Advancing Construction: Unveiling the Power of Mass Timber Post and Plate Efficiency

On the left, you can see with post and plate, we have managed to get in an extra floor level. And of course, this is governed a little bit by the development approval process of local government. They’ll give you or there’ll be issues around what’s the achievable envelope height. And so if you’re a property developer, you want to get the maximum letible yield – a square meter each of floor area.

There’s also less complexity for mechanical, electrical and plumbing services, unlike deep post and beam scenarios where the beams have to have deep or large penetrations for ducting and wiring, and for example, cold water supply or where cold water supply might even dip under the beam and then come back up, which is unsightly.

This situation deals with that quite nicely. With this site, labor we’ve essentially, apart from the fact that it’s a pre-fabricated approach, it means that we’ve cut out a whole level of construction in removing all the beams. So less layers involved. This means a smaller kit of parts. And I guess you could say some from stock.

So if we go back to that previous slide here, you could imagine that these columns could pretty much come from stock. And the sort of spider connected by Rhodoblast is also from stock. It, uh, it’s flat packable. So in terms of delivery from potentially long distances away, you get quite a lot of economy in the delivery. It’s not like large 3D volumetric pods and things like that, where you’re essentially transporting quite a bit of fresh air.

That’s not an issue here. It’s a pretty repeatable design platform because it’s a grid layout. It has its economic sweet spots in that grid layout, but in terms of a production platform, you can fairly readily stretch it within, um, within a or to a degree within that economic sweet spot. It’s its economies of scaling that it can handle a 3D model from CAD that goes into a computer-aided manufacturing with a CNC approach fairly readily.

Brock Commons was perhaps one of the first using this system. It’s a Canadian project. It’s a bit of a benchmark mainly for process. It’s not a particularly exciting architectural building, but it was quite engaging as a uh, the way things could or should be done in terms of on-site process.

You can see here and if you follow my cursor for a minute, you can see that the columns, they’re relatively closely spaced. We’re only talking three to four meters, or in fact, probably not even four meters. But you can see these posts a bit like legs and then on your opposite side, and they’re supporting a table of sorts. So very easy.

And you can see the spacer bars cup Hold the top of the column grid in place, which meant the columns work were fairly rigidly located for landing these um, tabletop pieces. Um, it meant that there was relatively little propping of the columns that you can see here.

This was adding story student housing, and that’s why there’s such a close column spacing. It’s not that different compared to say a um, a jail cell. Um, they did two floors plus per week. That’s very fast. If you were doing this in concrete, on a 930, a 930 square meter floor plate, you’d be looking at a much longer than that. For this scale of building, you might be looking at more like 10 to 14 days or something around there.

There was only a nine-man install crew, six hours for floor panels to be placed. And remember, you’re close to a thousand square meters here. As little as three hours for the columns. You can pick that up that they used a semi-manual or a manual method, which as you get to larger scale buildings, is not really achievable because the columns become too heavy. But it worked a treat in here where they just plopped the columns into the uh, the metal plate on the floor.

They used the for the process modeling, which is essentially what we did. We thought that was a good idea. It seemed to work very well. Everyone on that project really liked the 4D modeling to assess their understanding of how the process would work in detail.

So we wanted to realize this potential, uh, in an extended application of post and plate for construction Australia. We were looking at new connector systems. We used a number of them, but because of time constraints today, I’ll just focus on the Rhodoblast spider connector. Uh, there’s some new mass timber concepts. One of those was the um, the band beam system that Katie just spoke about.

We wanted to apply it to different building types because even though Brock Commons worked well for student housing, it’s a very small column grid. Um, and so quite restricted.

We wanted to look at more like suburban office development, institutional buildings, and even to some extent multi-story residential, which can be used here or it might be used in for flat pack concrete here, because it allows the developer to not commit to floor light wall layouts until late in the project after they’re, or it means it’s flexible right up until late in the project which they can adapt to suit changing market conditions.

As we go to longer spans of six and seven meters, and and perhaps even longer, the Brock Commons scenario is is uh, tested because we need more temporary support. And I’ll talk about that a little bit later.

The Australian conditions that we designed our process for looked at safety, uh, compliance, regulations, and different supply chain conditions. Uh, our team for it was myself obviously, Andrew Dunn, uh, he’ll speak after me, Paul Winter who helped with um, the actual uh, uh, expertise in in converting what we wanted into a 4D or digital 20 model. XLam and Rotoblast.

Said before, we wanted to focus on process, um, and having that incorporated in the engineering design to make sure that this was fast, efficient, reliable, and above all, cost effective.

One thing I think we need to remember is that the construction contractor makes up by far, far, far the lion’s share of the price. And if they are seeing it as being an awkward process, they’ll add on extra cost. And so we’re trying to make this a very doable process. And what I mean by that is where you’re probably all familiar with floor cycles. So this ground floor, first floor cycle, first, second, third, fourth and so on. 

You want that to be very reliable and very fast for Timber to have a competitive advantage ove rthe in- situated concrete alternative. Within that we have a number of subcycles if you look at that in terms of concrete you’d have form Rio and poor and probably some post-tensioning as well. But we’re looking at trying to minimize and simplify these subcycles and so this.

This say for example two-day floor cycle is going to break down into four subcycles, we’ll talk about that a little bit later. And they need to have in production terms flow balance buffering just in time and limited queuing.  And what that means is the process these four subcycle processes have flow in the as one finishes that it can pass on like a fast Batten change in a in a Sprint Relay race.

We don’t have someone saying I can’t come till tomorrow or that bloke’s not around for the next week. This has to be a continuous process. We need balance insofar as whatever this first subcycle does is producing whatever it does at a rate that is ready for the next subcycles. Take it on we don’t we want a little bit of buffering so that this subcycle if  hings go occasionally wrong.

Not systematically wrong but occasionally wrong. We can we can deal with that or they can deal with it as a crew but we sizes to make sure everyone’s occupied, no one’s hanging around doing nothing, and that it’s a smooth process. And that to some extent incorporates this just in time approach.

The method that we use for undertaking the project, I won’t dwell on it too much because we’re short for time. We made a 30 by 20 model building floor plate and projected that across, I think was five floors for memory, maybe Andrew can pick up on that one. We looked at three different construction systems where we tested and refined them.

We did some field research and and technical inquiry to identify uh obvious issues, one in particular was around how we temporarily supported things. Then we had an industry workshop with our tier one and tier two contractors and various members of supply chain and specialist mass timber fabricators.

Within us, Paul, to create, or Andrew and I did a lot of work on this, a lot of debate between us, uh, created flow, uh sorry, uh Gantt charts of timing, crew sizes, things like that, put it together in a draft digital model, ran a second workshop for refinement, and then moved on to a completed digital model, which I think Andrew can give you the details on how to find.

Here’s the three versions. Here’s the center support infill system. You can see that it’s just got a minimal number of columns down the center line of the support, uh, the the main panels. We have to place these and then we can drop in an infill panel in the middle in between them.

So I’ll talk about this a little bit more, but you can imagine that having a table like this supported by only three legs is far less stable than the sort of Rock Common scenario we looked at, where there was six legs on every table on both edges. Um, so that that became a temporary construction or on-site problem. Then we have to get this first one in place, second one in place, and then to drop the third center, uh, infill panel. In fact, we found it better to put all of these tables in place and then do the infill panels as a as a latter process.

This is the one we’re going to focus on. This uses the spider connector that we’ll talk about more. This is a band beam system that we looked at with cross panels. I think Katie probably talked about something similar. I guess one thing here, um, is you need a lot of screws to create composite action, like many many screws, which we found to be on the critical path or getting it off the critical path was important.

The edge support system, um, pretty similar to Brock Commons but using different connector systems. So as I said, I focused on this one, or we, I’ve seen folks in this six by 6.8 by 8, 9.44 meter grid. This used the spider connector which you can see on the right. I think the thing I wanted to point out here is that this, this cylinder on the left hand side is typically going to be prefabricated onto the top of columns, poured outside.

And it’ll punch through our CLT floor plates. But then you need to place the spider’s arms. By the way, it’s obviously an insect. It’s only six legs, not eight. Um, but they need to be placed on top and then they fall into this locking ring, whereas this piece here is sitting on the underside of the next column to go up.

And we found that this created certain complexities on site. Um, it, it would probably be better for these to be to have studs, uh, like I’m thinking like an engine block type stud where they come down through the plate and this, this particular ring can be more easily fixed to it on site.

We found it from our point of view in Australia. We’d prefer to do fixings from underneath over on the deck, if you follow my cursor fixing up underneath than having someone on a live deck. And so this tended to dictate the process that we adopted for this particular approach. It was different for the other two systems.

And obviously, as I’ve said before, there’s a lack of stability during, um, fabrication. You might have a, a 12 by two meter long or three, three meter wide panel here only supported by perhaps one or two legs. How’s that going to sit there unless you use sky hooks? It’s not going to work. So we had to think a lot about our temporary support and how to minimize that because it’s not, it’s not evaluating cost in the overall process. It’s something that you have to have that you want to minimize it.

I’m going to play you a video now. But before I do, this ended up being about a 1.5 to 2 day floor cycle. We had a 12-man crew and I’d break that down. We had a tower crane that did what you might call heavy lifting. And a mini crane that was used for um, some fabrication bays, which I’ll talk about later.

Features the spider collection connection. And if you look during the video, you’ll see floor four subcycles in our overall floor cycle. One was fabricating bays outside the um, the the footprint or the the floor plate.

And we refer to those as T tables because they’ve got legs down the middle or columns with the spider connector on top. Then we lifted those large assemblies into place on site. Then we did that for all of the T tables. Then we put our infill panels in. And then finally, jointing and screwing.

I’ll just click the video and hopefully we’re in business here. [Video plays] So you can see the fabrication base set up on the side of the building. The orange part is uh edge protection for a safety. You can see temporary push pull props stopping the tables from falling over. You can see the connectors going in. They take a lot of time. Um, and they’re, they can be on a critical path, so looking at the number of connectors is, is actually quite a very important thing if you’re looking at time and process.

[Video continues] So this is what happens inside the making the T tables in the fabrication bays. You can see the spider connectors going together. The spider acts like an umbrella of sorts that holds up the panel and reduces the or helps assist in having a reduced span or deals, sorry, with the overall span. It makes longer spans more achievable.

Andrew will talk more about connectors, but these CHS connectors and these sort of bow tie shakes things help pull the panels together and help deal with tolerance variance between the different panels. The couplers help try and transfer load. They have to be connected on the underneath and on top. So for us it means we’d put a team in early underneath because they’re working in a safe area with edge protection, whereas we had to wait a bit longer or um, or the clock top time to get on because we need our edge projection in place.

We need to put edge protection around every single panel. As you can see, every T table. Um, so they could start putting in the spider connectors without having to wait. But once we got to a certain point when the infill panels are placed, which will happen in a second, we’re able to start removing the uh, the edge protection around the T tables. So you’ll see that happen now. And the reason we want to do that is so that we can start preparing for that to be in stacks that we can then eventually crane up to the next level for the next floor cycle.

Okay, so I think we’ll leave that there as it’s just going to keep repeating. It gives you a feel for the spider connection, um, and and the four different subcycles.

So as I’ve listed on the left there, we have this fabrication bays for the tables. This has an underlying assumption that there’s room on the site for that. We could reorganize this to um to have a single fabrication bay. The fabrication base were mainly there to help remove safety issues of putting the spider connector in place in situ or our lack of the difficulties in creating Edge protection.

I think you’ll see in a moment that the two fabrication bays aim to help make sure this spider crane and indeed the tower crane that you see on the left here busy. So once this guy finishes making his T table, the tower crane will then spend time with a typical crane cycle putting it, bringing it up and dropping it off. And in the meantime, this second fabrication bay has already started its process of making the next T table. And then in the meantime, or once this crane has finished dropping fabrication bay once, it then picks up from fabrication bay two.

Crane cycle tends to rule your uh, your maximum speed. Um, here you see the sort of Gantt chart. I know this is complicated so I won’t dwell on it, but this breaks down our um, our subcycle one. So for example, the blue ones here are the first fabrication bay. They overlap very neatly in this nice consistent line.

The green ones are the second fabrication bay. And you can see that our crane, which is crane cycle, which is in orange, the moment this guy’s finished, the crane can pick it up. The moment the crane is finished, um, that dropping the blue one in, the green one’s ready and it picks up and there’s no wasted time in the crane cycle throughout the process.

Then we move to putting, dropping in the infill panels in black. Um, and again that’s on the same angle of trajectory which shows a pretty good cycle. Where we take up a lot of time is these ones down the bottom. These are essentially our CHS connectors.

This one is where we started on the underside of the panels, and we could manage to do that without affecting our critical path. But from this point onwards, these things are taking up time at a reasonable amount. So if we’re looking at speed, we really want to try to look at connector systems that reduce that. And the ones were used were available on the market, but I think there’s potential to improve.

So coming to a sort of conclusion and findings, pretty fast 1.5 to 2 day and reliable floor cycle. Our actual bar charts are just about 11 hours was the minimum, despite it dictated the overall process strategy. Tower crane cycle rules the process and that’s why we use two fabrication bays.

We tested, I didn’t mention it before but we tested using adhesives instead of mechanical connectors between the panels, which is really interesting technology. But I think for practicality, the things that affected are the curing time. So for example, if you have to wait two days, which was a sort of typical time where you can get around 80 percent strength, um, you, you, that’s just added onto your floor cycle because you can’t start your next one.

Um, and then the inspection test plans may also take increased length because, uh, you may need to do some level of lab testing. That’s unclear at the moment. And if you do, that’s more or less a separate subcontractor in the process. They have to wait for their results, which may all add to the floor cycle.

Um, as mentioned, aim for balanced crew sizes, repetition and synchronization between crews, start fixing as soon as possible. I think the digital twinning, actually we found it to be quite effective in seeing problems with things and showing people. So it was actually quite useful in seeing, clarifying and deciding areas for future development.

I think, well, one which we’re aiming to look at is in a project forthcoming hopefully is managing stormwater during construction, and to a lesser extent or to some extent wind because it affects crane, but stormwater can reduce unwanted variables, um, reduce fixing and jointing, screwing robot is, is a definite idea and craneless column installation, possibly using something like a glorified, um, uh forklift.

Okay, that’s me. Thank you for listening and I’ll hand it back to you Carla. Thanks so much Carrie. All right, well um, following that representation will flow on to Andrew who’s going to talk about some of the connection considerations in person player system. So off to you Andrew.