How the ground fails

Hydraulically caused failure according to EN in coarse-grained soil


1 Publications of the Basic Construction Institute of the Technical University of Berlin, issue no. 58, Berlin 2011, S Lecture on the 7th Hans Lorenz Symposium on hydraulically caused failure according to EN in coarse-grained soil Hon.Prof. Dipl.-Ing. Dr.techn. Lothar Martak well master and sworn and court-certified expert for foundation engineering, well construction and tunnel construction Dipl.-Ing. Dr. techn. Robert Hofmann engineering consultant for civil engineering, lecturer for geotechnical engineering FH Campus Wien, sworn and court-certified expert for foundation engineering and soil mechanics Dipl.-Ing. Petra Drucker Institute for Geotechnics, Vienna University of Technology Summary The fundamentals of the calculation models for buoyancy and hydraulic ground failure have not changed in any way with the introduction of the EN. The partial safety considerations recommended in the Eurocode EN remain in the wealth of experience, but at the lower level of the past decades (Terzaghi Peck, Davidenkoff Baumgart etc.). The calculation methods have also remained the same, i.e. the levying against the uplifting of a structure (buoyancy criterion) and against hydraulic ground failure, for which the levying with effective (effective) stresses and with total stresses are on an equal footing, although they lead to different results. New in the norms, but not always in daily geotechnical practice, is the introduction of erosion or suffusion ice in order to avoid hydraulic gradients that do not meet the physical requirements of the respective loose rock subsoil. The type of post-ice remains free in EN and is the responsibility of the planning geotechnician, as has already been and continues to apply to the determination of the hydraulically decisive seepage to determine the hydraulic gradient specified in the calculation.

2 182 L. Martak, R. Hofmann, P. Drucker 1 Theoretical Basics 1.1 General The EN deals with the section post-ice procedures and partial safety hazards when floating up (UPL) and in the section post-ice procedures and partial safety hazards with hydraulic bed failure (HYD). In addition to failure due to floating and hydraulic heave, the EN also knows failure due to internal erosion and failure due to piping. 1.2 Hydraulically caused failure By flowing seepage water, pressure and shear stresses are transferred to the grain structure. These tensions and the resulting (destabilizing) flow forces S dst = i [kn / m³] (1) are greater, the higher the hydraulic gradient i. The hydraulic gradient is calculated as follows: Pressure head difference i = = length through which the flow passes ΔL (2) The flow forces lead to the loosening of grains from the grain structure, especially in the area of ​​the air-side sources. As a result of the successively decreasing shear strength, the grain structure gradually loses its bond, which is referred to as a hydraulic fracture. Essentially, the following possibilities exist for the formation of erosion channels through hydrodynamic soil deformations in the subsoil: Hydraulic ground fracture Hydraulic breakthrough (piping) Contact erosion (retrograde erosion) Suffosion 1.3 Principle of total and effective stresses The re-ice required in EN to prevent hydraulic failure is the same. The constant stabilizing effects and the decisive destabilizing effects from own weight, buoyancy, pore water pressure and / or flow pressure are safeguarded by differently sized partial safety factors.

3 Hydraulically caused failure according to EN As explained in EN Section 10.1, Note 5, the conditions for a hydraulically caused failure of the soil can be formulated in total stresses and pore water pressure or in effective (effective) stresses and hydraulic gradient (see Figure 1). Total stresses σ r are made up of the force of the solid mass of the ground under buoyancy and the force of the water above the horizon under consideration, as well as external loads, if at all: σ r = σ + u (3) The pore water pressure u results from the hydrostatic pressure head in the horizon under consideration: u = (4) h Effective stresses σ 'are effective stresses which prevail in the horizon under consideration of the grain structure of a partially or completely water-saturated soil: σ = σ ru (5) z G tot = r ΔV G eff =' ΔV G s = (1-n) s ΔV R s = ΔV R s = ΔV R = R s G = (1-n) ΔV G = n ΔV total stress = effective stress + pore water pressure σ r = σ` + urz = z + z with: weight under buoyancy: = (1 n) (s) r weight of the water-saturated soil: r = + weight of water s weight of solid mass n pore volume = (1 n) + ns Figure 1: Possibilities to write down the forces that act on a particle in the water-saturated soil at depth z. EN specifies in Section 10.1, Note 5 that in the case of re-ironing against failure by floating up, the approach with total tensions is perished. The safety beds according to A.4 (1) and A.4 (2) are used to examine the equilibrium ice according to the buoyancy model (UPL), if necessary with the addition of an additional restrained excavation rubble.

4 184 L. Martak, R. Hofmann, P. Drucker For the investigation of the safety against exceeding the hydraulic gradient (HYD), according to EN, section both the assumption of total and effective stresses, with the assumption of the partial safety level according to A. 5 (1). However, these two admissible proofs lead to different dimensioning results, as should be shown as an example in Section 2 of this article. It should be noted that the post-ice guidance against hydraulic bottom failure has so far been carried out according to the principle of effective stresses, although the global safety concerns are coordinated with this (see Tables 1 and 2). Table 1: Compilation of the global safety coefficients that have been customary up to now. Soil gravel, medium density EAB 1994 loose sand, silt / clay, oak clay, at least stiff gravel, medium density Terzaghi / Peck loose sand, silt / clay, calibrated η global 1.5 2.0 1.5 1.5 2.0 2.0 3.0 The specified global safeguards can be represented in the partial safety concept as follows (see Table 2): G; dst η global = (6) G; Table 2: Compilation of partial safety zones for impacts and loads. Table A.5 EN DIN 1054: or EAB 6.1, Annex A6 LF 1 LF 2 LF 3 SIA 267 G; dst 1.35 1.35 1.8 *) 1.3 1.6 *) 1.2 1, 35 *) 1.4 1.6 **) G; 0.9 0.9 0.9 0.95 0.9 G; dst / G; 1.5 1.5 2.0 1.44 1.78 1.33 1.5 1.56 1.78 Comments on table 2: *) DIN 1054: lower value for G; dst with a favorable substrate. **) SN SIA 267 G; dst = 1.6 for silty and fine-sand soils, G; dst = 1.4 for all other soils at risk of breaking ground. After these investigations, it is up to the planner to determine an overall safety of G; dst 1.35 η global = = = 1.50 according to EN (or ÖNORM B) as sufficient 0.9 G; or when the flow force (effective tensions) is applied, it is necessary to use higher safety factors. The associated increase in the embedment depth of the excavation shoring also means a reduction in the hydraulic gradient, which is inevitable in the case of soils at risk of erosion.

5 Hydraulically caused failure according to EN For investigations against hydraulically caused failure, e.g. on the bottom of the excavation pit, it must be decided in advance whether a failure due to the bottom floating up or a hydraulic ground failure will be decisive. The decision criteria for this are the hydraulic gradient, the permeability range k f, the grain distribution and the cohesion of the soil. 1.4 The hydraulic gradient This article does not deal with the effects of the construction pit or excavation geometry, in the sense of the known narrow or wide construction pits, the receding or receding corners, etc., since the illustrated evidence relates to the coverage of the required equilibrium between the Soil, its stratification, water permeability and the ground water level inside and outside the excavation. For this, however, the shortest conceivable infiltration regimen along the construction pit enclosure and thus the greatest conceivable hydraulic gradient is assumed, as in most cases of construction practice an after-ice on the safe side equates to. It is important to note that the actual reduction of the flow pressure along the shortest seepage around the foot of a terrain jump protection does not take place linearly, ie the flow fields determined by means of numerical analysis in Figure 2 illustrate. According to the potential line distribution, the hydraulic gradient changes along the flow path (see Fig. 3) Fig. 2: Potential line networks at = 12 m. Left: embedment depth t = 8 m, global security η = 1.5. Right: t = 6 m, η = 1.0. Figure 3 shows the change in the hydraulic gradient at a groundwater overpressure height of 12 m along the shortest seepage L around the edge of the building pit. In the case of a 6 m deep embedment (η = 1.0), the flow pressure reduction determined on the terrain side describes an i min ~ 0.18, on the excavation side an i bottom ~ 0.72 and below the wall foot of i max ~ 8, 0. If no hydraulic soil deformations occur with these flow pressure conditions

6 186 L. Martak, R. Hofmann, P. Drucker, the soil at the excavation level next to the wall must be able to withstand a level of ~ 0.72 suffusion-proof. If the Sickereg e.g. extended by 4 m (embedment depth 8 m, or η = 1.5), the hydraulic gradient in the bottom area of ​​the excavation is ~ 0.63, which can be quite stable for uniform gravel / sands and sands (Bažant, Terzaghi, Harza and Müller-Kirchenbauer). Fig. 3: Course of the actual hydraulic gradient i from the potential networks of Fig. 2 along the shortest seepage around the terrain jump protection. It is not the hydraulic gradient applied linearly over the entire shortest infiltration regimen that is decisive for the safety against hydraulic ground failure, but the value immediately below the excavation level must be taken into account. Its admissibility results from the post-ice of the suffusion security of the existing soil. 1.5 Example for the determination of safety according to effective and total stresses The arithmetical equilibrium by applying effective or total stresses is only possible with partial safety values ​​G, dst = 1.00 and G, = 1.00 (or with a global safety level η = 1 , 00) numerically identical. For all other sizes of the partial safety areas, ie they ex. can be found in the appendix of the EN, there are serious differences. The example shown in Fig. 4 shows, based on the ascertained safeguards against the formation of a comparatively ice-impermeable silt layer on the excavation base, that a total safety of η tot = 1.21 corresponds to an effective safety of approx. Η eff = 1.5. (Note: the application of total stresses is possible at k f 10-5 m / s, while for the application of effective stresses a flow and thus a permeable subsurface is assumed.)

7 Hydraulically caused failure according to EN Water level difference: = 12 m = 9.81 kn / m³ sand, medium-tightly stored: '1 = 11 kn / m³ k f1 = 10-4 m / s clayey silt:' 2 = 10.5 kn / m³ k f2 = 10-6 m / sd = 16.8 m Fig. 4: Situation of the post-ice excavation of the buoyancy safety of a dam layer For the present example, the global buoyancy safety factors (which can be multiplicatively split into stabilizing and destabilizing partial safety factors) are grounded according to the concept of total Stresses and compared with the concept of effective stresses: rd (10.5 + 9.81) 16.8 η A, tot = = = 1.21 (7) (+ d) 9.81 (. 8) d 10 , 5 16.8 η A, eff = = = 1.50 (8) 9.81 12 If one now varies e.g. the embedment depth d (= thickness of the damming layer), it can be seen that the safety level according to the concept of total stresses does not rise above the quotient of the saturation density r = + and the density of the water. The safety level determined by applying the effective forces is, however, directly proportional to the embedment depth d and can therefore theoretically also be grounded infinitely: η eff. Figure 5: Development of the securities according to the total and the effective concept for the soil experts in the example in Figure 4.

8 188 L. Martak, R. Hofmann, P. Drucker As a decision criterion, whether in the case of stratified subsoil the floating of the sealing layer is decisive (i.e. the post-ice must be guided with total stresses and corresponding partial safety beds) or a hydraulic ground failure (and thus the post-ice with effective stresses are recommended and in any case other partial safety factors are to be applied) the ratio of the permeability can be used: kfm / s and k f1 >> k f2 failure due to floating up k f2 ½ k f1 failure due to hydraulic ground failure (Sickereg only in the area of ​​k f1) 1.6 Limitation of the hydraulic gradient to avoid internal erosion According to ÖNORM B, the post-ice of the resistance to hydraulic ground failure must be supplemented by the post-ice of the resistance to internal erosion, suffosion and colmation. Colmation (deposits on the surface or in the pores and cavities, e.g. due to silting up of dewatering measures) can occur, for example, when the flow pressure drops, e.g. caused by low water in the groundwater ladder. If the flow pressure rises again (high water), this can result in changed (shortened!) Flow or seepage. In Section 10.4 (1) P, the EN states that filter criteria must be applied in order to limit the risk of soil being carried away by internal erosion. Which geometrical proofs or filter criteria are used for this, however, is left to the specialist knowledge of the person in charge, as there are numerous models for this (e.g. suffusion ice in the leaflet Application of Grain Filters on Waterways (MAK) of the German Federal Institute for Hydraulic Engineering). A method for assessing the protection against erosion of weirs is the method according to Chugaev (Table 3), published as early as 1965, which specifies critical hydraulic gradients for different types of soil. However, the gradients only apply under the prerequisite of a homogeneous, isotropic soil and an exclusively horizontal flow as well as for dams with predominantly horizontal sealing elements. In addition, there is no security definition in this procedure. Table 3: The critical hydraulic gradient i crit for sands according to various authors [Brandl / Hofmann 2006]. Istomina Brauns Chugaev i krit [-] 0.30 0.40 0.20 0.40 0.12 0.30

9 Hydraulically caused failure according to EN Fig. 6 and Fig. 7 show approximate values ​​for critical hydraulic gradients for assessing the risk of suffusion and contact erosion depending on the grain size. i crit 1.0 i crit 0.6 0.8 loess 0.6 0.4 0.2 sand silt clay gravel 0.5 0.4 0.3 0.2 sand silt mixture SU SU sands 5 GW d [ mm] d U = 60 d 10 0, GE 0.1 0.2 0.3 0.4 0.5 0.6 0.7 d 50 [mm] Figure 6: Assessment of the risk of suffusion in soils with heat according to Istomina (1957 ) [Brandl / Hofmann 2006]. Figure 7: Contact erosion criterion for stratified soil according to Brauns (1985) [Brandl / Hofmann 2006]. d E ective grain diameter The stratification of the subsoil and the associated rapid change in permeability and the ratio of vertical to horizontal permeability increase the risk of erosion. This fact could be taken into account by a further partial safety factor. According to a suggestion by Brandl / Hofmann (2006), the after-ice of adequate protection against erosion is fulfilled if the existing hydraulic gradient i vorh is less than or equal to the dimensioned hydraulic gradient id: id, vorh i krit, kid = (9) i , jsid, vorh i krit, ki, js existing hydraulic gradient id, vorh = ik H characteristic value of the critical hydraulic gradient Partial safety level according to EN or national application document Partial safety level which takes into account the stratification of the subsoil Brandl / Hofmann (2006) have a possible one for flood protection dams Allocation of protection class, dam construction, soil type and seepage to partial safety beds and permissible hydraulic gradients, made on the basis of experience with high water protection dams, various publications, leaflets and guidelines, see Figure 8.

10 190 L. Martak, R. Hofmann, P. Drucker Figure 8: Summary of the critical hydraulic gradient for erosion for different fine-grained soil types A, B, C, D and E according to Brandl / Hofmann (2006). In table 4, the global and partial safety assessments to be applied for the follow-up according to DWA M 507, draft 2007 and DIN 1054 are compiled for comparison. i vorh η and i crit SW i vorh H, ii crit H, Eros (10) Table 4: Safety factors η or partial safety-related H according to DWA M 507, draft hydraulic criterion Global safety concept Partial safety concept Safety factor effects H, i (i prev) Resistances H, Eros (in advance) η LF 2 LF 3 LF 2 LF 3 Contact erosion 1.5 1.20 1.05 1.25 1.25 Suffosion 2.0 1.20 1.05 1.65 1.50 Erosion base 1.5 1.20 1.05 1.25 1.25

11 Hydraulically caused failure according to EN Case study on post-ice against hydraulic failure according to EN Water level difference: = 12 m = 9.81 kn / m³ sand, medium-tight storage: 'k = 11 kn / m³ kf = 10-4 m / s What is required is asked Embedment depth t according to EN to avoid hydraulic ground failure. Fig. 9: Case study of a sheet pile wall completely surrounded by a flow in homogeneous soil. 2.1 Security against hydraulic bottom failure after ice in accordance with EN, equation 2.9 b, with effective forces (flow model): S G (11) dst; d; d Dimensioned of the flow force: S dst, d = dst it A (12) Dimensioned of the excavation base under buoyancy: G, d = t A (13) t A = 1 Area of ​​the considered soil prism according to Terzaghi 2 With the simplified assumption of a for the potential reduction decisive length of also RVS in draft) the hydraulic gradient results to: L = 2 t (see i = L i = (14) 2t And the approximate after-ice is thus: bz. dst dst t A 2t 2t t A ( 15) (16)

12 192 L. Martak, R. Hofmann, P. Drucker With the partial safety factors dst = 1.35 and = 0.9, the effective required embedment depth is calculated: t erf, eff dst 12 1.35 9.81 = = t erf, eff = 8.02 m (17) 2 2 0.9 11 In this example, the hydraulic gradient i is approximated to: 12 i = = = 0.75 2t 2 8.02 (18) The length of the hydraulic seepage ( L = 2 t) is based on the assumption that the hydraulic overpressure compared to the pumped-out excavation is reduced only along the integration of the excavation enclosure, and in a linear manner. With the simplification '= 10.0 kn / m³ and with = ird: dst t erf, eff dst = = Δ h = 2 ti = = 1, 00 (19) 2 2 2t This approach corresponds to the rather cautious post-ice according to Davidenkoff ( quoted by Müller-Kirchenbauer, 1964) for uniform, loose to medium-density sands, with a hydraulic gradient of i = 1.00 (see also Fig. 8). The authors Bažant, Terzaghi and Harza (quoted by Müller-Kirchenbauer, 1964) put Δ h = 2.74 t, 2.83 t and even 3.14 t. According to the authors, this results in permissible hydraulic gradients of i = 1.03, 1.06 and 1.17, depending on the type of soil and the density of the soil. According to DIN 1054 or EAU / EAB, a fictitious hydraulic gradient i r is determined with the aid of the fictitious water overpressure at the foot of the terrain jump protection h r and a previously selected (estimated) embedment depth t ge. For the present example, let t ge = 8.0 m and hr is calculated according to EAU as follows: h 12 = = 5.09 m (20) t 8 r = ge A comparison with the exact determination of the pressure potential at the foot of the terrain jump protection using numerical means Analysis of the flow pattern shows that the approximation formula for hr agrees well with the exact value h = 5.20 m, according to Figure 2 (left). h 5.20 The hydraulic gradient is i vorh = = = 0.65 and the post-ice against hydraulic t 8 ge base failure is fulfilled according to ÖNORM B at an embedment depth of t = 8.0 m:

13 Hydraulically caused failure according to EN S t = i exist t = 0, kn / m (21) 2 dst, k = G = t A = kn / m (22) S, k = dst; k G (23) dst; k 208 1,, <318 Post-ice fulfilled (24) Note: The calculation shows how sensitive the post-ice flow HYD is to the choice of: With = 10.0 kn / m³ the post-ice is also still fulfilled (281 <288). For comparison, the post-ice should be grounded with total stresses according to EN, equation 2.9 a, (model no flow): u dst; d σ (25); d Assuming that the pore water pressure increases as a result of the flow around the construction pit on the earth side to reduce, should in the eventual. Ground example on the construction pit side for the dimensioning of the destabilizing pore water pressure on the underside of the considered prism: u dst, d = (26) dst The dimensioning of the stabilizing total vertical stress on the underside of the considered soil prism on the construction pit side is: σ = t = ( +) t (27), dr Nacheis: (+) t (28) dst With dst = 1, 35 and = 0, 9 results for the total required embedment depth: t erf, tot dst 1.35 9.81 12 = = t erf, tot = 8.49 m (29) (+) 0.9 (11 + 9.81) st The results according to the two possible verifications according to EN are not exactly identical. As explained in section 1.5, there is the approaches of total and effective stresses only with partial safety factors = 1 (and = W) equivalence. dst = With dst = dst η = = 1 it follows for the ggst. Example (Fig. 9): t 12 9.81 9.81 12 = = 5.35 m and t erf, tot = = = 5.66 m (30), 81 erf, eff =

14 194 L. Martak, R. Hofmann, P. Drucker The results according to the two permissible post-ice guides are only completely identical if the approximate simplification '= 10.0 kn / m³ is also taken. Then we get: t = 6 m and t erf, tot = = 6 m (31) erf, eff = 2.2 Post-ice of the security against suffosion etc. The post-ice of the security against suffosion of the sand from the present example (Fig. 9) is shown in Fig. 10 the leaflet Applications of geotextile filters on waterways (MAK). The leaflet stipulates in Annex 1, Sheet 2 under point 2.4 Suffusion safety that soils with constant grain distribution curves and irregularities C U <8 may be regarded as suffusion-proof. If C U> 8, the after-ice of suffusion safety must be grounded. A reduction in the hydraulic gradient simply by increasing the partial safety slope dst from 1.35 or 1.50 to 2.0 or 3.0 can delay a possible suffusion process in the ground. The depletion of fine particles in the soil then takes longer (reduction of the fine grain load), but the risk of hydraulic ground failure usually remains with longer-term exposure. In particular, loose rocks with intermittent grain distributions often lack adequate suffusion security, which is why the post-ice (or the dimensioning of an unbound grain filter) should be shown on such a sample soil in accordance with the German BA for Hydraulic Engineering. Fig. 10: Post-ice of the suffusion safety according to MAK of the German BA for Hydraulic Engineering at an exemplary grading curve with C U = 20 / 0.005 = 4000.

15 Hydraulically caused failure according to EN First, the existing grading curve is separated into base material (d B) and filter material (d F), for example for the separation diameter d T = 2 mm, and with percentages by weight (G BT) of the soil at the selected separation cut d T from G BT = 45% examined. Based on the test example, the base fabric has a degree of nonuniformity d 60 / d 10 of eta C U, I = 0.14 / 0.002 = 70 and for the filter fabric C U, II = 32/9 = 3.56. According to the diagram by Cistin / Ziems (see e.g. MAK of the German BA for Hydraulic Engineering) the permissible A 50 <8 results for the selected example. An existing A 50 = (d 50 filter) / (d 50 Base) = 30 / 0.068 = 441 mathematically estimate. The specified soil for the separation area G BT = 45% would be extremely endangered by suffusion (existing A 50 >> perm. A 50 ~ 8). 5.2 With i prev = = 0.62 (see Figure 2 left) and based on the lower and upper bound 8 of the internal erosion stability according to Brauns for sands from Table 3, this results in the present example: i prev = 0.62> 0.4> 0.2 for i crit. (32) According to the criteria of Braunsär, the example for sand is not erosion-proof, since i is too high and therefore the chosen embedment depth of t = 8.0 m too short. In order to reach the upper bound with i exist <0.4, approximately t would have to be grounded 15.0 m. With the safety factors η or H (partial safety factor according to DWA M 507, draft 2007 according to Table 4), the following results for load case combination 2 with contact erosion: i crit G; dst ηglobal = = H, i (i exist) H, Eros (i exist ) = 1.20 1.25 = 1.50 i exist G; (33) For the post-ice according to Table 4 against the risk of suffusion in dams and long-lasting, high water pressure differences, ηglobal = H, i (i prev) H, Eros (i prev) = 1.20 1.65 = 1.98 (34 ) to be observed.

16 196 L. Martak, R. Hofmann, P. Drucker 3 Summary and Conclusions According to EN, in connection with hydraulically caused failure, appropriate proofs against hydraulic ground failure, other hydraulic failures (internal erosion, suffosion, piping, etc.) as well as against floating are required . The previous global safety factor is split into the stabilizing and destabilizing partial safety factors in accordance with the EN partial safety concept. According to EN, the post-ice against hydraulic ground failure may be grounded by means of both effective and total stresses, but this leads to different results. In accordance with other rules (e.g. DIN 1054), the authors recommend a process in which the effective flow forces are compared to the stabilizing face forces determined with the buoyancy views (application of effective tensions). The hydraulic gradient, which is decisive for the flow force, or the length of the flow through which is available to reduce the prevailing pressure potential difference, must be determined in an expert manner. The post-ice of the subsoil's resistance to erosion or suffusion (other forms of hydraulic failure) is particularly important for long-term unprotected building structures. If this post-ice is not met, the hydraulic gradient must be reduced by deepening the shoring (e.g. according to Table 3) and a geotechnical expert must be called in. In critical cases, a mere increase in the partial safety bed to the size of the former global safety bed for fine-grained soils according to Table 1 is, in the opinion of the authors, not always sufficient for the post-ice of the erosion and suffusion resistance. What effects the possible forms of hydraulic failure can have in practice is shown in Table 5, which compares the required embedment depths of an excavation edge determined for a case study. Table 5: Compilation of the embedment depths t calculated according to EN of a completely flow-around construction pit in a homogeneous soil and with a water pressure difference of = 12 m (case study Fig. 9). Forms of failure Hydraulic ground failure Internal erosion Floating σ eff σ tot (sand) σ tot t = 8.02 m (i = 0.65) t = 7.00 m (i = 0.72) t = 8.49 mt = 15, 00 m (i = 0.40) (t = 6.28 m)

17 Hydraulically caused failure according to EN The hydraulically required embedment depth of a terrain jump protection has thus already remained apart from the excavation shape, the ratio of excavation depth to excavation width, etc., the relevant fine-grained soil (grain distribution, cohesion), the uniformity or non-uniformity of the water permeability (Soil stratification) and the thickness of the horizons located in the seepage reg of the groundwater (partial or full flow). In summary, it should be pointed out again that the choice of the correct soil mechanical or hydraulic model is of particular importance for the post-ice against hydraulically caused failure. Literature Federal Institute for Hydraulic Engineering (BAW) Karlsruhe, leaflet application of grain filters on waterways (MAK). Brandl, H .; Hofmann, R .: Erosion stability and stability of high-water protection dams in torrents, conference proceedings, Siegen Davidenkoff, R .: Inferiority of Stauerken, Wernerverlag, Düsseldorf DIN 1054:: Ground safety dikes in earthworks and foundation engineering; Supplementary regulations to DIN EN EAB recommendations of the excavation work group of the German Geotechnical Society, Verlag Ernst & Sohn, Berlin, ÖNORM EN:, Eurocode 7: Design, calculation and dimensioning in geotechnical engineering, Part 1: General rules. ÖNORM B:: National specifications for ÖNORM EN Martak, L .: Stability considerations for subsidence and sealing, safety concept of Eurocode 7, written contribution to the course Sealing instead of lowering in earthworks, civil engineering and hydraulic engineering, Technical Academy Esslingen Martak, L .; Hofmann, R .: Hydraulically caused failure according to section 10 of EN: 2006 in connection with section 4.10 Hydraulic ground failure (HYD) of ÖNORM B: 2007, introduction of Eurocode 7 part 1, lecture seminar, Vienna RVS guideline open construction T02 urban tunneling Raum, Österreichische Forschungsgesellschaft Strasse Rail Transport in preparation Terzaghi, K .; Peck, R.B .: Soil Mechanics in Building Practice, Springer Verlag Berlin 1961.