All right, we're coming up on almost two minutes past the hour, so we're going to get started. Good morning to the West Coast, and good afternoon to those out East. Welcome back again to the Post-Tensioning Institute's monthly webinar for June, obviously. As I'm sure a lot of you guys are as surprised as I am, we're in June already. A little bit of a shocker when I was looking at it today, and it was even a bigger shocker was that we're almost halfway through 2024 here in a couple weeks, so yeah, all of us busy people, it's June, but my name is Kyle Boyd, I'm the moderator of today's session. I'm also the chair of the Education Committee, EDC 130 for the Post-Tensioning Institute, and that's the committee that launches this monthly webinar. As we've talked about in previous webinars, this webinar happens the same time every month, which is that second Wednesday of the month at this exact time, whether you're on the East Coast or the West Coast, depending either in the morning or the afternoon. As you can see, today's topic is resilience of post-tension box girders, and we're pretty excited because historically throughout the year, we've been talking about buildings, and today we're going to go away from buildings and focus more on bridges, and what's even more exciting is today's presentation even has some slides that show fire, and I would say most of us would agree that whenever we see some PowerPoint slides that show fire, things get a little more interesting, so we're pretty dang excited about today's presentation to get into the world of bridges on there. Before we do that, we do have to go through our housekeeping slides, just like we do every month, the first one being the continuing education. If you're logged in through your registration and you're on for the entire hour, you'll get this continuing education credit emailed to you through RCEP, which is the agency that we use on here. If you do have to leave, client calls or somebody comes into your office and you need to step away for part of the presentation, no need to fear, you can go online and you can watch a prerecorded or this recorded presentation afterwards and fill out a couple questions at the end, or if you straight up warn your colleagues misses today's presentation, they can go online, watch it, answer those questions, get that credit, all for free, right? Good stuff. So copyright material, this is just our slide that makes lawyers happy, says thou shall not copyright, however, if anybody wants some information from today's presentation, we'd be more than happy to provide it, just reach out to the presenter or us at PPI and we'll give you that information on that. Then the protocol for today's webinar, we talked about the continuing education already, so the other ones talked about is the fact that all you attendees, you guys are just in listen only mode, we cannot hear you or see you. So because of that, if you have any questions, use the question feature in Zoom here and at the end, we'll go through the questions, if we've run out of time for the questions, once again, there'll be contact information for you to get ahold of us for the presenter to go through those. So with that, I can now introduce our speaker. Our speaker today is Mr. Brian Merrill, Brian works for WJE. Most of us would agree that WJE is a globally recognized firm for their technical expertise and Brian's a principal for WJE, he's out of their Austin office and he focuses in the bridge sector. He's an expert when it comes to bridges, he's an expert in design, forensics, repairs, troubleshooting, essentially, you name it in this market, he's your man. He has over 40 years of experience, as you can see, and he's very involved in the Post-Denture Institute. So involved that he's a fellow, and for all of you that aren't as involved in these different extracurricular groups out there, fellow just means you put in way too much time over your career helping out an organization and you're being recognized for that. He's also a chair of a committee and active on two other committees, and on top of that, he's now part of the Post-Denture Institute's Technical Advisory Board, or we call it TAB. And TAB is really the most senior technical experts within the Post-Denture Institute, and a lot of what they do is review the technical content within PTI before it gets published to ensure that there is, one, it's correct, and two, that we're going along with a strategic direction of PTI and not trying to bring in marketing or anything else like that with outside sources. So his expertise within the bridge industry has made it to the point that he's even a member of the TAB committee that reviews everything there. So with that, we really can't think of anybody out there who we would say is more qualified to be presenting today on this topic, and we're going to hand it over and let him go with it. Thank you, Kyle. We'll arm-wrestle about that expert term sometime later. As Kyle mentioned, I'm going to present a couple of case studies on performance of grouted post-tension box girders, and I'm going to talk about four primary subjects, which are on the screen, the corrosive environments, construction inadequacies, which in real terms means poor grouting. He did advertise the fire damage, and I've got a couple of pretty good slides on that one, and then even some demolition challenges where we had to take out a post-tension box girder bridge, and to me, that also helps explain some of the resilience of these structural systems. So getting right into it, we're going to start with the corrosive environments, and I'm going to talk about the JFK Causeway. This is a bridge I've spoken about in some ASB presentations, but I think this is the first time we brought it to PTI. If you're not familiar with the JFK Causeway, it's located on the Texas Gulf Coast near Corpus Christi. The two red stars show where the bridge is in the world, and it's about a mile as the crow flies from the Gulf of Mexico, so a very highly corrosive environment and a warm environment as well. What's unique about the bridge is it was the first precast segmental bridge built in the U.S. We opened it in 1973. It's not a very large bridge, it only has a 200-foot main span, and it was built using balanced cantilever precast construction, but the important thing was it was the very first one. It kind of started the whole process. Some design and construction aspects that were unique to the bridge at the time, they did use epoxy-jointed segment faces on both faces. The specifications didn't have a lot of information or requirements on the metal ducts, so they ended up using some rather thin-walled metal ducts, and I'll describe the impact of those in a slide or two. They did specify that grouting was to be performed within 48 hours, which is an aggressive approach compared to today's practices, but they felt it was important. The other thing is, this was early 70s or late 60s design, they didn't take any special mitigation measures to prevent or mitigate corrosion. For example, all the steel in it is black rebar, and in most cases we had one inch of clear cover. So not an unusual design for the time, it would be unusual for the current state of the practice, but that's the way they were doing it at the time. The grouting system was pretty straightforward and pretty simple, using a combination of cantilevered and continuity tendons. So the cantilevered tendons are the blue ones I've shown here, and we're going to talk a little bit about those anchor blocks that I've circled there on the left, that anchor roughly mid-height in the web, and they're now covered up, they're no longer visible or accessible to us. Then we have the continuity tendons. On the left are the in-span continuity tendons, on the right in the red are the mid-span. You can see there I've highlighted that those tendons actually anchor up at the deck level, and that's a design that's no longer in practice, but does have an impact on the performance of this bridge. Here's a couple of pictures and images from the original plan, showing the cantilevered tendon anchorages on the left, and the anchorage in the form system, and then the continuity tendons on the right, showing the deck level anchorage from the contract plans, and then the stressing hardware that they had to use to stress those continuity tendons. The little recess there was just filled with regular cast-in-place concrete, with no special treatments, additives, bonding agents, anything. They just filled it full of concrete. I mentioned the thin-walled metal ducts, and that caused some challenges during construction with duct alignment. They first noticed it in the form of some high friction values when they were starting to stress the tendons, and they actually had to replace several tendons due to some installation trouble. One of those troubles was when they pulled the tendon bundle through the duct, if it snagged on a seam or a lip on that duct, it literally pulled the metal duct out with the tendon in a number of cases, and they documented that really, really well in a series of research reports done by the University of Texas. So that was one of the challenges that they had during installation and stressing of the post-tensioning system. They also had some other issues during construction. The most serious one was some web cracking that developed during stressing of the cantilever tendons. By the time they started seeing these cracks, all 88 segments had been cast, and so if the decision was made to make a change, that was going to have a significant impact. This shut the project down for about nine months, and the University of Texas, who actually designed the bridge, did a pretty exhaustive study on the mechanism that caused the cracks and then also its impact, and they determined that it was due to insufficient local zone reinforcing and the fact that the fabricator terminated some rebar in these regions to allow the anchors to be installed, and unfortunately, that allowed some cracking to occur. They didn't think it was going to be a long-term performance issue, and so they recommended epoxy sealing, and they kept going. The image on the screen is a current-day image of the cracks. You can see the epoxy seal, but you can also see there's been no distress associated with those cracks. So in 2020, we were hired by TxDOT to go do a pretty in-depth condition assessment, and that started with a review of the contract plans and the very many TxDOT reports on this bridge. The University of Texas had six different research reports, I believe, on this design, construction, and even after construction on the bridge, and so there was a lot of information in there. And then we conducted a visual inspection of 100% of the interior and the exterior surfaces, and then we used a combination of ground-penetrating radar, some impact echo, ultrasonic tomography in an attempt to locate the PT system and any anomalies that might be found within that system, and anomalies being typically voids in the grout. Then we did some destructive openings to inspect the grout and tendon conditions, and we sampled and tested a number of samples of grout to determine their current properties and their impact on corrosion performance. As part of this effort, we were asked to do some service-life modeling, and so in addition to the PT system, now we're looking at taking cores for chloride testing and petrography. We did a pretty good clear-cover survey using the GPR. I mentioned that they had one-inch clear cover on this thing, and so in quite a few cases, we found that that was often busted. We also did half-cell corrosion potential at select locations and then corrosion rate testing at those same locations, just trying to characterize the state of the structure at that time with respect to corrosion. And TxDOT's goal for all of this was to try to get about a 30-year extension for the bridge. Right now, it's 51 years old, so they're looking to get it to roughly 80 years. The images on the right just show you some of the half-cell testing and just a photo of the surface. You can barely see the marks where we marked out rebar and that sort of thing for our corrosion testing. We only found one anomaly type in the bridge, and that was we were able to find those missing ducts. I'll show you a photo of what that looks like here in a bit. We found only two grout voids. Now, we didn't look at every tendon, but we looked at quite a number of the tendons, and none of the voids really exposed strands that didn't have a coating of grout on them. When we did open up the ducts and the anchorages, we found that the strands were in like-new condition, which, considering this bridge's exposure, is encouraging. We also found evidence of their regrouting practices, and so I'll run through a couple of quick photos to show you what we saw. This is one of the two grout voids we found. This was at one of the deck-level anchors. You can see that the strand does have a coating of grout. This void goes back about nine inches, but there was no corrosion associated with it. This is also one of the cantilever tendons that we opened up at the deck level, and you can see the condition of the strand. It's really quite good, considering this is a 50-year-old bridge in a marine environment with no corrosion protection measures at all. We did do some phenolphthalein testing for pH. The top stain is the concrete surrounding the duct, and then you can see the very high pH of the grout before we chipped it off. This is one of the ducts. It's inside one of the webs where the duct had been removed during installation of the PT strands, and you can see the shape that the duct left behind in the concrete, but there was no metal duct to be found. Once again, it had no negative impact on the condition of the strands. This was also a common site when we opened up a lot of the ducts, and we were initially a little bit concerned about the corrosion staining visible on the inside of the ducts until we read one of the UT research reports, and they mentioned that they had observed this during construction. What happened was these segments were cast in Corpus Christi and stored near the bay, and so some of the ducts during that nine-month period, while they were waiting to resolve the cracking issue, began to corrode on the inside. Obviously, on the outside, they were encased in concrete, so they transferred the staining to the grout, but it's been no negative impact of that corrosion on the duct. Here's one of the deck-level tendons. We removed the pour-back concrete, and you see it's in very good condition. That hole in the middle that's filled with grout is what we drilled out to do our bore scopes with, and then finally, here's the evidence of re-grouting. They did document tendons they re-grouted, and the yellow arrow is pointing to the first level of grout, and then the blue arrow is pointing to the second level of grouting they did. So, they took extreme care to grout these tendons really, really well. On the interior, we were able to notice the diaphragm cracking at the piers, which isn't unusual for a bridge of this type. We'll talk about the web cracking, and then the moisture that we saw at the deck anchors, but the most important thing was we saw no leakage of the joints, which was really important, because when you've got post-tensioning crossing these joints, that's certainly a problem to be concerned about. Final view of the interior, you can see the web cracks from the cantilever tendons. Here's a close-up of one of those, and you can see that moisture stain up at the top, and that was something we were looking hard at, and we did do some testing of this concrete. We removed some of this concrete. There was a little higher moisture content, but there was no corrosion associated with that moisture, and so that was an important takeaway from this. Once again, just another shot, and you can see the epoxy that was squeezed out on the inside of the segment joints. On the exterior, once again, we're looking closely at those web cracks along the tendons to confirm that the epoxy seals were intact. We did observe some spalling of some of the end-tendon anchors at the expansion joint, and then there's obviously some corrosion spalling, primarily due to low concrete clear cover. Here's a couple of slides. Here's one of those web cracks. No extension of this crack. Some of them extended slightly, and we think that's just because they didn't seal it completely. This crack here on the upper right of the image is not associated with the corrosion on the left. The corrosion on the left is actually a mesh that they added after shop drawings had been prepared, and it was not included in the shop drawings, but they clearly didn't have sufficient clear cover to protect it from corrosion. All this is black steel again. Here's one of the bottom slab anchors at the end underneath the expansion joint. This is probably the worst area we saw, where we were getting some corrosion of the anchor plate itself. Immediately inside the tendon, we didn't see any corrosion, but we saw some corrosion to the anchor plate. Here's the inside view of a similar location, and another one. You can see we've got a little bit going on underneath the expansion joints, which isn't surprising. We can also see the rebar chair feet as they contacted the surface. All in all, the bridge was doing pretty well. Here's another example of some of that mesh that's corroding due to very, very low clear cover. The lessons learned on this, though, is that, and it's not any surprise, there's no substitute for clear cover, because where we had adequate clear cover, even with the black rebar, we were not getting measurable corrosion rates. We obviously need to protect the anchors at the ends near the expansion joints, and that's an issue that the industry is still working on, trying to develop methods to do that. At the time this bridge was built, and in the 20 years after it was built, there was some debate in the industry about whether or not you need epoxy for the faces of precast segments, and if you do need it, do you need it on both faces? Well, the answer is absolutely yes, and that's in the current code. We need to pay attention to grouting, and they certainly did on this bridge, and I'll show you some slides in the next case study where they did not. The deck anchors for the post-tensioning system could potentially cause a problem, but right now they're not. One of the things we did in the rehabilitation of this is we're putting on a polymer concrete overlay that will provide a much better waterproofing system than the asphalt overlay that was on the bridge. And it's our opinion that this post-tensioning system will easily get this bridge to 80 plus years. Switching over to a couple of newer bridges, 1986 construction and 1994 construction, both with very similar designs. And I've titled this slide construction inadequacies, but really it's going to be, the subtitle is poor grouting. The San Antonio project was roughly 16 miles of segmental box girders built between 1985 and 1991 or 2. It primarily uses a lot of continuous spans with a combination of internal and external tendons. The grout used at the time was a cement and water grout with an expanding admixture, and I'll talk about that in a bit. The Austin project was built, it's about six miles of box girder, built almost a decade later, but with simple span design for some reasons that I'll explain in a few slides. Still using a combination of internal and external tendons and the same grout specifications were carried forward from the previous project. We didn't make any changes to those. So why use the expanding admixture? Because we certainly don't use them in the current grouts in the M55 specification other than to control expansion and shrinkage of the grout, but we don't typically put a large quantity of an expanding admixture in there. Well at the time there were some reports from the 60s and early 70s that you could get up to 15% bleed in a vertical tendon, and it's not as well known or defined, but we could expect plus or minus three percent for a horizontal tendon. And then the shrinkage for a neat cement grout is not unknown, and so the concern was that those two processes, the bleeding and the shrinkage, would leave voids in the duct. So they decided to put in an expanding admixture and it was specified to get between two and four percent expansion. I can tell you we had trouble keeping a uniform dosage rate to meet that specification, and the admixture itself was a gas forming aluminum powder, which in itself has some challenges and can cause some problems. The bottom line though is that it did not work as intended. First of all, we often got well more than four percent expansion and that ended up splitting a lot of the ducts, not early on but certainly later on, and I'll show that in a slide. We also ended up with some foamy or highly porous grout at high points. If you've got air or gas bubbles in your grout, it's going to find its way to the high point with the bleed water, and so not only did we get bleed water, then we had foamy grout within the bleed water, and so that wasn't a great thing. And most importantly, we still had voids. The grout still bled because the gas forming expanding admixture doesn't have any impact on bleed. So here's what split ducts look like, and they were fairly commonly found in both bridges. So if you can imagine, if we put this expanding grout in there and it expands as the grout begins to set up, that induces some radial stresses in the ducts, and those stresses are locked in. And then the problem is when an HDPE duct gets cold, it becomes a little brittle. There's a term called environmental stress crack rating, and if that's not specified correctly, they can split. And those splits can often occur many years after construction, so it may take a while for them to show up. It just depends on how many cold weather cycles you get and how much expansion you had in the duct to start with. So let's talk a little bit about some of the details before we get into the grout voids. So in the San Antonio project, we had basically two similar or different designs, but a similar mindset, and that is to use continuous spans to make a two to seven span continuous unit, and that eliminated the number of joints, obviously, and reduced the exposure of the end anchors to joint leakage and that sort of thing. But we use two different methods. So the top one is a system where we've got overlapping tendons over a pier with two different closure pores on either side of the pier segments, and then in the bottom we've got a single closure pore over the pier with some simple span tendons, and then we've got a continuous or sometimes coupled tendon to make the spans continuous. Cross sections of these are a little bit crude, but we had six projects on the San Antonio segmental bridges, and so six different designs, but they were largely grouped into three groups. So we had one group on the top left where all of the external tendons or all the drape tendons were external, and that was one of the earlier designs. In the bottom middle was kind of a next generation where they used external tendons for the drape tendons again, but we had some of them that were coupled and some that were simple span. And then the last generation on the right was a combination of draped internal and external tendons, and the thought process behind that one was the internal tendons give you a lot better ductility and we could actually reduce the number of strands to a degree, but obviously we've got some construction challenges about putting draped tendons in the web, so those all had to get worked out. In the Austin project, we took a different approach. We decided we were going to go with all simple spans, but we're going to make them into continuous units of usually two to three spans using that little link slab at the top, and I'll show you a detail of that. And it's a surprisingly simple concept and it worked really, really, really well. One thing it did was it simplified our tendon profiles at the pier segments and so it made their casting a lot more easily accomplished. It still was able to reduce the expansion joints and we were able to protect the tendon anchors over every pier with that link slab. So in these designs, because we're dealing with simple spans, we ended up with more bottom slab straight tendons, and in the end we used roughly 10% more strands than the continuous design from the San Antonio project, but because it simplified pier segment casting and some other operations, it actually got us a lower cost per square foot for the entire bridge. So here's what that link slab detail looks like over the spine, and so it is just very simply a cast-in-place concrete slab. We had a precast option in the plans but they elected not to use it, so they went with a precast, excuse me, a cast-in-place, and then out over the wings it's a similar detail but slightly smaller gap. And here's what it looks like on the inside. The black rectangle or squares are pourback covers in grout or concrete, and then the metal decking you see on the top is forming that link slab. And this picture was taken not too long ago. You can see we're not getting any water leakage through these. They're not having any cracking. They're working really, really well. So back down to San Antonio in 2002, they did a full length of the 16 mile box inspection of the PT system, of the external system I must say. So there's roughly 5,300 tendons, and they found voids in about 60 percent of those. And so on the left you could see they obviously knew they had a void during construction. They pookied it up with some nice looking epoxy. The bore scope taken on the right was from 2002, and that's not an atypical condition where you see partial coating of the strands with grout, and where you even had a partial coating, they seem to be pretty well protected. We also found roughly 84 tendons without any grout at all or with very minimal grout. This is one of them, and you can see we had a little bit of grout in the bottom of the deck. I have no idea where it went. They obviously tried to pump some grout through there, but didn't quite make it. But that's a large number of tendons with no grout, and you can see the condition of the strands. This was roughly 20 years after the bridge had been put in service, so not horrible, but we certainly need to do something about it. And the one thing I wanted to point out, and this is another example, this one had a little better coating of grout on the strands, and it didn't take much to protect those strands. Now keep in mind this is San Antonio, Texas, so it's not a place where they use a lot of de-icing salts, but it is warm and it is humid, so that can impact our rate of corrosion. We did have some pretty ugly looking ones, not very many, and this is one of the worst. But even when we cleaned off those strands with a Scotch-Brite pad, they cleaned up fairly decently, so in some cases it almost looks worse than it is. And then in the following year, they did another assessment, a similar assessment in the Austin project, and they found even more external tendons with voids, almost 80 percent. A little bit fewer 20 ducts with no grout, but keep in mind this project was six miles of box girder, San Antonio was 16, so if you ratio it up, it's about the same with no grout. There was still some corrosion of the exposed strands, and it ranged from no corrosion to moderate. I'll show you a few slides of what that looks like. A bore scope, once again very similar to the other picture I showed, that if you get that fine, in fact this may be the same picture, you get that fine coating of grout on there, it actually can protect the strands. Here's another example, this time there's less corrosion visible on the strands. And I have to point out that because we're shoving a bore scope through some really small openings, sometimes the bore scope twists and up is no longer up, and so you have to kind of think about the orientation of the camera. This is in the trumpet region, so up near an anchor. We've got a coating of grout, but once again no corrosion, and this one had very little grout in it, but also no corrosion. But not so this one, and keep in mind we're looking at a bore scope, so it does tend to magnify what you're seeing on the screen. So for the San Antonio project, TxDOT tested the grout in 2011. They went to 40 different locations and opened up the tendons, and they found obviously, and not unexpected, some non-uniform properties of the grout, some stratification, this transition from a good solid gray grout to this kind of what we would call a punky grout with a lot of air voids, and then sometimes a with a lot of air voids, and then sometimes a very white chalky grout. They found large air voids in the grout samples due to the expanding admixture. We discovered that using petrography of the grout samples. There were no chlorides and very little carbonation detected in the grout samples, and we sampled and tested the grout, and the pH of all the samples was greater than 11. Even the really poor quality grout had a pH greater than 11, and at the locations we were looking, and we were looking at really bad locations, there were no outbreaks. So they re-grouted this project from 2014 through 2016 using the pre-bag grouts that are specified in an M55 grouting specification, and they used a vacuum-assisted grouting process, and once again, the plans were set up to expect 60% voids in the tendons, but we didn't look at all of them obviously at the time, so this was what the contractor was told to expect, and then 35% we suspected had major voids, and another 25% had smaller voids. They also repaired all the split ducts. There were some broken drain pipes and some other hardware inside that needed to be addressed, but the most important thing was they opened up every single accessible anchor and bore scoped it before they did any work, and so the end bill, if you will, was roughly 20 million dollars, and I know the folks that did the re-grouting and did some of the other work, and they spoke of the really difficult access. I mentioned that some of these units are two to seven spans long. Sometimes they only have an access hatch at the very end, so if you imagine an access hatch 700 feet apart with seven spans or six spans, you've got some challenges getting your people and equipment in there. So just this year, we went back into the San Antonio Bridge to verify the quality of the re-grouting process and to verify that it has worked, and so we went back to 24 of the original 40 locations and opened them back up. This time we're still using a bore scope, although there weren't very many voids to bore scope, and we did some half-cell potential testing of both the new and the old grout to see if there were any differences in that potential based on the pH or chemistry of the new grout versus the old grout, and we determined that any existing corrosion that we observed almost 20 years ago has not continued, and there's been no new corrosion, and so we're characterizing this as a really successful re-grouting process project. So here's a couple of slides of the assessment itself, and so here we're using a wire saw to cut the duct. Once we cut a section of the duct off, we slide it out of the way. We would do a phenolphthalein test, usually top and bottom, trying to capture the top new grout and the bottom old grout. If that was the case, this image is not, but in some cases it was. Here we're doing half-cell potential. Our lead is on the middle of the tendon out at mid-span, and so we ran some wires to do that. We did chip off the grout and sample it and retest it, both the old grout and the new grout, and then we slid the sleeve back over it. We actually mixed up some grout and filled that void left by our grout removal operations, and then we wrapped it with this really thick duct repair tape. It's not your Home Depot duct tape. It's a really high-quality duct tape, about $500 a roll. So that's the process we followed to do the San Antonio inspection. In Austin, we also went back in this spring. We're looking at 12 locations. Once again, the same locations from that 2003 inspection. This time, because they haven't re-grouted in Austin, we're basically just doing visual inspections with a bore scope to see if we can tell if the corrosion has progressed from 2003. So this is one of their original openings, and they did use a Home Depot duct tape to seal it up. So less than ideal, because it does allow moisture to penetrate. But we were able to compare these photos to the older photos, and in most cases, we didn't see a lot that was of concern. Here's one of our openings that we drilled in, and you can see the condition of the strands, just a few little spots of surface corrosion. Here's another one. This was one of the original 2003 openings that we've reopened. Once again, the strands are in like new condition. This is one of the 2003 openings, but there's been no additional corrosion. This was what it looked like similarly in 2003. So not a significant change. And then here's another case where the strands have been coated in some type of punky-looking grout, but still adequately protected from corrosion. So the summary of at least the work we did in Austin and San Antonio is that voids can be re-grouted using the current grouts. You do want to test the original grout for pH to make sure that it has the chance to be compatible with the new grouts. And in the case in San Antonio, it was, and in the case in Austin, it will be as well. Vacuum or vacuum-assisted grouting grouting is going to be needed to fill those voids adequately. You're really going to want to follow the N55 grout specs as far as mixing and placing, installing the grout. And the primary reason for that is to avoid the formation or development of soft grout. If you try to overwater even a pre-bagged high-quality grout, you will form soft grout, and that is not something you want to deal with. So once again, lessons learned from these two projects are, yeah, we still need to protect the anchors. The designs on these two bridges did that in two different ways. We certainly need to pay more attention to grouting. We got a little lax, it looks like, on those two projects. I say we. I was actually involved in some portions of the construction of both projects. The simple designs are easier to inspect and repair. So that's something I think designers might want to keep in mind is providing access to the interiors and making things easy for inspectors, easy for those that are going to be doing the repair work going forward. So now I'm going to switch to the cool stuff, the fire damage. Now when most people think of fires on or under bridges, this is what you might think of. In our case, we're talking about fires inside the bridges. There are at least three reported locations, two in San Antonio, one in Austin, all the same bridges that we've been talking about. And trespassers entered the boxes through access hatches, and the yellow circle on the right shows a little rope that they had rigged up to open the hatch. Textile did have a lock on the hatch, and they cut it off and put their own lock on there, but that's how they were getting inside there. Only one of the fires was actually reported to the fire department. The other two fires were noticed during subsequent inspections. No reported injuries on any of the three fires and no tendon damage, at least in the Austin bridge. So the first San Antonio span was a ramp span, and there was traffic on the span during the fire. This is one where the fire department was called because they saw smoke billowing out of the access hatch. And I'll show you what the damage looked like, but there was significant damage to the main external tendons. There were lots of other materials or equipment in the span. I think they may have had an air conditioner in there that was wired into the bridge lighting system. From a resilience standpoint though, even though they're going to lose a significant number of their main structural tendons, there was no cracking and there was no deflection observed to this span. So here's a couple of inside shots. This is the damage to the top slab. This is a pre-stressed, transversely pre-stressed top slab. And so that was a big concern structurally because there's really no way to recast that and kind of restore the pre-stressing. The two yellow ovals show where the tendons used to be because they're not there anymore. Here's another shot of some of the failed tendons. A couple of the wires or strands did not break, but by and large, almost all of them broke or relaxed significantly. Here's an elevation of this span. This is one of the designs that used the overlapping and continuous tendons over the pier. And you can see we've got the draped external tendons, T1s and T2s, and you've got some internal tendons in the webs, plus a few top and bottom slabs. And here's what it looks like in the cross section. So we've had two draped tendons in each web, plus two external draped tendons in the void. Those are what were missing or gone from an analysis standpoint. And so that was a bit of a problem to try to figure out how to replace four draped external tendons. Plus the big challenge was the top slab damage to the pre-stressing because those strands were at seven inches and they were a very prominent portion of that structural system. So ultimately though, they decided they needed to replace this span, primarily due to the damage to the pre-stressed top slab. But they did lose half of the draped tendons. Because this was part of a four-span continuous unit, the analysis had to consider the impact of removing that span on the other three. And they've got a solution for that. I think they're gonna replace the tendon. Or no, that's the other span. They're gonna replace the span with pre-stressed girders and they may have to do some retrofit post-tensioning in the remaining now new end span. But I wanna point out though that despite they lost half their draped tendons, there was still no cracking or deflection observed or measured in that span. I think that was amazing. The second span, also in San Antonio, also a ramp span. This fire was never noticed. And it turns out when we went in it, there were actually three different fires in three different locations. And it's possible traffic was on the span for who knows how many months after this fire occurred. Fire department was never called. This is a shot of the interior on the right. And they did lose one draped external tendon. The two on the left with the green checks, we determined were not of concern. But the one on the right with the yellow check was the one we were looking the hardest at because it was also in the vicinity of the fire, which occurred largely right underneath where that X is, where the strands snapped. This one had minimal damage to the concrete. And you'll see some grout removed in some of the photos. That was all done during post-fire cleanup to check the strands. None of the grout spalled off due to the fire. And similar to the other span, there was no cracking or no deflection noted, measured, observed, anything like that. And we were looking really closely at this one. So we shored the bridge to leave it open to traffic while we did the assessment and then while repairs are gonna be implemented. So we went inside under the short condition to do a detailed concrete assessment. We're looking at the grout. We're looking at the strands. We're gonna test and sample the duct. We're looking at every one of the joints to see if the epoxy was still in the joints because the epoxy's melting temperature is around 170 degrees. And that's not very high when you're considering a fire. And like I said, we actually discovered that there were three fire locations. So I'll show you what a few of those look like. Here's one of the smaller fires. It was actually in the end segment near the abutment. It didn't do a lot of damage. It just created a lot of soot. On the lower left, you can see the severed tendon. It didn't sever in this region. It severed where the larger fire was. But you can see what it looks like when it's severed. This is the tendon with the yellow check. It's the one above the fire region. And we knocked those grout samples off. And you can see the cracking in the grout. But you can also see the grout, and this was facing the fire, fortunately had pretty good cover over the strands. And we think that that did a lot to protect the strands inside the duct. This was the same tendon a little further down. You can see how the duct had melted here or something like that. The lower duct is actually the one that severed and snapped back to the right of the image. And the metal hardware is a folding table that got caught up in the fire. This is a close-up of the fractured strands at the break location. And the green arrow is pointing to a strand where it necked down pretty significantly, but the yellow arrow is pointing to a pretty clean break, a break under stress rather than under relaxation, right? So as the strands necked down and released and snapped, it transferred the remaining load to the right of the image. It transferred the remaining load to the other strands, and they failed more brittly and more suddenly. And so we were able to take a lot of samples. We took actually 23 strand samples for testing, and I'll talk about that in a little bit. So from an assessment standpoint, we were doing, for the concrete testing, we looked at using a rebound hammer just to see if we could characterize the impact of the fires on the concrete surface properties. We did take some course for petrography to be able to determine the kind of the depth of the impact of the fire. We did a lot of sounding, which isn't much different than a rebound hammer. The rebound hammer just gives you a number. For the grout, we took a lot of samples for petrography. We also sampled the grout from the yellow-checked tendon further down the span where it was not affected by the fire. And those, we were able to heat soak in an oven so that we could compare various levels of heating on the grout properties to give our petrographer kind of a fingerprint of what he might be seeing when he's looking at those heat-affected grout samples. We tested the failed strands. We tested the duct samples, and I'll show what that looks like here in a bit. So here's one of the grout samples we took, and you can see the discoloration of the grout at the bottom, and I'll show a close-up of that. Here, we're coring into the wall. Had to use a GPR to make sure we don't hit any rebar. Rebound hammer, you can see we're dealing with about a four-foot, 10-inch high box, and James there is 6'3". He was a little sore at the end of the day. Here, we're removing strands, and like I said, I'll explain the testing we did on the strands. And here's a close-up of one of those grout samples, and the red line kind of shows the line of demarcation of the heat-affected grout at the upper portion and the more normal or non-heat-affected grout in the bottom portion. And once again, we were able to characterize that. I'll show a slide or two of that later using the scanning electron microscope and some other imaging techniques. This is a sample of one of the concrete cores. You can see the microcracks from the paste dehydration due to the fire, but we had minimal depth of impact of concrete, roughly an eighth of an inch. These were not super-hot fires. They were not super-large, and that works in our favor. From the grout petrography standpoint, once again, there's one of the grout samples on the right, and then our petrographers under the polarized light were able to see the impact of the heat on the calcium hydroxide crystals left in the cement paste or the grout paste. And so they were able to determine, based on our heat-soaking samples, that we got to approximately 850 Fahrenheit up to a half an inch in some grout samples. Now, I mentioned Fahrenheit because later we present things in Celsius, and unfortunately we had to convert back and forth between units because some of the references are in various units. So in the strand testing, we basically just did an ASTM, I think it's 416, test on the strands, which is the test for new strand to evaluate its yield point, its ultimate strength, and we compared all those to the reference value, which is in the specification. You see, we're usually below the reference value, but in most cases, not by a lot. There was a couple of them, Tendon 6, for example, or Test 6. It was pretty well below both reference values, but it also was the strand that we suspect was closest to the fire, so it was probably most affected. One thing to point out is every one of these tests failed in the grips or very near the grips, closest to where they frailed in the bridge. And so that gave us some confidence that the remainder of the strands were likely in better shape, and we could count on their properties. The chart on the right is just a comparison of the calculated level of stress on the strand versus the temperature, which you might expect that strand to fail. And we've got, those are all the references that were used to develop that chart. And then similarly, you can correlate exposure to temperature to relaxation of the strands due to that elevated temperature. And so what we're looking at, at an assumed exposure temperature, and this is extremely worst case of 450 degrees Celsius, that we might get almost 88% of, or 12% relaxation of the strands from their original condition. So this might affect just that one tendon that we checked in yellow in one of the earlier images. We also heated up some duck samples. We pull these out of the bridge at various temperatures to just kind of see what they would look like at various exposure temperatures. The surprising thing we found was on the right was when you lit it with a match, it stays ignited. And that might explain some of the duck conditions we saw where there wasn't a lot of grout damage, but the duck burned away. So from a finding standpoint, the concrete was relatively unaffected. Exposure temperature was roughly 850 at the very worst case. And while some strands were exposed to the fire, others certainly were not. We were not concerned about ultimate strength, but we did have some possible loss in the residual stress in some of the strands on one tendon. And like I said, the left side tendons in that one picture were not exposed directly to the fire. We're not that concerned about them. So they're gonna keep the span. They're gonna replace the failed tendon. And they will be doing some refined analysis on the remaining tendons to make sure they can adequately account for that heat relaxation that occurred due to the fire on the one tendon. So now let's go into some demolition stuff. And I'm gonna go a little bit quickly because I'm running long. This is one of the Austin spans, the same bridge we've been talking about. They had to reconfigure the bridge. And so we had to remove four spans. They found it was gonna be the quickest and cheapest to do explosive demolition on three of them, but they had to do or wanted to do conventional demo on the other, primarily so that we could keep some samples so that the University of Texas can study them. So those are the samples out in the CBEI yard at the Pickle Research Campus in North Austin. Same designs that I presented earlier with the link slab. On the right, you see for this span, we had three rather large bottom flange tendons or bottom slab tendons, and then either two or three external drape tendons. And so for the explosive spans, we cut them where we wanted to fracture them where the two blue arrows are. We needed to protect the piers. And so we wanted those N3 segments to literally fold over onto itself when we dropped it. And so to do that, we had to weaken the concrete section. And so in the lower image, everything in dark gray is concrete we removed. And so you can see how little of the remaining section we had left, but we did the analysis to determine it was still okay to remove that much concrete from the section and still have it remain upright and stable. On the image on the left, you can see we removed a good chunk of the wing and the wing tendon. They had an access hole in the top slab so they could get inside there and put the charges. The image on the right, the red arrows are pointing to the charges that sever the tendons. So we've got charges on the, in this case, three or six, excuse me, draped external tendons and the three bottom slab tendons. And then the yellow arrows are pointing to some holes we drilled into the webs and they put some small charges in there. We needed to fracture the webs so that they would release as well, but we didn't put any charges in the top slab because that's gonna be our hinge. So here's what it looks like when it blows up. It's always nice to have a plan go as you want it to. In conjunction with that demolition though, we also took out the external draped tendons for the last span that we were gonna do conventional demolition on. We didn't wanna run the risk of trying to have folks in the box, trying to cut those tendons with saws or anything like that. And so we ran the analysis and we figured out we could lose all of the draped external tendons and the structure would remain up. And then this was our demolition scheme for this last span where we removed the wings for those middle six segments. And then we were gonna have them kind of chip out this little wedge right there in the middle in the bottom image. And once they weaken the concrete compression zone, it would fall and we had a pile of soil underneath to catch it to protect the strands or the tendons and the segments, I'm sorry, excuse me. Because the ones on the left in yellow are the ones we're gonna keep. So here's what it looked like. Those are the tendons that were removed with the explosives but the rest were remaining. The image on the right, you can see one of those tendons released. All of them were released but only one of them kind of jumped a little bit. Didn't look like it had a lot of grout in it. And then on the left, you see the wing removal limits on the upper image and then how much concrete we were moving at a given time. And when they got to step A in the bottom image, that's the slides I'm gonna show you in the next couple of slides. And we wanted them to start with those yellow trapezoids or polygons and work their way downward. They didn't actually do that. This was before they started chipping. For some reason, the guys operating the whole ramps it was real easy to chip through the webs. We kept telling them, that's not where it needs to be weakened. And so it took them a little while to start working on the top. But you can see how much concrete they're removing. And at this point, this whole span is still being held up by just three tendons. And it's still not opening up the joints. It's still not deflecting. But finally, they got enough chipped out on the top that it was able to land and they were able to start removing the segments. And to me, this just shows just how tough these bridges are and provides a really good example of the resilience of post-tension box girder bridges. So that's the end of the presentation. I finished just barely in time. And I think I'll turn this back over to Kyle. Yeah, absolutely. Thanks, that was great. I think my takeaway is the only way to get rid of a bridge now is to blow it up. I recommend it. All right, well, we don't have much time. We do have a handful of questions. I'm just gonna take one of the questions. It has to do with when you're drilling into the duct to do that bore scope, do you have a recommended repair just to temporarily go back and patch that duct up prior to any re-grouting or anything along those lines? Yeah, what we did, we were actually looking at a number of duct repair techniques. And so we were using a hole saw that we used when we cut a hole in the duct and we kept the little disc that comes out with the hole saw. And when we repaired it, we put a little bit of caulk around the perimeter of it to kind of hold it in place because it's now a little bit smaller than the opening. And then I don't recall the name of the duct repair material, but we contacted one of the duct suppliers and we were actually looking for heat shrink tubing or fabric that we could use. And they recommended, it's a really thick, probably 50 to 75 mil, very adhesive plastic duct that we wrapped around it to seal it up. And it provides a really robust repair. And I do think it will survive. If you need a little help with a vacuum or anything like that you could put a couple of steel band clamps over it and I think it would do the job for you. Awesome. We have a few other questions about grout, how it's changed over the years and both from the material standpoint and also from the installation standpoint, but we are out of time. So for those that do have questions, you can see on the previous slide that had the information that you can reach out to us. And we need to quickly talk about the next few webinars that we are going to be having over the next three months. Like we always do, we present the next three months worth of webinars that are coming up. In July, it's going to be the valuation and repair of unbonded post-tension concrete structures. So we're really going into repair procedures and have a look at it. This is a popular concept, especially in regions of high corrosion there or areas where there's retrofits. A lot of good technical information there. In August, we're going back to buildings and we're looking at some advancements in the design of elevated unbonded post-tensioning and that's going to really a banded-banded layout instead of a banded uniform layout on there. Then September, we're going to go to PT bar considerations and small alternatives, which are some new tech notes that are being published right now. And so we're kind of bouncing around again, away from buildings into buildings, then back away there. But we do have a pretty good lineup for the next three months. We look forward to seeing everybody again this same time, same date, second Wednesday of the month. And with that, enjoy the rest of your guys' day. Thanks for chiming in.

Post-Tensioning Institute

monthly webinar

resilience

post-tension box girders

bridges

corrosion

grouting challenges

fire damage incidents

demolition techniques