Mass Timber Multi-Storey UAMS Build Diary – Lessons Learned

In this aerial view, you see the two buildings – building A to the north, L-shaped, and building B to the south, a C-shaped that are linked. They’re each five stories tall and they’re linked in the middle by this one-story cabin with the major social spaces. It’s worth noting there was significant topography on the site, a 25-foot change from the north to the south, and that kind of fed into some of the construction logistics as well.

So since I’m the architect on this panel, I have to show you some of the pretty pictures. These are the completed project. So here you have the view from Stadium Drive that passes to the east of the complex. And as I said before, celebrating wood is really a key design goal. So what we’re really proud of here was the ability to expose structural timber as much as possible and marry that with locally sourced finished wood.

qAt the street level, you’re seeing exposed glue glam trusses and columns. Here in the cabin social space, you’re seeing the exposed CLT ceilings in the residential floors above. And you’re seeing exposed locally sourced Arkansas cypress that creates the finished soffit and highlights points of entry.

Here again at the central courtyard, looking into the cabin space, you’re seeing that wood-on display is a major design element. And during the day as well. So just again to point out what you’re seeing here, this is glulam columns, glue and beams and trusses. This is exposed CLT slab here on the edges combined with Arkansas cypress uses finished material on the walls, almond’s wood grill in the ceiling to conceal the major mechanical spaces.

Moving on down the site through the project, a series of major academic spaces cascade down the hill. This is a performance space. And one thing that I’ll come back to this later, but we actually were able to use some of the salvage CLT. So really a CLT skid that’s a delivery system for glulam and CLT, and recreate that in the custom furniture on the site. So these bench tops here are actually salvage CLT from the process. And here in the woodshop area, that same CLT skid was used to create custom tables that were part of the workshop space.

Again, that one thing that we’re really proud of here was we have some beautiful exposed columns and beams and a really pristine CLT slab up above that we worked really hard on to keep this clean so we could expose it in the major public areas and in the student residential spaces above. Again, the intent was to have wood and natural materials exposed in every room.

So with that kind of background, I wanted to take you back in the project and share some strategies that we use to anticipate construction hurdles. There’s really five key takeaways here. Some of the things that made this project successful were that we created a shared foundation of knowledge, that we made cost-effective design choices, that we evaluated code issues early, considered species and fabricator differences, and leveraged the pre-fabrication potential of mass timber.

So I wanted to start with this one. One of the most impactful early steps on this project was to create a shared foundation of knowledge through a site visit that included mass timber building precedence and actually visiting CLT fabrication sites in person.

I mean, this group of people you’re seeing moving through here includes the design team, but it also includes key decision makers in terms of the client side, the University of Arkansas housing and University of Arkansas architecture department, as well as the CM itself. So this is really bringing everybody together to understand what some of the challenges and potentials of this fabrication method might mean.

So then making cost-effective design choices, the client really initiated the mass timber interest on this project, but we were trying to create a design that was cost effective and that could be realizable in either steel or wood. So we developed the structure on parallel tracks through schematic design with cost estimates really at every phase to compare conventional steel, excuse me, conventional seal concrete construction to mass timber construction.

One of the key decisions in making mass timber cost effective here was looking carefully at the student room and, together with the overall structural modules. So we looking here and thinking about the the typical eight foot by forty foot CLT panel available in North America, we really worked back and forth between program and structural modules to make sure that we were creating kind of cost-effective decisions at every point that satisfied the program as well.

So moving here from the typical student room to the typical residential wing, where again you’re seeing the, you’re seeing how that resolves into an 8 by 40 CLT panel, how we’re able to effectively use that modular dimension, minimizing waste and kind of maximizing the structural potential of that.

Another key strategy was to evaluate a wide range of structural frameworks. So you’re kind of illustrating and honing down on where the optimal choices are in math timber. So we examined mass timber with a variety of core and shear wall structures and compared that to steel. And what you’re seeing here is that the residential wing that we developed took advantage of structural efficiencies in mass timber with one-way beams as opposed to a two-way steel structure with composite deck. So there were some efficiencies to be found there.

Another key takeaway was to evaluate code issues early. So we met early and often with the University and state fire marshals, beginning in pre-design. And this project would be a pioneering use of mass timber in the state for a student residence hall. And we wanted to make sure that everyone was on board with and understood, and that all concerns were kind of vetted early in the process.

So the local fire marshal made it clear that they would not entertain any variances for this project which ended up influencing key decisions around construction type and concealed spaces. Again we use the kind of comparing to conventional structure back and forth um to help drive decisions and illustrate points about construction type. Here the massing the area and the desired room count led us towards a five-story building we illustrated what that would look like in a steel frame building type 2a protected non-combustible construction one.

The local fire marshal made it clear that they would not entertain any variances for this project, which ended up influencing key decisions around construction type and concealed spaces. Again, we used the kind of comparing to conventional structure back and forth to help drive decisions and illustrate points about construction type.

Here the massing, the area, and the desired room count led us towards a five-story building. We illustrated what that would look like in a steel frame building, type 2a protected non-combustible construction, one hour fire rated structural frame. And then compared that kind of apples to apples with an advanced timber frame building, type 3b non-combustible exterior walls but unprotected combustible structural frame, floors, ceilings and roof.

One question that frequently comes up here is, you know, why not type 4? Isn’t that the natural fit for mass timber? But again, we understood that the client, the university, did not want to expose mechanical systems in student areas for reasons of protection and longevity and maintenance. So we knew that there would be some drop ceilings and concealed ceilings in these spaces. So again, we were looking for construction types that would allow that flexibility.

Early conceptual details again identified key code issues such as the separation of a combustible structure from a non-combustible exterior, and understanding how a glulam column can form part of a rated partition and how that non-combustible exterior enclosure meets that rated partition. And so again, this is very early in the pre-design phase, but just to try to identify as much as possible any of those issues that would need further vetting.

At the end of schematic design, the client made the decision to continue in mass timber. And at about fifty percent CD, the mass timber package was awarded early. Bender Holtz and Holtzback won the contract, and this kicked off a design assist phase to fully detail the timber and coordinate with the specifics of Bayer Wood Products. This is an important phase because there were several critical changes that needed to be understood and addressed.

A key takeaway is the importance of considering species and fabricator differences as you’re making your kind of final selections. Among the key differences that this design assist phase really unearthed were differences in the typical member size.

So the basis of design was a Doug fir caught dark fur glulam beam, 14 and a quarter inches wide and 18 inches deep. The basis of construction beam was a spruce pine fur glue lamb, again 14 and a quarter inches wide but 20 and a half inches deep. And this was largely due in differences in species strength and European versus North American standards.

Some people say, okay, you know, two and a half inches, is that a big deal? And in most cases, no. But in one very specific instance, this was a big deal. There was a key pitch point that happened in each of the typical student rooms. You’re seeing kind of a reflected ceiling plan mechanical diagram on your right and an axon of the typical seal of the typical student room on your left.

And the key issue here is there is one point where a supply duct needed to dive underneath the glulam beam in order to kind of feed this student wall here. So that was the one place we were most constrained in overall ceiling height. I mean, that was the one place where this two and a half inches really was a major issue.

So the compromise solution was choosing a beam that was a little bit fatter, understanding, you know, as Greg just highlighted, that might have some cost implications there, but was significantly less deep. So we could continue with the kind of existing mechanical system and that coordination, and avoid, avoid any negative impact on the ceiling height that would really, you know, ripple through all of the student rooms.

Another decision was the issue of factory installed cast beam hangers versus field installed beam hangers. So our basis of design had originally called for factory installed beam hangers with tight fitting male female sliding or sherpa connections. And given the size of the project, Bender Holtz had recommended substituting a bearing bracket that you’re seeing here for interior connections, and a tight fitting pin connection for perimeter connections.

These connectors are both field installed but do allow for greater adjustability, which ended up being important on a project of this size. So ultimately, I think this field adjustment was important, but there was a significant addition to the field labor associated with the change, which I think again, a big room that can help you vet through all the implications of those decisions is really a key part of the communication process.

The fabricator had also recommended substituting multi-story columns, up to 40 feet long, in lieu of the single story columns that the design had originally specified. And this change represented a savings in overall material and also in kind of effort of installation, fewer connections, but required a thoughtful re-detailing of the column to floor connection to ensure that, number one, the fire separation between floors is maintained. It was just a 30 minute separation. And two, to protect the continuous column from the moisture of the wet pour creek topping that was on top of each CLT slab.

The last piece here was that the design assist phase is an opportunity to leverage the pre-fabrication potential of mass timber. Think back to all those student bathrooms, one bathroom for every double student room, and how that kind of ripples through the entire project. So there was a combined effort to coordinate and factory drill all CLT, fabricall CLT penetrations. And that really involved the design team, it involved the CM, and it involved the fabricator.

We were reconciling the design model, the contractors coordination model, and the fabricators production model, trying to make sure that we all had all the holes throughout the project in the same place. So there were 41,113 penetrations in total. The CM called it the “swiss cheese slab”, but the key point here is that each penetration was located, coordinated, reviewed for clash detection, and cored in the factory before the timber was delivered to the site.

Looking forward in time, that these pre-drilled penetrations led to a lot of time savings in the actual installation of the plumbing. It expedited insulation, the plumbing, the mechanical systems in the field. But it did, this coordination process did require more input from the overall design team at a key point in time than in conventional construction. So I think one of the takeaways here is just thinking of that coordination process and thinking of all the people who need to be involved for it to be effective, and to build that into the schedule from the onset.

So as much as I would love to say that we successfully anticipated all the construction hurdles and Adohe, we did not. I wanted to highlight some key lessons from the field about addressing construction hurdles as they arise during the process of construction. So the key takeaway here, key takeaways here are really about protecting exposed structure, accommodating field tolerances, and addressing necessary field modifications.

Going back to the timeline, with the early release of the timber package, the timber fabrication transport overlapped with early sight work and concrete work in the field. And you’re really seeing that here in this ariel that shows the the overall construction sequence which moved from the south to the north of the site. It allowed sector by sector assembly of the mass timber and piggy backing of trades. But you see kind of before all that, these concrete cores, that were the key shear cores, preceded the entire timber installation.

This is a photo from the in progress timber insulation of the first wing. And a key lesson here was protect exposed structure, especially when that structure is moisture and light sensitive and will be exposed as a finished material. Both the CLT and the glue lam arrived wrapped from the factory, and our initial guidance had been to leave those factory wrap protections in place as much as possible during assembly in order to, you know, to preserve that protection layer that was already there.

However, what we found in the field was that columns are being partially unwrapped in order to install connector brackets, leading to uneven discoloration due to UV light exposure, you know, up and down the height of the column at the ground floor. We actually found that some of these factory wrappings were actually collecting, accumulating water at the base of them here, as you see in this inset picture.

So with with these issues in mind, and in consultation with the CM, we reverse course and recommended that each timber element be unwrapped at installation. And just noting here, the contractor team also installed bump protection around the ground floor columns to protect them from equipment damage during construction.

So our site visits highlighted the importance of craft and consistency, and we typically ended each site visit with a quick huddle with the CM, with the timber installer, and the design team to highlight and discuss emerging issues that we were seeing kind of in real time. One of the site, one of the site visits kind of caught this issue.

I know that we spoke about it earlier, but the importance of again, sealing against water, protecting all of these joints. You’re seeing here both the kind of spline joint, and again, that key column to slab joint. And so just making sure that all of those really were watertight, that we were protecting from water on the slab above, and again, protecting from that wet gypsum

Another key issue was accommodating field tolerances and i know this is not the first time you’ve heard about that today but again in this case study while the timber was on the ocean the contractor team performed a field survey of the cast and place concrete cores which as you can see were already erected here.

And what they found is that a couple of them were off by as much as one and three quarter inches from top to bottom which is well beyond the anticipated tolerance and again this is a an issue where you have really highly precisely fabricated timber and cast in place concrete and really kind of dealing with some of the joints between those. So we call the joint brainstorming session um to try and address the solution and create something that was workable in the field and leveraged the timber construction process and was going to have an aesthetically pleasing result.

So working with the mass timber fabricator and the contractor, the team created a field adjustment manual to adjust this specific condition. Where, as you can see here, the glulam beam is meeting a concrete core through a kind of steel fitting assembly.

So in the typical connection, you have 3/8 of an inch gap between the concrete core and the gluelam beam, and no gap between the concrete core and this custom steel bracket. However, if the concrete opens, if the concrete is pulling away from that, basically we needed to find a way to shim that and to manage those within the tolerance of each piece of this connection.

If the concrete core closed, if it was leaning basically toward the beam, then the steel bracket had to be cut. And this was just again a simple step-by-step manual of what needed to be addressed, what was okay, and you know, have this vetted by everyone throughout the entire process.

Ultimately, what we decided to do was that for future wings, we would not pre-drill these pin connections as the gluelam beam met the concrete core, and instead created a field drill jig so that this could really be adjusted on site to address each of these unique conditions and the unique field tolerances of the concrete core. Again, this was more site labor, but this built in some adjustability that we really needed at that point in time.

And finally, the multi-story column span had a couple of unintended consequences that, again, we just needed to work through all together as a team. And as you can see here, one issue was, there is some needed bracing. You have these 40-foot tall columns. They are, they’re out there. And so you kind of get the slabs in place and the beams in place to brace them. So that was one one thing that we just had to make sure was consistently applied throughout the entire project.

The other was looking at this corner slab detail right here. It made perfect sense right here where you’re seeing this is the pink CLT slab. It’s not to go around the multi-story columns. There’s going to be a filler piece in here, but there’s going to be a filler piece in here. In actuality, this was totally impossible to erect in the field, to thread the CLT slab over a 40 foot tall column, you know, basically to sink it in, because you have your tolerances going in different directions here. So the installer had to field cut the corner off the CLT slab, as you can see here, in order to actually erect it in place.

But from a design perspective, we needed that corner. We needed it. It was finished ceiling in that student room and was also support for the exterior enclosure. So the solution was ultimately inserting an L-shaped piece of CLT, with a custom bracket that was either mounted to steel or that there are mounted to the gluelam column in some cases. And again, this was, this was an exposed slab that we were hoping to keep clean, but it was not in highly visible locations. So ultimately it was, I think, not a perfect solution, but one that that just addressed the reality of the situation on the ground.

So this is a brief construction video taken from the north of the entire process of construction. So you’re seeing first the concrete work, the cores going up, foundation’s being poured, and then mass timber arriving the further the southernmost wing. And I think it’s really interesting for us to see how quickly those go up. And you can see actually, in the end, we’re doing building A and building B at the same time, with two different crews, going to really putting that all together.

And that takes us to the end. Sorry, thank you so much for having me. I look forward to the questions as part of the panel discussion.