May be a dumb question but are we sure there is only one per girder? I can’t think of why, as a designer, I would put the rods exterior of the two exterior girders? I would likely put a tie plate over the top of the girder and run two rods, one on each side, then all interior rods would end up inside the concrete diaphragm. Otherwise there is a lot of unbalanced load running down through the deck to the girder and back to the substructure. This would be consistent with the “Details of Stress Steel Tensioning Unit” shown on Sheet 139 that calls for a top plate 7” x 3” x 28 ½ I also just looked at the VHB report from 12/11/23. It shows on the second page the north face and south face of Cantilever C and it looks somewhat like there is one rod on each side. It could be two views of the same rod; tough to tell. Something is not right with the header on this and the following pages. Would be worth noting the rod ASTM, yield strength, ultimate strength, and percent of yield/ultimate represented by 120kips. Assuming we don’t have that information, we should note that we don’t have it. We could/should provide a table of original cross section, reduced measured cross section, original 120kip stress, reduced measured cross section stress. We can compare those to the UTS results either here or elsewhere in the document. Should also leave some wiggle room regarding the 120 kips. That was likely the force to balance the dead load and possibly some portion of the live load. This is where the rating results come in. Possibly for another work product in the future. I didn’t find this directly in the metallurgy report. We should reference typical PT materials/ASTMs from the late 1960s if we don’t have one called out in the contract documents. Not sure this is totally factual. What did the contract call for at the time? CVN may not have been included in the contract at that time. I believe the minimum CVN of 15 ft-lbs at 40 F was adopted in the early 1970s. We have had test results on plenty of steel components over the years come back this low. It confirms the material is brittle. It doesn’t necessarily mean the material did not conform at the time of construction. May be worthwhile to state the anticipated stress on these cross sections at the stressing force of 120K and then compare that to the UTS test results stated above. See comment above about adding a table of values. I think we need another bullet that would postulate why, given that there was no plastic flow, the corrosion and reduced cross section occurred near the surfaces of the concrete. I would leave this out. It says the same thing as the rest of the sentence and leaving it in begins to point direct fingers. Should probably expand upon this to say something like: This study was focused on the materials properties of the samples and a metallurgical evaluation of the rods and fracture surfaces. Further evaluation of how those test results fit into our larger investigation is ongoing. Also, if you take this approach, you could cut a lot of my comments above and in the metallurgy report out, and address them in another future work product. Mr. John Preiss Rhode Island Department of Transportation February 15, 2024 Page Wiss, Janney, Elstner Associates, Inc. 2941 Fairview Park Drive, Suite 300 Falls Church, Virginia 22042 703.641.4601 tel www.wje.com Atlanta | Austin | Boston | Chicago | Cleveland | Dallas | Denver | Detroit | Doylestown | Honolulu | Houston | Indianapolis London | Los Angeles | Milwaukee | Minneapolis | New Haven | Northbrook (HQ) | New York | Philadelphia | Pittsburgh Portland | Princeton | Raleigh | San Antonio | San Diego | San Francisco | Seattle | South Florida | Washington, DC February 16, 2024 Mr. John Preiss State Bridge Engineer Rhode Island Department of Transportation Two Capitol Hill Providence, Rhode Island, 02903 Forensic Investigation of Failed Post-Tensioned Tie-Down Rods I-195 SB Washington Bridge North (700) Dear Mr. John Preiss : Wiss, Janney, Elstner Associates, Inc. (WJE) has completed analysis of two post-tensioned anchor rods from Pier 7 of the southbound I-195 Washington Bridge north structure (700). The subject bridge is an 18-span structure comprised of prestressed/post-tensioned concrete multi-girder approach spans (6 west / 11 east) and a steel multi-girder main span (Span 7). There is also a three-span prestressed concrete box girder ramp to Gano street at the west end. The bridge was closed to traffic on December 11, 2023 after construction workers reported observation of a full fracture of one of the post-tensioned (PT) rods used to tie down the end spans of east and west approaches to the substructure . WJE has been engaged to perform a review of the events leading to the discovery and to evaluate the cause of the rod failures. Site Visits WJE personnel Michael Brown and John Cocca visited the bridge on December 16, 2023. During the site visit we observed the fractured bars in situ and the configuration and condition of the cantilever beams and structural walls at Piers 6 and 7 that the post-tensioned (PT) rods were designed to connect. The approach structures comprise post-tensioned concrete balanced cantilever beams on concrete piers that support prestressed concrete drop-in girders between ( Figure 1 ). The end cantilever segments are tied down with PT rods to reinforced concrete corbels on the outward structural walls of Piers 6 and 7 to either side of the steel main span, Span 7, to counterbalance the loads of the drop-in beams in Spans 6 and 8. There are six concrete girder lines (A-F), with a single tie-down rod for each . The tie-down rods are embedded in the ends of the cantilever beams at girder lines B–E and extend down into the corbels, making them thus not visible for inspection. However, at the four outward corners at girder lines A and F of Piers 6 and 7, the tie-down rods lie to the outboard side of the cantilever beam ends and are exposed from the top of the beam seat to the bottom of the concrete deck. Mr. John Preiss Rhode Island Department of Transportation February 16, 2024 Page Fracture near bottom of tie-down rod. Fracture at top of tie-down rod. Reported bouncing of tie-down rod. Figure 1 . Framing Plan and Elevation of Spans 6 through 8, southbound I-195 Washington Bridge North (700) During the visit, WJE observed the fractured rods at two of four corners of Span 7, at Cantilever A and Cantilever F at Pier 7. At the time, a third rod, Cantilever A at Pier 6, was suspected of also being fractured due to observed movement of the cantilever/rod, but a fracture could not be visibly confirmed. A report noted bouncing of cantilevers, gaps at bearings of adjacent girder lines at the Pier 6 and Pier 7 connections. 1 VHB Visit Findings 12-11-23 (Additional Info) (Reduced).pdf 1 Photos of the fractured bars taken by WJE on December 16, 2023 are shown in Figures 2 and 3 . The bars were noted to be corroded on the exterior perimeter and the cross-section was reduced/tapered in the regions around the fractures. According to the original design drawings, the prestressing rods were to be 1 3/8” diameter threaded high strength pre-stressing rod to be tensioned to 120,000 lbs . each (Figures 4 and 5 ). Figure 2 . Pier 7 Cantilever A fractured tie-down rod Figure 3 . Pier 7 Cantilever F fractured tie-down rod Figure 4 . Section A-A excerpt from Detail Sheet 4 (Sheet 139) of original design plans (1967) Figure 5 . Section A-A excerpt from Detail Sheet 8 (Sheet 143) of original design plans (1967) Metallurgical Evaluation On January 5, 2024, John Cocca returned to the site. With the assistance of the current contractor, three segments of tie-down rod containing the identified fractures were obtained. The first was a long section of rod from Pier 7 Cantilever A below the fracture. Since the fracture occurred nearly flush with concrete at the top of the rod, it was not feasible to obtain the mating top face of the fracture. The second and third segments were those below and above the fracture at Pier 7 Cantilever F. The rods were wrapped and transported for shipping to WJE’s metallurgy laboratory in the Janney Technical Center in Northbrook, IL. The PT bar segments were catalogued, measured, and photographed before being subsampled for testing and metallographic evaluation. Short segments containing the fracture surfaces at the ends of the bars were cut off for further microscopic evaluation and chemical testing. An 18-inch segment from the longer portions of both bars were removed and shipped to Product Evaluation Systems (PES) in Latrobe, PA for mechanical testing (tensile and Charpy V-notch (CVN) impact). The details of visual, mechanical, chemical and metallographic evaluation are presented in Appendix A . Findings The results of the metallurgical evaluation are as follows: The tie-down rods were confirmed to be made fabricated from chromium-alloyed, high-carbon steel. Ultimate tensile strength (158.9 and 165.5 ksi) and associated elongation (10.7 and 13.5%), as well as hardness of the steel (HRC 32.1 and 34.6) tested at room temperature were consistent with material used for post-tensioning applications . The steel elemental composition and pearlitic microstructure are consistent with the reported material properties. Toughness of the steel (3.7 and 4.3 ft-lbf) as indicated by CVN tested at 40°F is well below that expected of a steel for this application. Significant corrosion existed on both the exterior perimeter and the fractured faces of the rods. Chemical composition confirmed iron and oxygen rust product with silicon, chloride and bromine in measurable concentrations; Thus, corrosion was likely induced by a combination of roadway deicing salts and spray from marine water sources, which this is consistent with the fact the highway bridge crosses a tidal portion of the Seekonk River near Narragansett Bay. At the locations of fracture, bar diameters had been reduced to 0.986 to 1.006 inch diameter and 0.882 to 0.925 inch diameter in Cantilever A and Cantilever F rods, respectively. From a nominal design diameter of 1.375 inch, this represents losses of 48% and 57%, respectively, of the original cross-sectional area. Metallographic evaluation showed the microtopography of the fracture faces had been degraded by corrosion and the thickness of corrosion product on the fracture faces was almost half that of the corrosion on the exterior perimeter of the bars. Observation of the pearlitic grains structure showed no evidence of plastic deformation (i.e., yielding) of the steel near the fracture surfaces and secondary cracking adjacent to the fracture surface is consistent with rapid, brittle fracture. Based on the findings, WJE concludes that it is highly probable that the two cantilever PT bars evaluated for this study fractured due to tensile overstress of remaining cross-sections that had been progressively reduced by corrosion. The substantial loss of cross section caused by corrosion combined with the steel’s poor toughness limited the bars’ capacity to accommodate a temporary overstress with plastic deformation and strain hardening, and thus the failure is believed to have been rapid and brittle in nature. Given the substantial loss of detail and the thickness of corrosion product on the fracture faces, it is highly unlikely that these fractures occurred within 6 months of the time the fractures were discovered . as suggested by them not being identified in the July 2023 routine safety inspection; however a more precise estimate of time of fracture is infeasible. WJE is continuing its investigation into the circumstances surrounding the failures and will provide our assessment of the broader factors in a separate report . Sincerely, Wiss, Janney, Elstner Associates, Inc. Michael C Brown, PhD, PE, FACI Associate Principal Mr. John Preiss Rhode Island Department of Transportation February 16, 2024 Metallurgical Evaluation