interface stability of granular filter structures under …...interface stability of granular filter...

10
Interface stability of granular filter structures under currents Henk Verheij (Deltares) Co-authors: Gijs Hoffmans, Kees Dorst, Stefan Vandesande August, 2012 ICSE-6 Paris 2012

Upload: others

Post on 21-Jan-2020

16 views

Category:

Documents


0 download

TRANSCRIPT

Interface stability of granular filter

structures under currents

Henk Verheij (Deltares)

Co-authors: Gijs Hoffmans, Kees Dorst, Stefan Vandesande

August, 2012

ICSE-6 Paris 2012

Granular filters

ICSE-6 Paris 2012

df15/ db85 < 4

(Terzaghi/Peck 1948)

• Geometrically closed (sand-tight) filters

• Stable Geometrically-open (sand-tight) filters

85

15 85

0.161

f f fF

f f b b

n dD

d n d

Wörman

Geometrically closed filter structure

Stable geometrically open filter structure

Unstable geometrically open filter structure

1989

1989 Klein-Breteler

1994 Bakker-Konter

mechanisms

ICSE-6 Paris 2012

Shear failure Winnowing Edge failure

Shields (1936) Simultaneous

movement

Filter material

moves first

Base material

moves first

50 50f bd d

15

ff

Dd

Principle idea behind the new formula

New formula

ICSE-6 Paris 2012

50 ,

15 50 ,

1ln

1

f f c f fFd

f b b c b b

d VD

d d V

expf

b d

k z z

k L

Hoffmans (2012)

(Klar, 2005)

50

15 50

lnf f

d

f b

D d

d d

Laboratory experiments

ICSE-6 Paris 2012

Test

db50

base

[m]

df50 filter

[mm]

df50 / db50

[-]

DF / df15

[-]

Thickness

DF [mm]

remark

T-01 309 7.1 23.0 3.29 20

T-02 309 21.3 68.9 1.44 27

T-03 309 21.3 68.9 3.29 61.5

T-04 633 21.3 33.6 1.44 27

T-05 309 15.1 49.0 0.57 8

T-06a 309 15.1 49.0 3.01 40

T-06b 309 15.1 49.0 3.01 40 High

turbulence due

to sill

T-06c 309 15.1 49.0 3.01 40 High

turbulence due

to piers

T-07 309 15.1 49.0 4.13 57

results

ICSE-6 Paris 2012

0 50 100 150 200 250 300 350 400 4500

2

4

6

8

10

12

1

2

3

4

5

6

7

Geometrically closed filters

d =1.2 (value proposed by Hoffmans)

d =0.86 (upper limit)

d =0.7 (90% lower)

d =0.28(lower limit)

df50

/db50

[-]

Df/d

f50 [

-]

van de Sande [2012], filter material eroded before base material (with test number)

van de Sande [2012], simultanious erosion (with test number)

van de Sande [2012], base material eroded before filter material (with test number)

Wörman [1989]

Dixen et al. [2008]

Other tests, filter material eroded before base material

Other tests, simultaneous erosion

Other tests, base material eroded before filter material

conclusions

ICSE-6 Paris 2012

1. The relative layer thickness fits better with df50

than with df15

2. Simplified design equation:

3. The alpha value proposed by Hoffmans [2012] is

too high (αd = 1.5)

• Deterministic approach;

• αd = 0.82 (safe upper limit)

• αd = 0.69 (90% confidence bound)

• Probabilistic approach;

• µ(αd )= 0.47

• σ(αd )= 0.04

50

15 50

lnf f

d

f b

D d

d d

recommendations

ICSE-6 Paris 2012

Extra tests:

• Damping in the filter

• Infuence of Turbulence on the αd value

Large stones – no extra turbulence Large stones – with extra turbulence

Thank you for your attention

More information

Email: [email protected]

ICSE-6 Paris 2012

Granular filters

Relevant equations for designing geometrical open filters

1) Relative strength

2) Relative load

3) Damping length

4) Critical phase

If

and combining Eq’s 1, 2, 3 and 4 then

and c Fz D , ,50

50 ,

c bf c bb bc

c f f c f

gd

gd

exp

f

b d

k z z

k L

15 with 1.5d d f dL d

and c Fz D

, , 50 15, and / 1.2f b c f c b f fd d

50

50 50

1.2lnfF

f b

dD

d d