All-metal joints studies related to the mechanical and irradiation Przemyslaw Lutkiewicz Supervisors: Christian Rathjen, CERN Cedric Garion, CERN Blazej.

All-metal joints studies related to the mechanical and irradiation

Przemyslaw LutkiewiczSupervisors:

Christian Rathjen, CERNCedric Garion, CERNBlazej Skoczen, CUT

Answer the question How leak-tightness is made How all-metal joints behave How irradiation and inelastic deformation influence on

joint behavior (CDM theory extension) Understand the material and geometrical influences on the

sealing properties Develop tools and methods helpful in the joint design process Apply that knowledge to the new design

Answer the question How leak-tightness is made How all-metal joints behave How irradiation and inelastic deformation influence on

joint behavior (CDM theory extension) Understand the material and geometrical influences on the

sealing properties Develop tools and methods helpful in the joint design process Apply that knowledge to the new design

“All metal joints studies related to the mechanical and irradiation load”

Thesis scope

Przemyslaw Lutkiewicz AT/VAC

ConFlat jo int studies (about 4600 units for LH C)

new jo int design for CLIC (about 460000 units)

Continuum D am age M echanic theory approach M echanical load (a lready

w ell know n) Irradiation (m y scientific

input )

ConFlat joint studies (about 4600 units for LHC)

new joint design for CLIC (about 460000 units)

Continuum Damage Mechanic theory approach Mechanical load (already

well known) Irradiation (my scientific

input )

PhD thesis subject:

ConFlat joint


CLIC joint pa

rticl

e be

am fl

ow

material under mechanical load and radiation

Mechanical load

1. ConFlat joint studies

Joint consists of:1. “soft” gasket (OFS Cu)2. “hard” flanges with knives (SS 316LN)3. bolts, clamps or chains

ConFlat join setup

ConFlat joint concept

Bolts push the “hard” flange knifes into “soft” gasket

A high contact pressure acts between the knife and gasket surfaces

Plastic flow occurs in the gasket

Gasket material fills gaps and makes joint leak tight

1.

2.

3.


1. CF joint studies

Strong material flow – high deformations (FE calculations and experiments)

High and quasi-constant contact pressure (FE calculations)

Strong localization of high stress (FE calculations)

Locally high plastic strains (FE calculations)

A crack was observed under the knife tip for sharp-edged design (experiments)


Przemyslaw Lutkiewicz AT/VAC ; [email protected]

Experimental Material properties definition

Damage tensor results for anisotropic material

Damage tensor results for isotropic material

Proceeding with the publication

CF joint anisotropic damage analysis

Fig. 9. Anisotropic model: a; Distribution of micro-damage (D11, D22, and D33) compared to evolution of plastic strain intensity (EPSE), b;

Evolution of micro-damage components as a function of depth

Fig. 4. Damage parameter D and elastic unloading modulus E as a function of strain at 4.2 K

1. CF joint studies

http://en.wikipedia.org/wiki/Image:CERN_logo_400x400.gif

1. CF joint studies

Two stage behavior Stage I – “elastic” Stage II – “plastic”

Two stage behavior compares well with bilinear gasket material model

Good correlation between experiments and FEM

Sealing force evolution


FE calculations experiments

1. CF joint studies – material influences

Sealing force evolution

Material properties strongly influence the joint behavior

Results were obtained using Finite Element modeling (ANSYS)

Based on those results the phenomenological formula was defined (2008, JVST A): Describes evolution of

displacement with force Depends on elastic-

plastic gasket material properties E , 0, H


1. CF joint studies – experiments


Locating pins

Experiment setup

1. CF joint studies – experiments

Leak rate evolution Tightness during compression


Compression tests

1. CF joint studies – resume

Different material properties for gasket material in different directions (anisotropy)

High plastic strain concentration indicate macro crack propagation possibility

Tightness is done on the lower “knife” slope, second “knife” slope is not at the contact with the gasket

Joint behavior is related strictly to elastic-plastic gasket material properties and can be described by phenomenological function

Joint is leak tight when gasket is plastifed through the thickness

ConFlat joint is leak tight for about 0.1-0.15 mm imprint



1. Reliable joint - high sealing performance

2. No plastic deformations on flanges

3. Smooth flange-gasket-flange transition ( no gaps)

4. Simple shapes and preferably symmetrical joint

type

5. Easy in assembly

6. Cheep in production and low sensitivity on

machining tolerances

2. CLIC joint


SLAC X-band joint Stainless steal WR90 flange

(male and female type) 31.9 x 19.2 mm copper

gasket, 4.42 mm width and 2.03 mm high

CLIC gasket vs 2 CHF coin

Joint conceptDrawbacks (things to be improved): Large and non constant

displacements into gasket aperture (0.25 mm = ¼ of the gasket thickness)

Different contact pressure distribution for upper and lower surface of the gasket

Trapped volume and grove weakness

0.25

0.25Lateral profile of the maximum inner displacementsContact pressure between upper gasket surface and flange

Contact pressure between lower gasket surface and flange

No sealing

Sealing

No sealing

Sealing

Upper surface

Contact pressure between upper gasket surface and flange

No sealing

SealingTrapped volume

Contact pressure near grove

No sealingSealingNo sealing

No sealingSealingSealing

2. CLIC joint

Rectangular in cross-section OFS copper gasket

Symmetrical, “knife” based design Conical side – higher slope Flat side – possible low slop

Initial gasket position back from flange-flange aperture face

SS 316LN flanges with 8 screws and two locating pins

Symmetry

Existing design concept


gasket

flange

gasket

flange

flange

New CLIC design concept

2. CLIC joint

Homogeneous contact pressure – the joint is symmetric type (sexless)

Homogeneous contact pressure – the joint is symmetric type (sexless)


Cheaper production – simple and easy to machine gasket shape and one type of flange

Easier assembly – symmetric, self-placed gasket with additional pins centering joint system

Superior RF properties – smooth flange-gasket-flange transition

SYMMETRYNO SYMMETRY

New proposalOld design

2)

3)

1)

1)

1)

2)

1) Female flange2) Gasket with moderate shape3) Male flange

1) One type of flange2) Gasket with simple shape

self-placed gasket self-placed

gasket +

Pins centeringjoint system

0.25 mm

0.15 mm

Flange

Flange

Gasket

> 0.1 mm

Flange

Flange

Gasket

2. CLIC joint


R

P

w

AQ 306.0exp3400

2

21

111

QQQT

Theoretical leak rate

Total leak rate

Collar interaction

Thickness [mm]

Imprint depth [mm]

Alpha [deg]

Beta [deg]

Initial position

[mm]

Initial values Yes 2.0 0.3 30 5 -0.35

Final values No 2.5 0.3 35 5 -0.1

Alpha (Q1)Beta (Q2)

Gasket thickness

Initial position

2. CLIC joint

1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5 3.751.0E-16

1.0E-15

1.0E-14

1.0E-13

1.0E-12

1.0E-11

1.0E-10

1.0E-09

1.0E-08

Gasket thickness [mm]

Leak

rate

[mba

r l/s

]


Q1

Q2 collar interaction

no collar interaction

2. CLIC joint

The displacement into aperture as an input for future RF calculations


Displacement into the aperture

2. CLIC joint


Pressure distribution through the gasket circumferance

2. CLIC joint


The leak tightness test: Reference joint parameters:

1.4e-10 mbars L/ s ~60% to 100% of clumping disp. Confirmed 4 times

Finally defined joint parameters 5.6e-11 mbars L/s ~60-115% of clumping disp. Confirmed 3 times

The leak tightness level

Joint at the experiment

Displacement into aperture: Reference joint parameters:

0.08 mm @ 0.30 mm Flange’s knifes plastically deformed

Finally defined joint parameters 0.031mm @ 0.33 mm Flange - no plastic deformation

2. CLIC joint

Irradiation types:

Light charged radiation• Beta particles• Electrons

heavy charged radiation• Ions• Alfa• Protons

Light neutral radiation• Photons (Gamma/X-ray)

Heavy neutral radiation• Neutrons

3. New CDM theory (irradiation)

Irradiation effects:

Impurity production Ionization Heating Atom displacements – defects

creation


The irradiation level [dpa] can be calculate in two ways:

From NIEL using Norgett-Robinson-Torrens equation

NIEL from Monte Carlo codes like FLUKA, SRIM

Non Ionizing Energy Loss profile

dNRT

E

NIELN

2

8.0

V

Ndpa

at

NRT

2216.1~1cm

nEdpa



The irradiation level suspect at LHC project at most critical region (near the collimators)

If the all NIEL energy is consider to contribute into dpa production -> ~8 dpa

First calculations with FLUKA code -> 1e-4 up to 1e-3 dpa

Results based on the neutron flux value -> 8e-4

FLUKA calculations with the synchrotron light contribution -> ??? [dpa]


1 MeV neutron equivalent flow.


Irradiation effects:

Yield point increasing

Uniform elongation decreasing

Plastic flow instability

Tensile tests results for different irradiation dose [dpa]



NdGbMH Ty

Orowan mechanism

Yield point increasing, irradiation hardening H


Irradiation defects

Hardening as a [dpa] dose function


Offset in the material properties at true stress- true strain space



Representative Volume Element

As

YpD ,~ s

D

A

AD

Damage parameter definition

where

D

1~

)1(~

DEE

crDDfor

Damage influences

Effective stress

Effective Elastic modulus

Macro - Crack propagation

Mechanical load

AD

Elastic modules changes for OFS copper



parti

cle

beam

flow

material under mechanical load and radiation

Mechanical load



Irradiation influence on the material structure

SRIM calculation (Monte-Carlo method)

micro-voids cluster formation

Micro-voids cluster



Classical damage evolution

Irradiation damage – cluster nucleation and growth made by irradiation

Post-irradiation damage – cluster growth made by mechanical load

The assumption is made that:

)(dpafDr

,,, pDdpaDYpDD rrm

D

s

ppHpS

YD

eq

m

r

r pD

D

2

3exp57.0


Mechanical damage for irradiated CF gasket

Post-irradiation damage for irradiated CF gasket

Results pure mechanical

damage – 0.48 Irradiation damage

– 0.06 Post-irradiation

damage - 0.19



Damage results pure mechanical

load is weakly influenced by irradiation

Irradiation damage is critical around and over cluster saturation [dpa] value

Post-irradiation damage can be about two times higher than initial irradiation ones

Post-irradiation damage evolution for different [dpa] values



Post irradiation damage evolution theories:

Based on the cluster diameter growth mechanism [1,2]

Based on the damage parameter evolution [3,4]

Results differences for different approaches


Different irradiation types can be unified by [dpa]

Even small [dpa] dose influence on the material properties strongly (change the components mechanical behavior)

Radiation increase the elastic limit (good) but also make the material more brittle (bad)

Radiation and plastic deformation are the source of micro-voids (decrease the mechanical properties and can be filled with gas – source of virtual leak)

Irradiation and post irradiation damage contribution is critical for around and over the saturation [dpa] value



With width like old design

With width like 1/3 of old design THE END

Thanks for Your attention


All-metal joints studies related to the mechanical and irradiation Przemyslaw Lutkiewicz Supervisors: Christian Rathjen, CERN Cedric Garion, CERN Blazej.

Documents

flange gasket slide

gasket circumferance

gasket shape

gasket aperture

aperture slide

upper gasket surface

collar interaction slide

lower gasket surface