IOP Con e ence Se ies: Ma e ials Science and Enginee ing
PAPER • OPEN ACCESS
E ec i e way o econs uc a ch b idges using
conc e e walls and ans e se s ands
To ci e his a icle: Ladisla Klusáek e al 2017 IOP Con . Se .: Ma e . Sci. Eng. 236 012058
View he a icle online o upda es and enhancemen s.
Rela ed con en
A nume ical p ocedu e o ailu e mode
de ec ion o mason y a ches ein o ced
wi h ibe ein o ced polyme ic ma e ials
S Galassi
-
Jobs mason y in LHCb wi h elas ic G id
Jobs
F S agni and Ph Cha pen ie
-
Resonance Con e sion as he E ec i e
Way o T igge ing he178m2H Isome
Ene gy
F. Feodo Ka peshin, M. B.
T zhasko skaya and Zhang Jing-Bo
-
This con en was downloaded om IP add ess 147.229.6.155 on 06/12/2018 a 13:46
1
Con en om his wo k may be used unde he e ms o heC ea i eCommonsA ibu ion 3.0 licence. Any u he dis ibu ion
o his wo k mus main ain a ibu ion o he au ho (s) and he i le o he wo k, jou nal ci a ion and DOI.
Published unde licence by IOP Publishing L d
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
E ec i e way o econs uc a ch b idges using conc e e
walls and ans e se s ands
Ladisla Klusáček
1
, Robin Pěkník
1
and Radim Nečas
1
1
B no Uni e si y o Technology – Facul y o Ci il Enginee ing, Ve eří 95, 602 00,
B no, Czech Republic
E-mail: klusacek[email p o ec ed]b .cz
Abs ac . The e a e mo e han 500 mason y a ch b idges in he Czech Road sys em and abou
2500 in he Czech Railway sys em. Many o hem a e c acked in he longi udinal (span)
di ec ion. The ba el aul s a e sepa a ed by he c acks in o pa ial mason y a ches wi hou load
bea ing connec ion in ans e se di ec ion. These cons uc ions a e abou 150 yea s old and
hey a e also oo na ow o he cu en oad sys em. This pape p esen s a s eng hening
me hod o mason y a ch b idges using ans e se pos - ensioning. This me hod is e y use ul
no only o s eng hening in he ans e se di ec ion, bu widening o mason y a ches can be
aken as seconda y e ec especially in case o oad b idges. Se e al b idges we e success ully
epai ed wi h he use o his sys em which seems o be e ec i e and eliable.
1. Failu es o he a ch b idges in he longi udinal di ec ion
Some mason y a ch b idges we e buil wi h cylind ical aul s. The ype o aul is usually a hal o he
ci cle o a pa o he ci cle in he longi udinal di ec ion. Fo calcula ion o hei load bea ing capaci y
a simple ame model is used (Figu e 1). The e a e only no mal o ces and bending momen s ac ing on
he c oss sec ion. The e a e no o ces in he ans e se di ec ion in his case. Bu eal aul s a e
sepa a ed by c acks in he longi udinal di ec ion c ea ing mo e sec ions (Figu e 2). These c acks a e
ypically si ua ed in he cen al a ea o he aul s. They a e also si ua ed nea he le and igh edges in
a dis ance abou 1.5 m.
Figu e 1. F ame model o a mason y cylind ical aul .
2
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
Figu e 2. Railway (on le side) and oad (on igh side) mason y a ch b idges; ba el aul s we e
c acked in he longi udinal di ec ion.
A shell model can be used o explaining he cen al c acks. De o ma ion o he aul using he
shell model is shown in Figu e 3. We ob ain in e nal o ces (no mal o ce and bending momen ) in
bo h di ec ions: in he longi udinal di ec ion (axis X) and in he ans e se di ec ion (axis Y) oo.
Because o e y low ensile s eng h o mason y in he ans e se di ec ion c acks can appea in he
cen al pa o aul s. Explana ion o he exis ence o c acks nea edges is no so clea . The e a e also
ho izon al o ces due o a ic load in hese a eas. These c acks a e la ge, hei wid hs we e obse ed
om 10 o 30 mm. Some imes hei size is abou 100mm and his pa o aul is ully sepa a ed om
he ace (spand el) wall. In his case he s uc u e is uns able and he b idge should be closed o a ic.
Figu e 3. De o ma ion o he shell model due o
li e load.
Figu e 4. A angemen o ans e sal pos -
ensioning.
2. Design o s eng hening using ans e se pos - ensioning
S uc u al s eng hening is based on an analysis o he aul s using he shell model. Tensile o ces in
he mason y ba el aul a e aken by ans e se pos - ensioning (Figu e 4). Pos - ensioning is
achie ed by ans e se cables. Cables usually consis o monos ands, each wi h wo o h ee pieces.
They a e pushed in o d illed cable duc s and slo s p epa ed in ad ance in he mason y. Diamond
d illing and sawing echnology should be used. The special d illing suppo has o be used o
p epa ing o cable duc s. Ancho ages a e posi ioned in o new widening ein o ced conc e e (RC)
walls. P es essing o ces a e dis ibu ed by hem o he c own and suppo ing abu men walls, so no
X
Z
Y
4.9
4.9
0.0
3
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
c ushing o s ones o b ick mason y due local p essu e akes place. New RC ace walls can also be
used o widening o he b idge. This is he second bene i o his me hod and can be used in case o
he oad b idges. Widening can be la ge because he span o he can ile e can be designed up o 3m
leng h wi hou di icul ies. S abili y o he new ace walls is blocked by o ces in he ancho ages.
Those p es essing o ces should be aken om o ce equilib ium in ho izon al di ec ion. Ac i e o ces
a e he ho izon al pa o ea h p essu e due o a ic load and embankmen o e he aul mul iplied
by sa e y coe icien 1.5 o he pa ial sa e y coe icien design me hod can be used. Passi e o ces a e
in ac p es essing o ces in he ancho ages. A check o he s ess in mason y aul in ho izon al
di ec ion is also necessa y. Acco ding o he expe ience a alue 20% o s eng h o mason y in
pe pendicula di ec ion can be aken in o accoun . The example o a eal oad b idge econs uc ion
using desc ibed me hod is gi en in he pic u es below (Figu es 5, 6 and 7).
Figu e 5. Example o design econs uc ion o a oad b idge using ans e sal pos - ensioning wi h
RC ace walls.
Figu e 6. Collapse o he old mason y ace wall due o wa e , os and ice.
3. Example o he econs uc ion o he oad s one mason y a ch b idge
This oad b idge was buil in 1880. I is si ua ed on 1s Class oad No14. The main s uc u e is a ba el
aul wi h ace walls. Span o he aul is 6.6 m (Figu e 5). Cons uc ion comp ises sands one
mason y.
Longi udinal c acks simila o hose shown in Fig. 1 and buckling o one ace wall we e obse ed
in 2002. The buckling became so ex ensi e h ough he yea s ha inally a pa o ace wall collapsed
(Fig. 6). The a ic o e he b idge was s opped. Design o econs uc ion using s eng hening o his
aul by ans e se pos - ensioning was made in 2003. I is based on ans e se cables wi h ancho age
in he new RC ace walls. Widening o he b idge was achie ed as a seconda y e ec . The
econs uc ion was made in 2003 by MADOS MT l d., he pos - ensioning wi h d illed cable duc s was
made by Mi enga-s a by l d. Fu he pic u es show he b idge a e epai wo ks (Figu es 7 and 8).
4
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
Figu e 7. Plan o he s uc u e measu ed du ing p es essing; le - c oss sec ion o he aul wi h
cables; igh - iew o new ace wall wi h ancho age duc s o ou cables.
4. P ocess o he pos - ensioning
This me hod o epai o he mason y a ch b idge is a ypical example o he p ocess o pos -
ensioning. Cable duc s and slo s we e p epa ed be o e new RC walls as he i s s age o
econs uc ion. Walls we e buil in he second s age and ancho ages a eas we e p epa ed in conc e e
oo. In he hi d s age he c acks we e g ou ed by mo a injec ions and ca i ies o e he aul we e
illed h ough d illed holes oo. The pos - ensioning was ins alled a e conc e ing. The cables we e
ensioned o a magni ude o 180 kN wi h 5 minu es elaxa ion using single-cable p es essing jack o
a ange up o 200 kN. The cables we e ensioned om one side and he ex ension o indi idual cables
was measu ed du ing he p ocess. Tensioning was applied ca e ully in 20 kN s eps simul aneously
inc easing o he o ce in all cables o one a ch. A e pos - ensioning he cable duc s we e injec ed
wi h common cemen mo a and he slo s we e g ou ed up.
5. Example o he econs uc ion o clay b ick mason y a ch oad b idge
The aul is cylind ical wi h i s cen e line in he shape o a ci cula segmen om mason y 450mm
hick. Lime mo a was used, pa ly lushed away, o s eng h 0.2 o 0.4 MPa. The span was 4.5m,
wid h o he aul 10m, leng h o newly ealized ace walls 18.4m. The plan o he s uc u e is in
Fig. 7.
The main eason o he epai was buckling o bo h ace walls and co nices wi h subsequen
occu ence o longi udinal c acks 20-30mm wide be ween he aul and walls downs eam and abou
10mm ups eam. In ela ion o he abo e men ioned ype o de ec s he s eng hening was designed in
such a way ha new RC walls we e conc e ed agains he o iginal ace mason y walls. RC walls we e
s abilized by ans e se p es essing. Load bea ing capaci y was se in he longi udinal di ec ion
(sui able o Class A) wi hou dec easing he calcula ed alues caused by de ec s in he ans e se
di ec ion. Face walls we e s abilized by ans e se p es essing wi h unbonded cables. The cables
comp ised h ee s ands Ls 15.7 mm NPE (p o ec ed p es essing cables agains co osion,
manu ac u e Aus ia D ah ).
O iginal s one aul was p o ided wi h cable duc s, diame e 52mm o se ing o p es essing
cables. Cable duc s o med a ame wi h inclina ion 10 deg ees o he longi udinal adial plane. They
we e d illed om he aul owa ds he aces. On he uppe su ace o he s one aul he cables we e
se in slo s 70 × 50 mm. New RC walls we e conc e ed agains he o iginal aces and emaining pa s
o he ace walls. De ec s o he o iginal s uc u e a e shown in Figu e 8.
Measu emen o ho izon al de o ma ion o he s uc u e was ca ied ou in h ee e e ence lines
2.2m long. The e e ence lines we e made o s eel ancho s ixed in he aul mason y and s eel pipes.
De o ma ions we e cap u ed by induc i e displacemen senso WETA 2mm wi h a ange 0.001mm.
P es essing o ce was measu ed by a es e P oceq 200 kN. De o ma ion was con inually moni o ed
by a compu e . The empe a u e du ing moni o ing was p ac ically s able +5°C.
P es essing o ces we e induced e enly in o he aul so ha in luence o une en p es essing on
displacemen o conc e ed walls was a oided. Fi s , he cable No. 1 was s essed in all ou h ee-

5
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
s and cables, ollowed by he cable No. 2 and hen No. 3. P es essing o he aul was ca ied ou by
a p es essing jack in 20 kN s eps up o he designed p es essing o ce. I was possible o apply he
ull p es essing o ce in he cables (180 kN).
Figu e 8. De ec s in aul and senso base; le - c acks be ween ace wall blocks and mason y caused
by buckling; igh – senso base in he c own wi h displacemen senso .
6. De o ma ion o mason y walls
Comp ession in he walls inc eased e enly in all senso e e ence lines in ime. Du ing p es essing
mason y was comp essed immedia ely a e p es essing was induced, a e ancho ing o each cable
he inc ease ended o s op. A e p es essing was comple ed (las cable was p es essed) he inc ease
in comp ession s opped ully.
Rela ion be ween he de o ma ion (comp ession) o mason y and magni ude o p es essing o ce is
in Figu e 9.
Figu e 9. Rela i e comp ession o aul mason y du ing ho izon al p es essing.
6
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
Following conclusions may be induced om he esul ing ela ion:
1) I was possible o s ess he cables up o he p edic ed and designed o ce 3 x 180 kN = 540 kN
in one cable. Such o ces we e su icien in ela ion o s abili y o he s uc u e desc ibed in
Pa ag aph 2. Wi h such o ces no nega i e impac occu ed (c acks, c ushing e c.). The
conc e ed s uc u e emained s ill wi hou mo emen . The mason y was comp essed
p ac ically linea ly, i. e. p es essing was pe o med in elas ic pa o he mason y beha iou .
I means ha o iginal longi udinal c acks in aul (be ween b ick aul and o iginal s one ace
walls) we e p ope ly illed wi h g ou om he s a ical poin o iew, because no displacemen
occu ed when he e was no p es essing o ce. Subsequen g ou ing can eliably h ough
d illed holes ill c acks and ca i ies ha we e c ea ed by de ec s in o iginal aul .
2) Le side o he aul (le abu men a ea) was sligh ly so e – and i was comp essed mo e.
The di e ence is 10 % compa ed wi h he a e age alue; and i is p ac ically o no
signi icance. Vaul mason y com-p essed ega dless o cohesion be ween aul and
consolida ed in ill ( econs uc ion was ca ied ou du ing ope a ion wi hou emo al o in ill)
and wi hou isible in luence o newly ealized walls on o iginal walls and in ill. Such
in luence is hus no big and wi h ela i ely small magni ude o comp ession only s i ness o
aul and abu men in longi udinal di ec ion may be impo an .
3) A e age max. comp ession eached in sp ingings was 0.27 mm/m´; in he c own 0.28 mm/m´.
I is almos cons an comp ession in he ans e se di ec ion upwa ds and p o es good design
o p es essing o ces and also good ealiza ion o he econs uc ion. De o ma ion o he
s uc u e is in compliance wi h assump ions gi en in Pa ag aph 2 and i is possible o conside
hem as con i ma ion o his sys em o econs uc ion o b idge aul s (Figu e 10).
Figu e 10. View in o he ba el aul ; he pic u e shows sand s one mason y o he aul and cable
slo s a e comple ion.
7. Example o he econs uc ion o he clay b ick mason y a ch ailway b idge
The b idge ca ies a double ailway line B no – Česká Třebo á. Due o long e m in luences o
ope a ion he aul was sepa a ed by longi udinal c acks in o wo independen ly ac ing pa s unde
each ack wi h hi d edge aul bel 1.2 m wide which was abou o collapse. Fi s epai was ca ied
ou in 2014 – new ace wall s a ically secu ed by ans e se p es essing. A se o a m ampli ie s
loca ed in ail ack No. 1 and 2 axes p o ided echnology o de o ma ion measu emen .
The supe s uc u e o he ailway aul b idge is o med by segmen b ick aul wi h a clea ance
7.540 m and a ise o 1.970 m. The sp inging is 2.100 m abo e he pa emen and he c own is 4.220 m
abo e he oad. Thickness o he aul is abou 0.840 m; his alue comp ises in lowe pa abou 35 –
60 mm o o c e e. The epai s o bo h aul aces we e o med by new RC ace walls (Figu e 11)
7
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
which ans e s he induced ans e se p es essing in o he whole aul sec ion. S abiliza ion o
sepa a ed pa s o segmen aul was achie ed by induc ion o p es essing in he ans e se di ec ion
by ou cables. All mason y in he aul was g ou ed by cemen g ou in a ma ix 600x600 mm.
Figu e 11. Moni o ed b idge aul , new cas -in-
si u walls.
Figu e 12. Loca ion o c acks and di ision o
aul in o h ee sepa a e pa s.
La ge c acks di ided he aul in o h ee independen bel s (Figu e 12) which o cou se in luenced
he whole s uc u e. The edge bel had also isibly d opped when compa ed wi h o he pa s o he
aul . Diagnos ics by p obes con i med ha he c acks ex ended in o mason y and ha he mason y
was sa u a ed wi h wa e . The c acks we e ound in bo h abu men s in app ox. e ical di ec ion. In
abu men 1 he e we e mainly hai c acks, mos ly below he sp inging, whe eas in abu men 2 hey
we e up o 2.0 mm wide and ex ended up o ¾ o he heigh om he sp inging. O he de ec s we e
co esponding o he age o he b idge. Su ace de ec s we e isible on bo h ace walls especially
below non- unc ional d ainage.
8. Me hod o s eng hening he s uc u e
The main aim was o s abilize he s uc u e and es o e cohesion be ween sepa a ed pa s in ans e se
di ec ion. As a logical s ep sepa a ed pa s o segmen aul we e s abilized by p es essing in he
ans e se di ec ion. A e g ou ing o c acks he p es essing in he whole s uc u e was induced wi h
he help o ou p es essing cables h ough new ace walls. T ans e se con inui y is supposed o
inc ease s i ness o he whole s uc u e and hus join in luence o he aul in he ans e se di ec ion
o li e load ans e .
9. Join in luence measu emen p inciple
9.1. Measu emen echnology
De o ma ion measu emen was ca ied ou in h ee places along he axis o he aul , simila ly o
measu emen be o e he epai . (Figu e 13) [1]. A se o a m ampli ie s was placed in he same
ancho age places. Fo he pu pose o his pape da a om a m ampli ie s placed below ack axes we e
also used. (RZ1 and RZ2).
9.2. Da a collec ion o measu emen
Fo de o ma ion measu emen i.e. s ain unde a ic load a ain wi h engine 560 „Pan og a “ (Figu e
14) wi h weigh o 37 was used. (mo e p ecisely – alues measu ed unde he i s axle o he engine
in he di ec ion o a ic). Measu ed and eco ded passages o his ain we e eigh in 2013 and ou in
2014. Du ing measu emen he di ec ion and acks we e moni o ed. Sampling equency was 20 Hz in
2013 and 50 Hz in 2014 (Figu e 15). F equency 50 Hz su icien ly documen s all ib a ions o he
aul du ing ain passages.
8
1234567890
BESTIn a2017 IOP Publishing
IOP Con . Se ies: Ma e ials Science and Enginee ing 236 (2017) 012058 doi:10.1088/1757-899X/236/1/012058
Figu e 13. Posi ions o measu ing basis RZ.
Figu e 14. T ain wi h engine 560 „Pan og a ”. Figu e 15. T ain passage – sampling equency
50 Hz.
10. Assessmen o he measu emen
Passages in he same di ec ion and on he same ack we e compa ed; he compa ison was based on
da a measu ed by a m ampli ie s RZ1 a RZ2. Measu ed da a we e subsequen ly con e ed in o s ain
on he c own bo h ace (RZx-D) and back (RZx-H) sides o he aul . Values o s ain a e clea ly
shown in he ollowing able (Tab.1).