SEISMIC TRANSIENT WITH LOW FREQUENCIES AT SHORT DISTANCES

CASE HISTORY: TUNNEL BLASTING IN URBAN SITE AT ISOLA LIRI (ITALY)

Roberto FoIchi (Mining Engineer, Consultant, Rome, Italy)

A b s t r a c t

This article deals with the excavatian of a tunnel below the town of Isola Liri, near Fmsinone, in centra[ Italy .

The tunnel was driven into a conglomerate formation, of lacustnne-fluvial origins, consisting in carbonate clasts with a sandy-siity matrix and calcitic cement. "Dnlling and blasting" had to be adopted in place of a tunnel boring machine which had showed uneconomica[ performances. Good advance speed rates were achieved notwithstanding the presence of some above structures to safeguard. The nearest strutture (an otd vouve chapel) was less than 10 meters (33 ft.) far from the tunnel crown; fhe nearest residential stnicture was at about 15 m (49 ft.).

Due to the dynarnic characterisfics of the propagation medium, unusually low predominant frequencies of tt-ie seismic transient were recorded at distances ranging from ?O to 90 m (33 to 295 ft.). Particular attention had therefore to be taken for the adoption of the peak ground velocity safety limit.

Continuous monitoring was perfamied while excavating. Measured data were computed with a two independent variables' power regression: "v MAX" (peak ground velocity) as dependent, "R" (distante) and "Q MAX" (charge weight) independent variables. FFT was also calculated. Results are exposed and diccussed.

INTRODUCTION

The tunnel is part of an interceptor which fundon is to link, in case of overilow, the upper and the lower section of the Lin dver, by-passing Isola Lin. This town was, in fact, frequently flooded dunng winter seasons.

- - - - - - - - - - - - W - -

Tables, figures and photos are following the t&

blast induced ground vibrations beiow prefied threshold values.

In lack of specific nationat regulations if was referred to the DIN 4150, part l l l . The following threshold values were so initially adapted:

residential buildings: 20 mmls (0.79 inls); chapel: 10 mmls (0.39 inls).

A peak ground velocity decay law, characferistic of the site (geologica1 forrnations, blast design, explosive, seismic path length and above ground marphology, etc.), was not known. Reference was so made to a conservative "general purpose" decay law. This was computed with data measured many tunnel blasting. Taking as reference the 95% confidente limit curve it was adopted a safety scaled distance "SDsti equa1 to 40 mtkg0.5 for the chapel and 27.7 r n ~ k g O - ~ for the residential structures. The maximum explosive charge blasted per delay was than improved by the factor IIEF, where EF is the ratio between the weight strength of TUTAGEX, a water gel from ITALESPLOSIVI (table 1) utilized for the tunnel excavation, and that of the "reference" explosive in the "general purpose" decay law.

After the first shots it was clear that the values of the predominant frequency of the seismic transient induced by the blast, was lower than expected. The threshold peak ground velocity limit for residential stniciures was so reduced to 15 mmls (0.6 in/s).

All the blasts were monitored and the safety scaled distance was gradually reduced. The last rounds were blasted with a SDs = 13.3 m/kgO-63.

RESULTS OF B U S T VIBRAT ION MONITORING

Blast vibration were monitored by means of digital seismographs VMS 500 (by THOMAS INSTRUMENTS INC. - USA). Vedical, horizontal longitudinal and horizontal transversal cornponents of the ground particle velocity were recorded.

Geaphones were placed inside the nearest stnictures, in the corner of the curtain walls. For the first hundred shots were recorded both seismic data (wave forms, ...), distance and charging data in order to compute a decay law. For the following shots were only recorded the seismic data to verify that the threshold value was not exceeded.

The recorded data (table 2, figure from 3 to 7) were processed in a power regression with velocity "v dependent variable, "R" distance, or seismic path length, and "Q MAX" weight of the explosive charge which has induced the peak partide value "v MAX", as independent variables (table 3, figure 7).

The high frequencies, characteristics of the blast at such short distances (in massive limestone the predominant frequency generally overpass the 150 + 250 Hz) were heavily filtered by the conglomerate, with energy transfer to the lower frequency components of the wave train.

COMPLAINS

Complain carne just fram one person. His house was at a minimum distance of l00 m (328 ft.) fmm the tunnel but "disease to the house stmcture showed by honzonfal and

Figure i. Pianimetry of the tunnel and trace of the nearest above standing structures io be'..*<- safeguarded (numbered from l to 9).

Photo 2 Structure n.2 (votive chapel).

Photo 1 Structure n ?

Photo 4 Strutture n. 5

Photo 3 Struct~!re r! 1

Photo 5 5itri~ctur.e r l . 9

Photo 6 iiaciure between ihe h o adjoining bodies in ihe house of ihe propie that cornplained for damages.

Figure 2. Geologica1 section of the tunnel (vertical yellow conglomerate of lacustrine-fluvial origins "P"

fongitudinal). The (form Pleistocene

forrnation consisted io recent times).

Table 1 . Characteristics of TUTAGEX 1 1 0 water gel explosive.

Density

Specific energy per unit mass

Detonation velocity

Weight strength

l l SO @/m3 2.79 MJ/kg

4,000 mls

85%

Photo 7 & 8 Coiiylornerate in intich afler the btast

Table 2. Data recorded during tunnel blasting.

VV vT vR W ff n* (mmls) (mmls) (mmls) (Hz) (Hz)

fR v VECT (Hz) ImWs)

Table 2 Data recorded during tunnel blasting.

vv VT VR fv fr no (mmls) {mmls) (mmls) (Hz) (Hz)

f R vVECT (Hz) (mmh)

Table 2 Data recorded during tunnel blasting.

vv VT VR fv n no (mmk) (mmls) (mmls) (Hz) (Hz)

st dev 5.72 3.09 3.59 30.26 67.14 mean 7.16 3.78 4.44 75.20 59.14 min 1.56 0.66 0.78 9-00 16.00 rnax 51.23 30.02 31.47 128.00 512.00

fR V VECT (Hz) (mmls)

Table 3. Resulf of statistica1 computations of the data recorded during tunnel blasting.

Standard errar of the rneasured value "s" (mmis)

Correlation coefficient "r"

Number of measures

Degree of freedom

R, Q MAX coefficients Of R, 'MAX

coefficients Equation of decay law (50% probability) Equation of decay law (95% probability)

C (Q exponent in SD)

v VECT

1.28

0.92

105

102

-1.45, 0.92

0.06, 0.07

393DS*'+45

648 DS-1 -45

0.63

VV

1.29

0.93

105

1 02

-1.52, 1.14

0.06, 0.07

31.1 SV1-52

490 DS-1.52

0.75

vT

l .42

0.79

105

1 02

-A.i6, 0.53

0.09, 0.1 0

1.13DS-1.16

7 88 DSi -1 -16

0.46

vR

1.44

0.82

105

102

-1.31, 0.45

0.09, 0.1 1

252DS-1-31

470 DS -1 3'

0.34

..: , .. , . . > - , i . * : . 1

. .... -L

. . 1-1- . , . 2 ... L < . * = : i ; 7 . - b k h - ! r - - , , v-'- i.;. .*=---:,--.4- . ;

- .. :L?:. . . . . T L

$ - C ' - :- > C .L . . --n --

f:6*hu.i,rbx i ' ?I!l?%c,,<:

--. ; "" ! ,,:i F- 5 .S. - q; fi-E-T-~%-;=:+r+-;;-T- . < I, . - . ., . v+ i

: F r e q 21 YI . ' .- L MIiInr ,

PW : .>!bd;<; , . , , Ij 25 n; p- " "' :. .. 'm+=-*.. ... - . .. 8

Tjv: 1? R7 , . . Figure 3. Wave form recorded

.- -. F O U B I E R B H A L Y G I S

I I , 8 , I I I L I L ! I

I

1 B.503 ! i u i l I Figure 4. ... and Fourier analysis of : E " 1 ,

I r i r e . a s i the vertical, horizontal transversal L i I ! i

; O - ' :j !, i- ! L ;',I! ;; :

and horizontal radial cornponents cif : C ..

3 : . , 1 1 J ? . . . m .> , I "'" p the giound paiticle velocity. , , ' ! ; q !! E .. ,.-,' i; ' 1 j .l , - .. .> :I ;I ! ; L g.iiie

I !ii{ ilj liliiji I j / $ 1.1 , 1 , t i ti \ \ / I I : I " !

; H Ì [i! I . L r 0.im i l , i ;i q ( , . A t.

l p, ,/:J L . ' v Y -7 T.-: i-iv-b-<!kkxi

I I l I I O.m j ! i 28 40 68 h i00 i F r e q u t n c g Hz !

I

I I I I I , l , I

i

: i C j C 8.3M - : I - . . - ' !

i -7 Ba23J i r t i

: i! ; , - : . " i ! : A!: - v i { , ;,

, , - - t 0 , im i

, ; ' + - - L

f i . .;, ! ' i * * I . , .> $ >-,-,.-. '.-,<..<>l I

1 . , l I Ì e.w

I 28 4: de BB :m I F p ~ p u e n c g Hz

vV MAX - 1'4 vT - fT

Figure 5. Peak values of the vertical (vV M U ) . horizontal transversal (vT MAX) and horizontal radiai (vi3 MA>(). componenis of the ground particle velocity and their predominant frequencies (fV, 'T, fR)

N - R

m IsO

m

14.3

m m ... IW

I*)

$*o

im t OO

m

W

40

20

O . . . . . . O m m m 1 m 1 2 0 1 . 1 1

[m)

f r - R fR - R m m

wa m 260 I 4

2IP 140

m 110

m - XD

1m I W

la E bm 1- t48

!m 1 1 0

*W 1m

110 M

m M

W .D

m M

P , . . . . . . . . . . . . . . , o O M m m i w 1 m I M j o E .D Q M 1m (1im l*

fl h) R l-)

Figure 6. Predominant frquency o i the peak vaiues of the vertical (fV), horizontal transversal (fi) and horizontal radial (m) components of the particle velocity, measured at various distancas (R) from the shot.

vV MAX ~ ~ ( ~ f ~ 0 . 7 5 )

VT MAX S D ( R / Q ~ + ~ ~ ) VR MAX SD(RIQ'.~~)

Figure 7. Decay laws of peak particle velocity (50% and 95% proba bility curves).

PROCEEDINGS OF THE

NINETEENTH ANNUAL CONFERENCE ON

EXPLOSIVES AND BLASTING TECHNIQUE

JANUARY 31 - FEBRUARY 4,1993

SAN DIEGO, CALIFORNIA, USA

Published by: International Society of Exptosives Engineers

29100 Aurora Road Cleveland, Ohio, USA, 44139-1800

(216) 3494004

Copyrig ht, I nternational SocIety of Explosives Eng ineers, l 993

All rights reserved. No part of this publication may be reproduced, stored in a computer, or transmitted in any form without perrnlssion in writing from the International Society of Explosives Engineerc.