Structural analysis of sewer linings by B. Falter (Trenchless Technology Research 11(2), 27–41, 1996)

✍ Scribed by J.C. Boot; J.E. Gumbel

• Publisher: Elsevier Science
• Year: 1997
• Tongue: English
• Weight: 426 KB
• Volume: 12
• Category: Article
• ISSN: 0886-7798
• DOI: 10.1016/s0886-7798(98)00033-9

✦ Synopsis

Structural analysis of sewer linings by B. Falter (Trenchless Techmhgy Research 11(2), 27-41, 1996) Sewer linings have been classijied

as':

(a) Type l-bonded to the host pipe and acting with it as a rigid composite structure (b) Type 2-flexible pipe which is not bonded to, but derives support from, the host pipe and surrounding soil.

Dr Falter's paper refers only to Type 2 linings of nominally circular cross section, and the title of the paper is therefore somewhat misleading. Nevertheless, unbonded flexible linings now dominate the international market, with over 12 000 km of Cured-in-Place-Pipe (CIPP) alone installed worldwide since its introduction in 197 1. New and more rational structural design guidelines are urgently needed for these systems, and Dr Falter's paper therefore addresses an important topic. Unfortunately, however, his approach to the relevant issues is fundamentally flawed in a number of respects.

The synopsis first states that new approaches are required, and yet then proceeds to define the problem in exactly the same terms as applied to conventionally buried pipe. No reference is made in the sequel to the origins of structural deterioration, or the hydraulic and structural functions of an unbonded polymeric lining; in consequence Dr Falter (hereinafter referred to as "the author") appears to be under several misapprehensions in this regard. These issues are. discussed elsewhereI and will be further elucidated in future publications by the writers. Accordingly it is sufficient to comment here that it can be conveniently assumed that the original soil-pipe system will stabilize at a small degree of ovality (usually taken to be in the range 2-5%) provided the lining restores hydraulic integrity and is capable of sustaining the groundwater pressure that must be assumed to build up as a result of lining installation. Thus there is no essential difference between the systems illustrated in author's Figs l(a)-(c) as far as lining design is concerned.

In particular, the conditions defined in the author's State III do not exist in practice. This is because even if the host pipe has deteriorated to become indistinguishable from the surrounding soil, the hole is still stable at the time of lining, and can only be rendered more stable by lining. Thus, the soil-pipe aspects of pipe rehabilitation are much more closely related to tunnel behaviour than that of newly buried pipe.

From Section 3.2 this paper correctly focuses on liner performance subject to groundwater loading; since the linings installed are invariably thin-walled and polymeric, satisfactory structural performance is ensured by providing an adequate safety factor against constrained creep buckling.

In Sections 3.2-4.3 the author considers buckling resistance assuming linear elastic behaviour. It follows from the foregoing that the dominant imperfection in field systems is ovalling as shown in the author's Figs l(b) and (c); when applying ref.2 to untracked systems, it is common to assume a minimum contingency allowance of 2% ovality. However, at no point in his proposed design methodology does the author consider an oval imperfection. Instead he takes a local imperfection (see author's Fig. 2) in the shape of the first eigenfunction (not smallest eigenvalue as stated) of a fixed-ended curved beam, with the span of the imperfection selected so as to minimize buckling load.

Any such imperfections existing in practice (due e.g. to a displaced brick or protruding lateral) are of specified, not variable, dimensions. Furthermore, they are active only over a very short length of pipe, and therefore dissipated by three-dimensional effects. Thus it is not possible to correlate the author's assumed imperfection with those defining the performance of field systems, and which correctly govern design, Certainly assuming x%eigenfunction IOD = x%ovality IOD where IOD = imperfection on diameter(author's ref. 13)3, will lead to gross overdesign. A further illogicality leading to significant overdesign occurs when the author combines imperfections by multiplying the reduction factors associated with the individual imperfections. Thus, in the author's terminology [author's eqn (3)]