The 600 mm thick walls of two anchor blocks of a sea structure suspension bridge had during construction suffered cracking due to freezing of water left standing in the cooling tubes installed in the walls to avoid thermal cracking during hardening.
In a pilot study radar and impact-echo was evaluated for detection of the delaminations. Only impact-echo proved to be successful and was subsequently used to detect the extent of the delaminations as well as the depth of the defects in four critical areas - each 60 meters by 7 meters - above the sea level on the vertical faces of the anchor block. Testing was done in very rough winter conditions.
Following the testing, 50 cores were extracted. All the impact-echo findings were confirmed in delaminated as well as in solid areas. The cracking detected was subsequently injected.
Shortly after completion of a sewage plants sludge basin rain fell heavily for three weeks. The rain penetrated the surrounding soil, lifted the basin and partly washed away the sand bed material below the fiber reinforced concrete bottom plate. Before the basin could be taken into operation the voids below the bottom plate needed to be detected and injected.
Radar and impact-echo was initially attempted used for detection of the voids. Only impact-echo proved to be successful.
Testing with impact-echo was performed on the entire plate, 65 meters by 6 meters.
Six major areas were detected as voided. Cores were drilled out for confirmation.
Injection with a pressurized mortar took place through the cored holes until the bottom plate was lifted 1 mm. Finally, the bottom plate was re-tested with impact-echo and no voids were detected. The sludge basin was put into service.
The concrete vessels for pulp production in paper mills are tiled on the outside and the inside on the 300 mm walls. The inside tiles protects the concrete as the pulp mass is slightly acetic. Traditional testing for debonding of the inside tiles is made by hammer tapping the tiles after the vessel is closed down and emptied for pulp mass. Such an inspection procedure is very costly as the pulp production needs to be shot down for 3-4 days.
With the vessel in service impact-echo was used from the outside of the vessel to detect the debonding of the inside tiles.
Detected loose tiles on the inside were confirmed by coring. Similarly, areas found solid by impact-echo were cored and found to be solid.
An impact-echo testing program was set up for regular inspection of vessels in service, involving considerable savings to the owner of the paper mills compared to traditional testing.
The bridge deck of a highway bridge had been subjected to de-icing salts during the winter for 15 years. The deck, 30 meter long and 12 meter wide, had a 300 mm concrete slab with an 50-100 mm asphalt overlay and a bituminous membrane in between.
Impact-echo detected the areas where debonding of the membrane had taken place, and, in addition, zones with delaminations in the concrete slab caused by chloride induced corrosion at the top reinforcing steel as well as at the bottom steel. Testing was performed partly from the top surface and partly from the bottom of the slab.
A limited number of cores confirmed the impact-echo test results.
A repair program was undertaken involving installment of a new asphalt/membrane overlay after localized repair of the concrete slab in the corroding zones had taken place.
Two supporting walls, 60 meters long and 5 meters high, of a major highway bridge were cast with a high performance concrete. After removal of the formwork the surfaces revealed a number of honeycombs, which were repaired by the contractor.
The owner of the bridge wanted to test the integrity of the 500 mm thick walls before acceptance. Impact-echo was selected for the testing.
The testing performed revealed two locations with honeycombs, otherwise the walls were evaluated as solid. The integrity of the walls was accepted by the owner.
In addition, impact-echo found an 8 mm bituminous membrane installed on one face of one of the walls against the earth-fill. This membrane was not indicated on the drawings. Installment had been made by the contractor to avoid water intrusion from the earth-fill.
Two 25 years old tunnels carrying railway tracks suffered from ASR (Alkali Silica Reactions). Prior to upgrading the tracks for high speed trains the extent of the cracking needed to be established, in particular to clarify if delaminations had occurred.
Impact-echo was performed. Three classes of signals were detected, solid concrete, concrete with multiple smaller cracks and one area close to a deck edge exhibiting a major reflection surface at a depth of 180 mm.
Coring revealed the reflection surface to be caused by a cast-in 40 mm in diameter rubber tube. The tube was not indicated on the drawings.
Coring also confirmed the solid concrete found to be solid and the concrete with multiple reflections to contain 5-30 mm long cracks, predominantly parallel to the surface.
An intended strengthening program was decided not to be carried out.
A 1 meter thick reinforced concrete bridge deck 30 years old, carrying a rail track, suffered from ASR. The span of the deck was 25 meters.
Impact-echo was used to detect the integrity of the slab.
A horizontal crack at mid-depth in the deck was detected across virtually the entire span.
The crack was confirmed by coring.
The bridge was judged unsafe and was demolished. During demolition the lower half part of the slab came down in its entire length.
The highway consisted of heavily reinforced 0.8 – 1.2 meter thick concrete slab sections, each 20 meter long and 16 meter wide. The slabs had a 100 mm dense asphalt overlay installed.
The slabs had heavy side walls. The structure was constructed below ground water level with a water stop installed in the joints. A bituminous material had, furthermore, been applied in the 2 mm wide joints between the slabs. The slabs were prevented from floating by connection to a large number of piles driven into the substrate. The slabs near the joints needed to be tested for delaminations prior to construction of a high rise building on top of the side walls of the highway.
The slabs were tested with impact-echo in an area extending 2 meters on each side of 16 selected joints.
Coring was performed, one on each side of the joints tested. The impact-echo findings proved to be 95% correct.
The delaminated slabs were repaired by removal of the concrete past the delaminations and casting of a new concrete well bonded to the concrete substrate.
Impact-echo was performed on tendon ducts in a post-tensioned concrete tunnel deck after injection of the ducts with a cement grout. The testing was made at the highest top point of the ducts in the middle of the 40 meters long span of the deck. The length of the tested cable ducts was 4 meters.
The 1 mm thick steel ducts had a diameter of 80 mm and were positioned 120-190 mm deep into the reinforced slab.
The impact-echo test signals were classified in three types of signals, fully injected ducts, partially injected ducts and empty ducts.
The test results were confirmed visually at six locations, two in each class, by inserting an endoscope through the holes of the top filling tubes.
Thirty cable ducts were tested with impact-echo of which 21 were evaluated as solid. The remaining ducts, partially injected or empty, were re-injected with grout from the top tubes.
The steel liner in a nuclear plants containment wall was tested with impact-echo for corrosion in the vicinity of the welding joints between the liner and the steel tubes containing the electrical outlets.
The 5 mm thick steel liner was positioned 20 cm deep in the 1 meter thick wall.
Corrosion of the liner cause an air gap and a reflection frequency of approximately 10 kHz for the 20 cm deep air gap. Similarly, if no corrosion has occurred, the impact-echo P-wave will be reflected at a frequency of 5 kHz from the steel liner.
The impact-echo testing mapped quickly the corroding areas of the liner. Coring confirmed the test results. The liner was repaired after removal of the 20 cm cover concrete in the corroded areas.
The 250 mm reinforced concrete slabs and decks of a waste water plant had been overlaid by a 70 mm – 150 mm reinforced concrete layer to ensure proper drainage of any water spillage. A bonding agent had been applied between the layers.
The specifications called for Bond-Testing to be performed to make sure a proper adhesion between the layers had been achieved.
Prior to performing the time consuming and destructive Bond-Testing, impact-echo was used to point out areas not bonded, quickly and non-destructively.
Coring at a couple of locations in non-bonded areas confirmed the impact-echo test results
In the areas found by impact-echo to be bonded, the Bond-Tests were performed subsequently.
An intensive fire broke out in a high voltage transformer room of an industrial building complex. The 300 mm reinforced concrete deck and walls were heated vigorously for about one hour.
Surface opening cracks showed up, especially on the underside of the deck, after cooling of the structure.
The deck carried a cooling tower in operation. The owner needed urgently to have the deck investigated non-destructively for depth of the surface opening cracks as well as for delaminations. Impact-echo was chosen.
The wave speeds of the deck ranged from 2200 m/s to 2800 m/s. The depth of the surface opening cracks were measured between 19 mm to 109 mm. Delaminations were detected at depths ranging between 83 mm and 270 mm.
The deck was subsequently supported and strengthened by a steel structure.
After casting of a 650 mm thick lightweight concrete roof, testing was performed with impact-echo to detect air voids beneath the heavy top reinforcement layers.
The concrete cover was 50-60 mm. At the top of the roof two layers of 32 mm reinforcement were present, closely spaced.
Prior to testing, the P-wave speed of the lightweight concrete was measured on the surface using two transducers spaced 300 mm. The P-wave speed was measured to be 2500 m/s.
Impact-echo detected, using a 5 mm impactor, the voids at a depth of 120 mm from a measured frequency of 10.4 kHz.
The solid parts of the roof were evaluated by means of the solid frequency at 1.92 kHz and the absence of the 10.4 kHz void peak in the frequency spectrum.
The thickness of a concrete road pavement was evaluated with impact-echo.
Prior to testing, the acoustic impedance difference between the pavement and the substrate was checked to be sufficient for P-wave reflection to occur from the interface.
The P-wave speed Cp was measured for every 10 meter on the surface using two transducers spaced 300 mm. Based on the P-wave speed the thickness was evaluated from the solid frequency fs measured, using the equation T = Cp/2fs.
800 measurements were performed. The thickness estimated was at 8 locations compared to the actual thickness. The thickness measured by impact-echo was within ± 8 mm from the actual measured ones.
The 200 mm design thickness of the pavement was met by 92% of the test results.
Fourteen tubes, each with a dimension of 2.25 meter in diameter, 2.25 meter long and 23 cm thick, were tested for flaws prior to putting the tubes into operation.
The 23 cm thick tube walls were reinforced in two layers, positioned 50 mm from the outer and inner surface. The tubes had been cast standing vertically in steel forms.
Each tube was tested from the inside in grid containing 20 test points.
The cause of the defects is believed to be pre-mature stripping of the steel shutters. The consulting engineer rejected all thirteen tubes.
The cable ducts were located accurately by means of radar. Subsequently, the DOCter impact-echo was used to detect if the ducts were fully grouted or voided.
The wavespeed of the concrete was found to be 3,820 m/s using the Longship, followed by impact-echo testing right above the cable ducts.
A number of voids were detected. At one location were impact-echo indicated a refelction surface 123 mm deep, the concrete was removed and the cable duct opened. The tendons were not encapsulated by grout and showed signs of corrosion. The actual depth to the void in the ducts was 122-125 mm.
2. s'MASH Impulse Response Testing Cases.
An existing reinforced concrete approach ramp to a freeway has been overlain with a 75-mm thick silica fume unreinforced concrete. Surface cracking appeared six months after construction, with debonding of the overlay becoming apparent at certain locations on the slab. s’MASH testing was brought in to determine the exact extent of the debonding using the mobility contour plot for the whole deck.
Testing was completed in three hours, minimizing traffic closure. Additionally, an estimate of those areas of the deck with incipient debonding problem was made by analyzing the s’MASH stiffness and mobility slope contour plots. This facilitated the decisions to be made for the repair solution and the quantity of deck overlay to be replaced.
A common form of prestressed beam bridge structure relies on cross-rods and grouted keyways between the beams to form a single span bridge. Each hollow core beam is usually between 20 and 30 m long and 900 mm wide.
s’MASH tested one of these bridges through a 100-mm thick concrete overlay, with much greater mobility and lower stiffness on one half of the bridge. In the half of the bridge with higher stiffness the transverse rod system is functioning as designed.
How ever in the other half of the bridge the opposite is true; either the transverse rod system was not properly installed or not present. It was reported that during recent renovation of the bridge, the crew had problems with one of the rods. s’MASH testing showed the ability of the method to detect changes within the bridge, such as a loss in functionality of the transverse rod system. This loss in the rod system can be detrimental to the overall load carrying capacity of the bridge. The speed of the s’MASH testing makes it a valuable tool for testing other similar bridges.
The s’MASH test evaluated a 9600 sq m industrial reinforced concrete floor slab for the presence of poorly consolidated areas, as well as voiding in the slab sub-base. The slab design thickness for the entire area tested was 200 mm. The s’MASH mobility and stiffness contour plots indicated areas with probable slab thickness less than 200 mm.
Coring of select locations and impulse radar traverses correlated overall floor slab thickness with core thickness.
The s’MASH Voiding Index when >2 showed voided/poor support areas beneath the slab. S’MASH also identified an area of potentially honeycombed or voided concrete. This was confirmed by coring at selected locations. In addition, areas of lower stiffness were located by s’MASH along construction joints, which is typical of jointed slabs on grade. Fieldwork was completed in 3 days!
Two 60 m high x 30 m diameter cement silos were being constructed by the slip-form technique. Low density and honeycombed concrete was observed on the silo surfaces immediately above the interior silo ring beam at 20 m above ground level when the forms were raised. The concrete on the interior and exterior walls immediately above and below the steel ring beam was tested by s’MASH, as well as the concrete at the level of the steel ring beam on the silo exteriors. No areas of high s’MASH mobility and/or mobility slope were detected in areas of visibly sound surface concrete.
No delamination of the concrete along reinforcement planes was observed from the test results. It was concluded from the s’MASH testing that the concrete in the walls was sound and integral, apart from those areas where rock pockets and tear cracks were readily visible on the concrete surface. Cores were taken in the concrete above the ring beam, including one core through a visible tear crack. All the cores contained sound, continuous concrete, confirming the s’MASH test findings.
A mechanical-draft cooling tower with four cells included 300-mm thick reinforced concrete walls 15 m high. The walls were lined with 2-mm thick plastic coating to prevent excessive water loss through the concrete. After 2 years in service, blisters were observed on the walls, and when pierced, showed zones of saturated, poorly consolidated concrete. s’MASH testing of the concrete through the lining mapped out zones of honeycombed concrete, and gave an accurate estimate of repair quantities.
All zones with poor concrete consolidation located by s’MASH were confirmed during the repair program when the old liner was removed. The speed of testing and analysis meant that no facility downtime was necessary, resulting in great savings for the owner.
Vertical cracking distress was observed in terra cotta clad column covers and mullions on the Wrigley Building, Chicago. s’MASH tests were conducted from the 4th to the 13th floor. The tested column and mullion were representative of similar elements around the entire perimeter of the building where the problem of vertical splitting or cracking of the terra cotta cladding was confirmed.
Features such as debonding of the unit from its supporting back-up masonry, and non-visible internal delamination or splitting within the terra cotta unit were detected by s’MASH. The stiffness recorded for each terra cotta unit was plotted versus the position or height of the unit from the 4th floor upward. For the column cover, the residual stiffness measurements indicated very strong concentrations of high compressive stress directly above and below the position of the steel shelf angles supporting the terra cotta at each floor line.
A 180 m circumference by 5 m high aeration tank showed water leakage at cold joints. s’MASH testing of the tank walls was used to locate and quantify the extent of any hidden poorly- consolidated pockets that could have an effect on the engineering performance of the structure. Those areas considered to have the lowest degree of concrete consolidation were located, and selected cores were taken.
The bridge was built in 1907, and the spandrel arches are approximately 25 m wide by 25 m long. Preliminary information suggested that the arches are approximately 1.1 m thick at the spring line, decreasing to 600 mm thick at the soffit, and that they are filled with soil to form the subgrade for the flexible paving bridge deck. The concrete in the piers had been placed in wooden forms in approximately 300 mm lifts.
In a pilot Study the anchor quality of granite panels to steel frame of a high rise building was evaluated.
The granite panels had been fastened to the steel frame by means of 6-8 epoxy anchors. Holes had been drilled into the panels to half depth of the thickness, the anchors epoxied in the holes and subsequently fastened to the steel frame.
By means of the dynamic stiffness and the average mobility the s’MASH classified the panels in three groups: loose panels, panels without regular anchoring and well-fastened panels.
The concrete weld joint technique has been developed to produce continuos monolithic slabs using pre-cast elements. Fiber-reinforced high strength concrete is cast in a gap between two adjacent elements with extended reinforcing bars. The element (3x6 meters) are ordinary precast, reinforced elements placed 100 mm apart. The reinforcement extends 80 mm into the gap. Transverse reinforcement is positioned in the gap as well.
Load carrying tests have shown that the weld permits transmission of bending moments equal to the capacity of the elements. Anchorage failure will occur as yielding of the elements reinforcement, not of the weld's. That is if the weld has sufficient strength upon loading and is flawless without around the reinforcement.
The weld joint cast in-situ was tested for maturity with the COMA-Meter, and with the DOCter Impact-Echo for flaws, on two projects in Denmark as part of the quality control.
A video of the weld joint technique and the testing has been produced. The video is available from Densit A/S, Rordalsvej 92, DK-9100 Aalborg, Denmark. Phone: +45-98167011, Fax: +45-99337788, Email: lan@densit.dk
Pull-out testing was specified on the Danish Great Belt Link for quality control of the cover. About 40,000 tests were preformed.
The testing is described in the booklet: "Pull-out testing by LOK-Test and CAPO-Test with particular reference to the in-place concrete of the Great Belt Link" available from the Danish Concrete Institute, Bredevej 2, DK-2830 Virum, Denmark. Phone: +45-45986182, Fax: +45-45986185, Email: xfb@ramboll.dk
The Experience with the pull-out testing has now been published by the owner of the bridge A/S Strorebælt in one of the Storebælt Publications called Concrete Technology. The book is available from Sund & Bælt Holding A/S, Vester Søgade 10, DK-1601 Copenhagen V, Denmark. Phone: +45-33935200, Fax: +45-33931025, Internet: www.storebaelt.dk
The program involved the construction of a series of full-sized concrete structures in the large Building Test Facility at Cardington where they where subjected to comprehensive tests during the building process and of their performance.
The principal partners of this program was:
- British Cement Association
- Building Research Establishment Ltd.
- Construct - the Concrete Structures Group
- Reinforced Concrete Environment
- Department of the Environment, Transport and the Regions
Key messages:
- Allows significantly increased efficiency of in-situ concrete frame construction.
- Enables early striking of formwork and its economic re-use. This is further explained in a companion Best Practice Guide, Early striking for efficient flat slab construction.
- Enables early prestressing with safety.
- Can give an indication of long-term strength, enabling early confirmation of the quality of the concrete as placed.
- Use pull-out inserts cast into the concrete to determine early age strength
- Horizontally cast members (e.g. slabs) - locate inserts, using a floating cup, on the top surface of the slab near the end of the pour.
- Vertically cast members (e.g. walls, columns) - locate inserts on the formwork, with provision for early access before striking.