design demands and i is he cheapes op ion. The Table 2 ep esen s
he pump ea u es.
Table 2: Hyd aulic pump ea u es.

31
4.2.3.2. Wa e ci cula ion hea e
Ci cula ion hea e s a e compac hea ing sys ems o luids and
gases, gene ally ha e mul iple elemen s moun ed on a sc ew plug o
lange i ing and placed in an insula ed essel h ough which liquids
passes. The medium is hea ed as i lows pas he hea ing elemen s.
These hea e s a e ideal o applica ions such as pu i ied and ine
gases, supe c i ical luids and liquids like de-ionized wa e o use in
semiconduc o and elec onics indus ies as well as o gene al liquid
and gas hea ing applica ions.
Elec ic hea e s wo k on he p inciple o esis i e hea ing in which
an elec ical cu en passing h ough a conduc o c ea es hea .
Acco ding o Joule’s Fi s Law, he hea p oduced by an elec ic cu en
is equal o he p oduc o he esis ance o he conduc o , he squa e o
he cu en , and he ime o which i lows.
Hea ing sys ems gene ally equi e p ecise ins umen s o each and
main ain app op ia e empe a u es and o con ol he la ge cu en load
need o c ea e hea om elec ici y. [17]
Ci cula ion hea e s a e made o a s eel o s ainless s eel essel
equipped wi h 3 nozzles (inpu -ou pu ) and wi h an imme sion hea e
32
(gene ally h eaded) equipped wi h space o add a con ol de ice
empe a u e. In ou case, a s ainless s eel essel was used in o de o
a oid changes in he composi ion o he demine alized wa e .
In his applica ion, a e de e mining he amoun o hea equi ed,
4.5 kW, and aking in o accoun he a ge empe a u e o be eached,
he needs o he ci cula ion hea e a e de ined and, conside ing also
he supplie s’ knowledge, he mos app op ia e de ice is he CETAL
RPM-C-045 (de ailed in o ma ion in he Wa e ci cula ion hea e
calcula ions chap e ). The Figu e 11 shows he hea e 3D model.
Figu e 11: Scheme o he hea e .
33
4.2.3.3. Flow secu i y
I is a con ol elemen ha is used o e i y ha he pump is pushing
wa e , o he wise i sends a signal o he equipmen and s ops i
immedia ely. Fo hese low con ol de ices, he KOBOLD b and has
been chosen o i s eliabili y and in his case he DUK-12G8H S300L.
4.2.3.4. Flowme e
I is a measu ing ins umen o he low o olume ic low o a
luid o he mass low. These de ices a e usually placed in line wi h he
pipe ha ca ies he luid. The e a e mechanical and elec ical e sions.
The la ge numbe o heo ies and models o lowme e s ha exis
con i m us ha he e is no ideal o mula o measu e low. The inal
decision mus be made based on he low a e, iscosi y, empe a u e,
chemical composi ion and p essu e o he luid we wish o measu e.
Each applica ion has a lowme e ha bes sui s you needs. In his case
we ha e decided o use he ul asonic lowme e o i s adap abili y o
he empe a u e, as well as because o he low a e and iscosi y o he
luid o be used.
KOBOLD DUK Se ies compac ul asonic lowme e s a e he
de ices chosen o ou es bench and a e used o measu ing,
moni o ing, and ba ching o low iscosi y (wa e -based) luids. These
34
compac ul asonic lowme e s wi h no mo ing pa s wo k on he
p inciple o un- ime di e ence. Ul asonic wa es in he media a e
in luenced by he a e o low. I ea u es wo senso s moun ed opposi e
one ano he in he pipeline, which unc ion simul aneously as
ansmi e and ecei e o ul asonic signals. I no low is p esen , he
un- imes o bo h signals a e iden ical. I he media is lowing, he un-
ime o he signal agains he low will be longe han he signal wi h
he low. This un- ime di e ence, de e mined by he KOBOLD DUK
Se ies compac ul asonic lowme e ’s mic op ocesso , is p opo ional
o low a e. [18]
Fo he design wo ul asonic lowme e s a e needed because o he
ac ha we wan o moni o , mainly, he inpu low o he chambe ,
and hen o con ol he es o he low.
4.2.3.5. Flow egula o al e
They a e de ices ha egula e o a ixed low, independen o he
luc ua ion o he wo king p essu es and s a low a es. They con ol
he ope a ion o pumps o egula e he pe o mance o all sys ems and
ins alla ions. Wi h hem one can con ol he p essu e and speed in each
b anch o he ins alla ion.
Fo he es bench h ee low egula o s ha e been ins alled, one
elec onic ha will be loca ed a he end o he ci cui and wo manuals,
35
one o hem will emain ixed once he sys em has been se up and
ano he will con ol he low o luid ha en e s he deionize o
main ain he same le el o conduc i i y in he wa e . The compu e
p og am will con ol he elec onic al e o decide he speed o he
luid impac ing agains he a ge s o he expe imen .
4.2.3.6. Senso s
As he idea o he design is ha i should modi y a iables such as
low, empe a u e, conduc i i y, oxygen con en and p essu e ( he low
is al eady con olled by he lowme e s), di e en senso s a e equi ed
in o de o ob ain good eadings and logging alues.
Fo he empe a u e, wo senso s a e chosen o be placed; one
be o e he inpu and he o he a e he ou pu o he hea e . The senso
be o e he inpu o he hea e will be he M12TXC-PT100-3MM-
24MM-G1/8 om OMEGA b and, because in his poin i is no needed
a lo o accu acy and his senso is jus a cheap ansmi e ha i is
possible o log he alues by he compu e . Fo he ou pu posi ion mo e
accu acy and isual logging a e needed, his poin will be he p ima y
in o ma ion abou he empe a u e, he KOBOLD TDA15H2R4 0 L3M
has been chosen.
A e a deep ma ke esea ch, conside ing all he high equi emen s
o he es bench, he conduc i i y and oxygen senso s ha e been

36
chosen in one ex e nal ansmi e , because i was he only op ion in he
ma ke ha ul illed all he equi emen s. These a e Dissol ed Oxygen
Senso HX438-01, Conduc i i y Senso 400VP-11 and In elligen Dual
Inpu Analyse 1056-03-20-35-HT om EMERSON b and.
The p essu e senso is a ansduce ha con e s p essu e in o an
analogue elec ical signal. Al hough he e a e a ious ypes o p essu e
ansduce s, one o he mos common is he s ain gauge based
ansduce .
The con e sion o p essu e in o an elec ical signal is achie ed by
he physical de o ma ion o s ain gauges which a e bonded in o he
diaph agm o he p essu e ansduce and wi ed in o a Whea s one
b idge con igu a ion. P essu e applied o he p essu e ansduce
p oduces a de lec ion o he diaph agm which in oduces s ain o he
gauges. The s ain will p oduce an elec ical esis ance change
p opo ional o he p essu e. The PXM309-007GI om OMEGA has
been chosen. Also, he e a e wo manome e s ins alled on he es bench
o gi e an easie isual con ol o he p essu e.
37
4.2.3.7. Fil e
Thei unc ion is o a oid pa icles in he ci cui s, which could
obs uc he ci cui o damage he hyd aulic pump. In his case, wo
il e s a e used, one a he end o he ci cui o collec he pa icles ha
can a i e om he e osion o he ma e ials o be es ed and ano he o
con ol he luid a e he ou le in he deionize .
The one ha is ins alled a he end o he ci cui consis s o a
s ainless s eel con aine in which he e is a ca idge o cellulose ibe
ha allows a il a ion o 50 μm. The ollowing is a special Y- il e o
demine alised wa e consis ing o an inne 0.5 mm mesh.
4.2.3.8. Piping and i ing
The design has been made based on he necessa y equi emen s in
he chambe , he minimum pipe sec ion wi h which he sys em can
ope a e was calcula ed and each pipe b anch was s udied sepa a ely in
o de o ob ain he op imal design. The size o he i ings is ob ained
by he pipe size.
38
4.2.4. Chambe
The chambe is a s ainless s eel p essu ed essel whe e he es wi h
he ma e ials will be pe o med. I will be placed on he op o he
aluminium s uc u e, be ween he wo holes because hey a e he access
o he pipes.
Two luid inle s will pe o m he je s which will allow he luid
impac on he ma e ials wall.
In he uppe pa wo co e s ha e been designed, one o which will
always emain closed by i s ixed connec ion wi h he ou le pipe and
he o he co e will be in cha ge o acili a ing access o he pa s o be
es ed.
4.3. Componen s calcula ions
In his chap e , he design decisions aken and calcula ions
pe o med in he design p ocess a e going o be desc ibed. In he Figu e
12 is possible o see all he componen s o he es bench.
39
Figu e 12: Tes bench componen s.
4.3.1. Calcula ion o he hyd aulic pump
The pump is esponsible o p o iding he necessa y p essu e o
he ci cula ion o he demine alized wa e in he closed ci cui o he
bench es . The p essu e o be supplied is a leas 4 ba as men ioned in
he Limi a ions o he Sys em The low o be d i en is he minimum
equi ed by he sys em equi emen s plus 20 % o o e size, 110 L/min.
Wi h hese da a and aking in o accoun ha he elec ical
connec ions a ailable a e 400 V and 50 Hz, conside ing he DAB
46
Thus, o he low a e o 90 L/min o be pumped, by imposing he
maximum speed he eo , he minimum diame e o he pipe is
de e mined immedia ely, which in his case co esponds o a nominal
diame e o DN25 mm o 1 in:
𝑆 = 𝑣
𝑄= 𝜋× (𝐷
2)2 (Eq.3)
Whe e:
Q: Flow
𝒗: Speed
S: Sec ion
D: Diame e
In any case, he no malized diame e should be chosen immedia ely
abo e he minimum alue. This alue is wha is known as he op imum
diame e o he pipe, because i ep esen s he lowes possible cos ,
mee ing he equi emen s as o he maximum speed o ci cula ion o
he luid h ough i .
Due o he design equi emen s o some de ices ha ha e been
p e iously selec ed, he pipes o DN40 mm o 1-1/2 in mus be chosen
in some sec ions. The Table 4 shows he connec ions o each elemen
o he ins alla ion.

47
Componen
Connexion
P essu e indica o
G 1/2 Male
P essu e ansmi e PXM309-007GI
G 1/4 Male
Tempe a u e ansmi e M12TXC-PT100-3MM-24MM-G1/8
G1/8 Male
Tempe a u e ansmi e M12TXSS-PT100-24MM-G1/8
G1/8 Male
Tempe a u e ansmi e TDA15H2R4 0 L3M
G 1/2 Male
Flow ansmi e DUK-12G8H C34PL
G 1" 1/2 Female
Flow secu i y DUK-12G8H S300L
G 1" 1/2 Female
Oxygen ansmi e
G1/8 Male
Conduc i i y
3/4" NPT Male
Pump
G 1" ¼ Female
Fil e s 5 µm – 200 µm
1" NPT Female
Hea e
G 1" Female
Expansion ank
G 1" Male
Wa e deionize
G 1/2 Male
Flow egula o au oma ic al e
G 1" 1/2 Female
Table 4: Connexion o he componen s.
Due o space limi a ions, he en i e ci cui mus be inse ed in o he
aluminium s uc u e and his means ha he e a e some changes o
di ec ion wi h hei consequen losses being will be s udied la e .
48
A his poin , each sec ion o pipeline o he ins alla ion will be
s udied. The Figu e 16 shows he inal layou o he bench es .
Figu e 16: Tes Bench design.
The i s sec ion, showed in he Figu e 17, consis s o he pump
ou pu o he hea e inle . The sec ions will always be designed as
s aigh as possible. In his sec ion, he low secu i y de ice DUK-
12G8H S300L is ins alled wi h which he low passage is con olled
o sa e y, hus a oiding a possible ailu e o he pump when i is in
ope a ion and he e is no low.
49
Due o he design es ic ions ha he sa e y de ice gi es us, we
will ha e o use a pipe wi h a diame e o 1-1/2 in and a leas 500 mm
o pipe leng h be o e and a e his o i s co ec ope a ion. Then, we
ha e ha he ou pu o he pipe is 1-1/4 in, we will use a educ ion
i ing om 1-1/2 in o 1-1/4 in. A e wa ds a lexible s ainless s eel
pipe will be used o abso b he possible ib a ions o he pump, hen
he sa e y lowme e is ins alled and inally a s ainless s eel pipe ixed
o he hea e . Fo he connec ion o he hea e , a educ ion a io o 1-
1/2 in o 1 in will be used. The empe a u e ansmi e M12TXC-
PT100-3MM-24MM-G1/8 is also ins alled in his sec ion.
Figu e 17: Fi s sec ion.
50
The second sec ion, showed in he Figu e 18, o piping would be
p o ided om he hea e ou le o he pump inle . In his sec ion one o
he low ansmi e DUK-12G8H C34PL is o be ins alled, so ha he
same es ic ions o sec ion 1 ha e been e-es ablished, he pipe has o
be 1-1/2 in and 500 mm be o e and a e he lowme e .
A he ou le o he hea e , a 1-1/2 in o 1 in educ ion i ing will be
used o connec he pipe o he de ice, hen he empe a u e ansmi e
TDA15H2R4 0 L3M will be ins alled and hen a T-connec ion will be
added, wi h an ou pu o 1 in which will be add essed in sec ion 3.
Con inuing wi h sec ion 2, he conduc i i y senso will be ins alled in
he nex co ne and hen a T-connec ion will be added, wi h an ou pu
o 1 in, o be explained in he sec ion 5.
Now he 500 mm inpu o he lowme e is conside ed, hen he
same a he ou pu and he di ec ion is aken o he pump wi h a sligh
inclina ion. In he las pa o his sec ion, he ollowing de ices will be
ins alled in his o de : low egula o au oma ic al e, 1-1/2 in T, 1-1/2
in o 1 in T, il e , 1-1/2 in o 1 in T and manome e . The connec ion
wi h he pump will be done by a lexible pipe o a oid he pump
ib a ions.
51
Figu e 18: Second sec ion.
The hi d sec ion, showed in he Figu e 19, s a s om he i s T
a e he hea e o he en ance o he chambe o he es bench. In his
case a 1-n pipeline will be used since we ha e no es ic ions due o he
equipmen ins alled and as p e iously es ima ed, he low necessa y o
each 5 m/s in he je s will be 50 L/min, so i is on he side o secu i y.
The p essu e ansmi e PXM309-007GI is ins alled in his line o log
he p essu e a he chambe inle . A 1-in al e is also ins alled o isola e
he chambe once he es is comple e.

52
Figu e 19: Thi d sec ion.
Sec ion ou , showed in he Figu e 20, consis s o he pipe
connec ing he chambe ou le o he hi d T-join o sec ion 2. In his
sec ion, a al e o isola e he chambe once he es is comple e is
ins alled, a manually low egula o al e is se up a he beginning o
he p ocess, he elec o al e con ols he low. The second low
ansmi e DUK-12G8H C34PL is ins alled o con ol he low h ough
he chambe . Following he equi emen s o he lowme e design, i
has o be conside ed a pipe o 1-1/2 in and 500 mm be o e and a e he
lowme e .
53
Figu e 20: Fo h sec ion.
The las sec ion consis s o a 1-in pipe ha connec s he second T-
connec ion o sec ion 2 wi h he ou h o i . This bypass has been made
o a oid, as a as possible, ha he oxygen senso suppo s oo much
p essu e since i s limi is 4 ba (Figu e 21).
54
Figu e 21: Fi h sec ion.
4.3.3.1. S ainless s eel pipe hickness
In his chap e , he calcula ions o he pipe hickness a e
demons a ed base on he in e nal p essu e using he ASME Code
B31.3 no ms.
The ma e ial selec ed o he pipes is de e mined by he expe imen
because he luid used is demine alized wa e and s ainless s eel is no
going o modi y he wa e cha ac e is ics.
The basic o mula o de e mining pipe wall hickness is he gene al
hoop s ess o mula o hin-wall cylinde s, which is s a ed as:
55
𝑡 = 𝑃𝑑𝑜
2(𝑆𝐸+𝑃𝑌) (Eq.4)
Whe e:
: pipe wall hickness, in.
P: in e nal p essu e o he pipe, psi.
dO: ou side diame e o pipe, in.
S: allowable s ess o pipe, psi.
E: longi udinal weld-join ac o .
Y: de a ing ac o .
Once he hickness esul s a e ob ained wi h Equa ion 4, one may
de e mine he minimum hickness conside ing he co osion and
e osion allowance. This is an ex a hickness called "co osion
allowance" and i is added o he wall hickness o compensa e he me al
expec ed o be los o e he li e o he equipmen . In his case, since
s ainless s eel and demine alized wa e a e used, he co osion is eally
low.
Compa ing he minimum hickness ob ained by he design wi h he
comme cial hickness o he pipes om he supplie s, he sa e y ac o
o 6.46 is sa e enough. The calcula ions a e pe o med in he Table 5.
62
This esul could be also app oxima ed analy ically aking in o
accoun he a ea o he inle s, 7.85·10-5 m2, and he olume ic low,
6.66·10-4 m3/s, hen:
𝑣 = 𝑉󰇗𝐴
⁄= 8.5 𝑚/𝑠. (Eq.5)
As can be seen in he Pic u e 26, whe e he eloci y dis ibu ion in
he a ge wall is shown, he maximum speed eached is 2.7 m/s, which
is lowe han he desi ed maximum alue o 5 m/s.
Figu e 26: Veloci y dis ibu ion in he a ge wall, i s i e a ion
con igu a ion.
The eloci y dis ibu ion along he wall can be also app ecia ed in
he ollowing g aph, whe e he wa e speed is displayed along a

63
ho izon al line loca ed in he a ge wall plane a he wa e inle s heigh
(shown in igu e 27). The a ge wall is delimi ed by wo lines, and i
can be seen once again ha he maximum eloci y eached by he wa e
impac ing he wall is a ound 2.7 m/s.
Figu e 27: Veloci y dis ibu ion along a ho izon al line placed in he a ge
wall. Maximum speed = 2.7 m/s.
These calcula ions a e done conside ing a wa e low o 80 L/min,
which is al eady close o he maximum alues ha will be used in
ope a ion. The e o e, he nex s ep o he analysis is o es ima e he
necessa y dis ance be ween he a ge wall and he box in o de o
ob ain he equi ed 5 m/s eloci y.
2.7 m/s
64
I is possible o ob ain he eloci y e olu ion along a ho izon al line
s a ing in each one o he wa e inle s and inishing in he a ge wall,
one igh in he co ne and he o he in he wall i sel , as shown in igu e
28.
Figu e 28: Ho izon al lines along wa e ajec o y used o ob ain he
eloci y e olu ion.
The Figu es 29 and 30 ep esen he eloci y e olu ion o he wa e
low along he wo lines. The eloci y o he wa e a Y=-0.15 m is
app oxima ely 5 m/s in bo h cases, which ep esen s a dis ance o 20
mm be ween he a ge and he box wall.
65
The e o e, he nex s ep o he analysis is o epea he calcula ions
wi h he a ge placed a a dis ance o 20 mm om he inle s, so ha he
wa e impac s he a ge wall wi h he equi ed speed o 5 m/s.
Figu e 29: Wa e low eloci y e olu ion along Y axis, om he inle o he
a ge co ne .
66
Figu e 30: Wa e low eloci y e olu ion along Y axis, om he inle o he
a ge cen e.
Final con igu a ion esul s
New CFD calcula ions a e pe o med wi h his con igu a ion, wi h
a dis ance be ween he a ge wall and he box inle s o 20 mm.
The wa e speed dis ibu ion in a XY plane coinciden wi h he
wa e inle s is shown in igu e 31. The maximum speed eached by he
wa e in he ci cui is a ound 10 m/s in his case. The eason is ha e en
i he inle a ea and he olume ic low a e he same, he p oximi y o
he a ge induces highe eloci ies o he wa e je coming om he
inle s.
67
Figu e 31: XY-plane eloci y dis ibu ion, a ge placed 20 mm a om he
box wall. 80 L/min wa e low.
Wi h his con igu a ion, he maximum eloci y eached in he a ge
wall is a ound 5 m/s as desi ed, which is shown in he Pic u e 32 and
33, whe eas p e iously he wa e eloci y dis ibu ion along he a ge
wall is displayed.

68
Figu e 32: Veloci y dis ibu ion in he a ge wall, inal con igu a ion.
Figu e 33: Veloci y dis ibu ion along a ho izon al line placed in he a ge
wall. 80 L/min. Maximum speed ~ 5 m/s.
5 m/s
69
As a i s conclusion o his calcula ions, wi h a wa e low o 80
L/min and a dis ance be ween he a ge and he box wall o 20 mm,
he equi ed 5 m/s eloci y o he wa e impac ing he a ge wall can
be eached. The eloci y o he wa e impac ing he co ne o he a ge
is sligh ly lowe han 5 m/s, howe e a highe alue could be eached
by inc easing he wa e low.
These calcula ions a e epea ed wi h a wa e low o 50 L/min and
25 L/min o ha e an idea abou he eloci ies ha would be eached in
he a ge wall o lowe wa e lows.
A 50 L/min he maximum eloci ies a e a ound 3 m/s, and a 25
L/min hey each 1.5 m/s, which is he minimum wa e speed
conside ed o he ope a ion o he es bench.
Figu e 34: Veloci y dis ibu ion along a ho izon al line placed in he a ge
wall. 50 L/min. Maximum speed ~ 3 m/s.
3
m/s
70
Figu e 35: Veloci y dis ibu ion along a ho izon al line placed in he a ge
wall. 25 L/min. Maximum speed ~ 1.5 m/s.
Conclusions
The CFD calcula ions pe o med allowed o de ine he necessa y
a ge dis ance o he a ge box wall (20 mm) in o de o ob ain he
equi ed wa e eloci ies impac ing he a ge . These eloci ies can be
ob ained by a ia ing he wa e low om 25 L/min o 80 L/min.
1.5 m/s
71
4.4. Sys em equi emen s.
The sys em o med by he hyd aulic bench es , objec o his
p ojec , p esen s he ollowing condi ions equi ed o i s co ec
ope a ion:
4.4.1. Powe supply o he sys em.
Fo he design o he powe supply o he es bench i mus be aken
in o accoun ha he powe abso bed by he main de ices is showed in
he Table 6.
Hyd aulic pump
1.5 kW
Hea e
4.5 kW
To al Powe
6 kW
Table 6: Powe supply componen s.
The e a e o he de ices ha wo k wi h elec ic cu en al hough i s
consump ion is negligible in his case. Following he ins uc ions o
ICE S anda d his consump ion is co e ed.
The sys em will be connec ed o a dis ibu ion boa d, which
acco ding o he IEC 62305 ins uc ions, mus ha e a gene al
omnipola sa e y swi ch, as well as sho -ci cui p o ec ion, o e load
78
4.8. Legal scope o he machine.
This ins alla ion is designed and manu ac u ed aking in o accoun
he ollowing s anda ds:
• EN 13480 S anda d. Me allic indus ial piping
• EN 13445: 2002 - Un i ed P essu e Vessels
• BS EN 286: 1991 Simple Un i ed P essu e Vessels
• EN ISO 12100 – Sa e y o machine y – Gene al
p inciples o design – Risk assessmen and isk educ ion
(Eu ope).
• DIRECTIVE 2014/68/EU OF THE EUROPEAN
PARLIAMENT AND OF THE COUNCIL (h p://eu -
lex.eu opa.eu/legal-
con en /EN/TXT/PDF/?u i=CELEX:32014L0068& om=ES)
Rega dless o wha is es ablished in his speci ica ion, all p o isions
on wo king condi ions, sa e y measu es, p oduc ion and manu ac u ing
p ocedu es e c, ully comply wi h he Regula ions, O dinances and
Laws in o ce.

79
80
Chap e 4
5. Conclusions
The wo k o his hesis ocuses on he design o he hyd aulic es
bench o he e osion and co osion es in o de o e alua e he
compa ibili y o a ious ma e ials in he n_TOF a ge egime.
A p esen , he a ge is unde design and a ious ma e ials a e
being in es iga ed o bo h he co e and he con ac pa wi h he
cooling wa e , in o de o subs i u e he p esen a ge which will ha e
eached i s end o li e.
The p ojec o he es bench, om he concep ual o he de ailed
design and manu ac u e he de ice, has been de eloped in
app oxima ely one yea .
The wo k has ca ied ou in collabo a ion wi h se e al g oups
ac oss CERN, in ol ed in he echnical speci ica ion, analysis, design
and manu ac u e.
The main phases o he p ojec can be summa ized in he ollowing
s eps:
81
• Analysis o he echnical speci ica ion and a
consequen ly ca e ul ma e ial choice;
• Concep ual design wi h Au odesk In en o ;
• Nume ical s udy, based on he Fini e Elemen Me hod,
o he implemen a ion o he p oblem in ANSYS;
• Selec ion o all he equipmen used;
• Discussion wi h he supplie s o he bes op ions;
• De ailed design o manu ac u e;
• P epa ing he echnical d awings o manu ac u e;
• In con ac wi h he echnical employees o he p ope
assembly.
The inal p oposal o his p ojec is a hyd aulic es bench based on
s ainless s eel any de ice in con ac wi h he luid o no modi y he
conduc i i y o his. All his moun ed on an easily mo ing aluminium
s uc u e.
The ci cui has a o al olume o 96 L, dis ibu ed in 50 L he ion
exchange , 21 L he chambe , 24 L he expansion ank, 1 L he piping
sys em. All his wi h a maximum p essu e o 4ba due o he es ic ion
o he oxygen senso .
82
A e di e en es s o he se -up, i is possible o say ha e e y hing
wo ks pe ec ly and ha , also, i is possible o ge he da a needed o
ca y ou he necessa y expe imen s success ully.
In he u u e, i could be a good idea o ins all a hea exchange o
con ol he low empe a u es, a his momen he hea e is ins alled and
i is possible o hea up he luid bu no cool down (jus spending ime
wi h he hea e u ned o ). Also, he pump gene a es hea so he luid
has a empe a u e es ablished by he de ice.
83

84
Chap e 5
6. Appendix
6.1. Ha dwa e manual.
This manual is a quick guide o he ins alla ion and use o he Tes
bench ha dwa e o he alida ion o he e osion and co osion o he
new n_TOF a ge . Fo addi ional in o ma ion/doub s he esponsible
pe son o he Tes bench a CERN mus be con ac ed.
6.1.1. Impo an no es
I is impo an o s ess ha he Tes bench is a p o o ype and is on
an ea ly s age o de elopmen .
Du ing ope a ion i does no equi e pe manen igilance o he
ope a o bu a pe iodic check is s ongly ad ised.
The Tes bench was designed o au oma ically send he epo s,
wa nings/ale s o e e-mail o he pe son in cha ge. Also, se e al
in e lock sys ems we e implemen ed in o de o p e en and ale
CERN abou Tes bench inco ec use o mal unc ions.
6.1.2. Tes bench ins alla ion and se up
The Tes bench mus be ins alled in a loca ion wi h access o no mal
ap wa e , sewe o discha ge used wa e , 380V iphasic elec ic
connec ion and E he ne connec ion o in e ne access.
85
Be o e using he Tes bench he ope a o mus pe o m se e al
ope a ions and e i ica ions. These ope a ions a e desc ibed in he nex
lis . De ailed explana ions o all he s eps a e p esen ed in he
subsec ions bellow. I any doub a ises du ing he usage o he Tes
bench, he pe son in cha ge o he Tes bench a CERN mus be
con ac ed.
1. Connec he chambe o he Tes bench.
2. Ins all he a ge pieces inside he chambe .
3. Open he al es V1, V2, V3, V4, V6, V7, VR1, VR2 and VR3
and close V5
4. Connec and/o u n on he cooling ap wa e .
5. Fill he ci cui wi h ap wa e un il 3-4 ba .
6. Connec he powe cable.
7. Tu n on he es bench.
8. Run he p og am.
6.1.3. Connec he chambe
The chambe mus be placed in posi ion on he Tes bench be o e
illing all he hyd aulic ci cui :
1. Place he chambe in he Tes bench acco ding o Figu e 39.
86
2. Sc ew he wo inne connec ions and he ou e connec ion.
Figu e 39: Tes chambe and connec ions.
6.1.4. Ins all he a ge pieces
The a ge s will be ins alled inside he chambe as decided be o e
he expe imen , as shown in Figu e 40.
87
Figu e 40: Schema ic p ocedu e o he es .
Then, he chambe mus be closed p ope ly. You should be su e ha
he O- ings a e p ope ly inse ed and sc ew all he bol s.
6.1.5. Val es
Make su e ha he al es V1, V2, V3, V4, V6, VR1, VR2 and VR3
a e open and he V5 al e is closed.
Figu e 41: Val e 1.
Figu e 42: Val e 2.
94
7. Re e ences
[1] "h p://home.web.ce n.ch/"
[2] C. Le e e, LHC: he guide.
[3] "h ps://home.ce n/abou /expe imen s/alice"
[4] "h ps://home.ce n/abou /expe imen s/a las"
[5] "h ps://home.ce n/abou /expe imen s/cms"
[6] "h ps://home.ce n/abou /expe imen s/lhcb"
[7] “h ps://na62.web.ce n.ch/na62/”
[8] “h p://wwwcompass.ce n.ch/”
[9] “h ps://espace.ce n.ch/hi adma -sps/Wiki%20Pages/Home.aspx”
[10] h ps://home.ce n/abou /accele a o s/an ip o on-decele a o
[11] h ps://n o -exp.web.ce n.ch/n o -exp/
[12] T. Thuillie , “Elec on Cyclo on Resonance Sou ce De elopmen
Lpsc”, G enoble Cedex, 38026, F ance
[13] h ps://home.ce n/abou /accele a o s/low-ene gy-ion- ing
[14] S. Sgobba, L. Ma ques An unes Fe ei a, “Ma e ial Selec ion and
Co osion S udies o a High Ene gy Physics Expe imen : The
Neu on Time-o -Fligh Facili y a CERN,” NACE Con e ence
Pape s: NACE In e na ional, pp. 1-15, 1 Ma ch 2011.

95
[15] “Repo o he Ex e nal Re iew Panel on he Design o he
Upg aded Lead Ta ge o he N-TOF acili y 14.2.2008,” Gene a,
2008.
[16] “h p://www.en ica esys ems.com/demine aliza ion-wa e -
ea men -plan s-pune.h ml”
[17] “h ps://www.elec ical4u.com/joules-law/”
[18] “h p://www.omega.com/ echnical-lea ning/ low-and-le el-
measu emen .h ml”