4 Test results
4.4 Bond influencing parameters
4.4.7 Surface pattern
Surface deformations are usually used in case of bars for internal reinforcement. If these reinforcements are used as NSM strengthening the effect of the surface patter on the bond behaviour is to be taken into account. Characteristic surface pattern of NSM reinforcements is shown in Figure 67.
Experiments were carried out with bars of round cross-section: with plain surface (epoxy coated C-8-S, Figure 67 a), with sand coating (B-8-SC, Figure 67 b), with helical wrapping and sand coating (C-6-SCW, Figure 67 c) with surface ribbing (G-8-RB, Figure 67 d).
Results of bond tests performed with the tensile-tensile test specimen and with the L-shaped specimen are presented in Table 8.
Figure 67: Characteristic surface patterns of NSM reinforcement bars
NOTATIONS.
C4: rectangular CFRP bar aspect ratio 10/10 C-8-S: CFRP smooth surface bar with diameter of 8 mm C2_4: CFRP strip with aspect ratio 2.5/15 C3_1: CFRP strip aspect ratio 10/1.4 (every point is an average of three measurements)
C‐8‐S C2_4
C3_1
C4
0,0 0,2 0,4 0,6 0,8 1,0
0,0 0,5 1,0 1,5 2,0
Mobilization capacity Ffu/ff∙Af
Perimeter to cross‐section ratio uf/Af[1/mm]
sand coated sand coated and spirally
wound ribbed
b) c) d)
plain a)
Reinforce ribbed re at the un epoxy co bond and capacity pattern a bond cap the C-6-S G-8-RB r along inc role simi the failur ductile fa of the rib brittle ch reinforce confining
Figure ement surface pa einforcement (G-8 nloaded end at lo oated reinforceme d shear propertie of the adhesive. F and shear failure o pacity but by the SCW bar where th reinforcement ha clined planes and lar to the case of re modes wherea ailure. Meanwhile bs or concrete cra haracteristics. She ment induces ge g effect of the adh
68: Failure mode ttern has influen 8-RB) almost no oad level of 84%
ent with smooth s es. Slip at unload
Failure was chara of the adhesive (F high tensile stren he bar nominal di ad intense surface
shear failure of s internal ribbed st s the ribs of steel e FRP reinforceme acking along inclin ar failure of the r enerally a less pr hesive usually ep
e of NSM mounte ce on failure mod slip was recorded and 87% for B1 surface (C-8-S) bo ed end initiated a cterized by mode Figure 68). The m ngth of the carbo iameter was 6 mm e ribbing which surface ribs (Figur teel bars in concre
bars are stiff and ent ribs are unabl ned planes forced ibs was mainly o onounced tenden oxy with a high t
ed reinforcemen des and on the b d at the unloaded -8-SC and for th ond mechanism w at 54%, 67% an erate cracking of t obilized tensile ca n fibres and the s m.
allowed almost re 51 and Figure 6 ete. The main diff d are able to shea le to shear the ep d by the radial co bserved at the lo ncy to splitting o tensile strength a
t C-8-S, C-6-SCW bond mechanism d length. In case o e C-6-SCW reinfo was governed by d 58% of the ult the concrete surro apacity was the lo small cross-sectio
no slip, failure in 67). In this case m ference was the re r the concrete be poxy adhesive in-omponent of the b
aded end where of concrete were and the concrete c
W, B-8-SC and G-stages as showe of the sand coated forcement at 88 a
the adhesion infl timate load. This ounding the groov owest for this bar on to bond surfac
nitiated with inte mechanical interlo educed slip expla
tween the ribs re between the ribs bond stress. This the slip was the h
we have to take cover thickness w
-8-RB
ed before. In case d specimens slip and 76%. In case luenced by the ad shows the defor ve with a fishbon explained not by e ratio in compar
ensive concrete cr ocking had an imp ined by the differ esulting in a gradu s and fail to shear
failure is explosiv highest (Figure 4 e into considerati which is normally
e of the started e of the dhesive mation e crack y a poor rison to
racking portant ence in ual and r failure ve with 3). FRP ion the higher
Bond characteristics of NSM reinforcements based on advanced test method PhD Thesis by Zsombor K. SZABÓ, supervisor György L. BALÁZS
for internal steel reinforcing bars. The mobilized tensile capacity was the highest for G-8-RB reinforcement explained by the high mean bond strength in comparison to the other specimens and the relative low tensile strength of the glass fibres in comparison to the carbon fibres.
B-8-SC and G-8-RB bars had comparable tensile properties, therefore we conclude that bond strength of ribbed bars is higher with over 40% compared to sand coated bars.
Table 8: Tension-tension bond test results (Appenix E) and results of the L-shaped specimen (Appendix C) for FRP reinforcement with different surface pattern
Reinf.
bar
Ultimate tensile
load
Ultimate loaded end slip slu [mm]
Initiation of the unloaded end
slip
Mean bond strength
Mobilization capacity1
Surface pattern
Ffu [kN] 50% 90% Fx/Ffu N/mm2 Ffu/ff·Af
C-8-S 40.7 0.298 0.763 0.54 5.4 0.29 epoxy coated, plain
41.5 0.328 1.220 0.67 5.5 0.29
42.7 0.374 1.990 0.58 5.7 0.30
C-6-SCW 33.2 0.372 0.905 0.84 5.9 0.53 sand coated
and helical wrap
34.5 0.420 0.961 0.87 6.1 0.56
39.12 6.9 0.63
L-shaped specimen
40.58 0.287 0.596 0.96 12.3 0.69 49.00 0.199 0.572 0.90 14.9 0.84
B-8-SC 28.45 0.472 1.337 0.88 3.8 0.57 sand coated
32.74 0.585 1.100 0.76 4.3 0.65
27.78 3.7 0.55
L-shaped specimen
38.50 0.254 0.635 0.96 8.8 0.77 38.12 0.184 0.553 0.92 8.7 0.76
G-8-RB 45.83 0.710 1.465 1 6.1 0.91 surface
ribbing
37.92 5.0 0.75
43.55 0.635 2.013 1 5.8 0.86
L-shaped specimen
53.81 0.139 0.72 12.0 1.05
48.65 0.679 0.009 0.72 11.1 0.97
1 comparison of values has to be made with regard on the different FRP material properties
Sand coated reinforcements were developed to be embedded in concrete as internal reinforcements. Generally high strength epoxy adhesives are used for adhesion in NSM applications. The deformations of the sand coated reinforcements should guarantee higher average bond strength in comparison to plain surface bars. But in some cases the shear failure of the deformations can cause bond failure at lower bond stress levels observed in case of B-8-SC bar (Figure 44). In a comparison of C-6-SCW and B-8-SC reinforcements (different modulus of elasticity) we notice the benefit of the helical wrapping which made possible an increase of the mean bond strength in this comparison. In case of the C-6-SCW bars the longitudinal splitting of the adhesive was observed starting from the unloaded end then cracks propagated through concrete. In case of B-8-SC bar (higher adhesive thickness) longitudinal splitting of the adhesive was not typical. Cracking of the concrete along inclined planes starting from the bottom of the groove triggered by splitting into two halves of the specimen was observed for the B-8-sc bar.
In specimens where secondary concrete failure was of concern bond behaviour was mainly influenced by the radial component of the bond stress triggered by the mechanical interlocking between bar deformations and adhesive. For plain or moderated deformations, bond is mainly influenced by adhesion or by the adhesive shear strength. In case of higher intensity deformations bond is controlled by mechanical interlocking between deformations and adhesive.