Beschreibung
Wichtige Erkenntnisse
- Discover preparations and procedures for large-quantity, multisource data collection
- Understand the processes for managing and merging massive volumes of reality capture data
- Discover the procedures to use reality capture data for BIM and photo model creation, and to model existing components
- Discover how to create visualization and virtual reality from BIM and photo models
Referenten
- DWDavid WinslowDavid Winslow, P.E. has worked for the U.S. Bureau of Reclamation for 33 years and is project manager for the unified model project at Glen Canyon dam. He has extensive experience in CAD software management, support and training along with experience in design and construction. He has provided support to personnel throughout Reclamation on CAD and drawing management for over 25 years. Dave holds a Bachelor’s degree in Civil Engineering from the University of Utah.
- COCARLOS ORONACarlos Orona is a Business Consultant for the Engineering / MEP / CAD / BIM industry. Over the course of his 18 year career, Carlos' has worked as Designer,BIM/CAD Manager, and engineer in the mechanical, electrical and plumbing/piping (MEP), water/waste (Headworks) consulting industry and has been involved in many large projects in and around the United States and international. He regularly conducts seminars on the use of CAD/BIM technology in building engineering where his MEP and engineering background brings a complete understanding to the process of helping companies effectively implement CAD & BIM in a production environment. He regularly provides CAD & Building Information Modeling implementation services, customization, content creation, training, and support for Autodesk AEC sofware.
This is a modal window.
The Playback API request failed for an unknown reason
DAVID WINSLOW: So by a show of hands, just to make sure everybody's still awake after lunch, how many were here last year when we did this for the first time? Oh, not as many as I thought. So that-- appreciate that. That helps me. And as we go through, I've got a couple of gift cards we'll hand out.
AUDIENCE: Hooray.
[LAUGHTER]
DAVID WINSLOW: Good. Good. Come on in.
AUDIENCE: David, go ahead and get started.
DAVID WINSLOW: All right, I think we're about at the point where we can go ahead and start. I know your time is important. I appreciate your being here. And I hope this provides you with good information. If you've got questions, please share those with us.
My name is David Winslow. And I work for the US Bureau of Reclamation. We're not the BLM. We deal with water, and dams, and power facilities, and canals. And we are in the Western US. I've worked for Reclamation for over 35 years. So I've been there a long time. I'm an old guy. It's just the way it is. My co-presenter today is--
CARLOS ORONA: Good if I turn that on. Right? My name is Carlos Orona, and I'm senior implementation consultant with Autodesk. And I have been in the engineering industry for now working on 27 years, so everything from petrochemical, industrial piping, water, waste water, and oil and gas.
DAVID WINSLOW: OK, so our session today details the work that we did with Autodesk. We partnered with Autodesk to create a comprehensive model of one of our larger facilities. That facility is Glen Canyon Dam.
It's located on the border of Utah and Arizona. It's on the Colorado River above the Grand Canyon. So for those of you who know, close to us here is Hoover Dam. Glen Canyon is above Hoover Dam.
And there is a compact, Colorado River Compact, and that compact says so much water goes to the Upper Basin states and so much goes to the Lower Basin states. And this is the facility that we use to make those required releases to the Lower Basin.
OK, what is Reclamation? Well, we were established in 1902. We have nearly 500 dams, around 337 reservoirs throughout the Western US. We are the second-largest producer of hydropower. So that hydro clean power, we produce a lot of it.
We are also a big provider of water. Now, most of us, when we get up in the morning, we stumble into where we can wash our face and use the restroom. We won't really think about where, when we turn on that light or the faucet, where that water or power is coming from.
It comes from a variety of sources. But Reclamation may well have provided the water and the power that you used this morning. We also provide a lot of water for farmers. And that water goes to produce a lot of the food that we eat annually.
Why is it so important? This photo here was taken in 1942. That's not that long ago in terms of the overall length of our nation. These people didn't have much water.
So Reclamation's mission was to provide that water, and then also power, as well as recreation benefits, and municipal, industrial water, provide all of those benefits to these people so they could actually inhabit the West. Because the West, unlike the East, doesn't have water really where you need it. It has to be stored, collected, and then moved to where it needs to go.
We have a big impact on the economy of the nation. We directly-- or because of what we do, a lot of people are employed. And there's a-- that power that I talked about earlier, that grosses nearly a billion dollars every year to help with our economy, to keep moving things forward. Does anyone know what this is?
AUDIENCE: Is it in Dallas where they killed Kennedy?
DAVID WINSLOW: What's that?
AUDIENCE: Is it in Dallas where the shooter--
DAVID WINSLOW: It is. It is. OK. You know what year that was?
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: Yeah, this was 1963. OK? That building was built-- it was actually built in 1899. It was rebuilt in 1902 after a lightning strike nearly burned it down. Reclamation was established in 1902. This dam, the concrete was completed in 1963, so really the same time period.
This is Glen Canyon Dam. That reservoir holds 27 million acre feet of water. It's a little smaller than the reservoir for Hoover, but very similar.
OK, so I don't know if anybody here has seen the great pyramids in Egypt. There's a smaller one just down the road here, and I don't know what it is. The volume of material that built that is a little over 3 million cubic yards of material, of stone, which is very similar to Hoover Dam.
And Hoover Dam is quite a bit higher than the pyramid. But Glen Canyon is a little bit shorter, but it has 50% more concrete in it. So this is a very, very large facility, nearly 5 million cubic yards of concrete in the dam alone. It's a big facility. Here you can see Carlos.
[LAUGHTER]
It's a lot larger than he is. That little speck there in the middle, that's me. It's got way too many of these ladders that are very scary. Here we are out on top of the power plant. What were trying to do here was just collect as much data as we could.
It's got places that are wet inside. The power of the water is really something. If you ever get a chance to go to Hoover or to Glen Canyon, I highly recommend you take a tour.
It also has a power plant. There are eight generators in there producing power on a regular basis to help us live our lives every single day.
When this was built in the '60s, we didn't have CAD. We didn't have PCs. We didn't have laptops. We didn't have tablets. We didn't have cell phones. But we had lots and lots of paper, and we have over 10,000 drawings for this facility.
A number of years ago, I was tasked with creating a model. And the idea was just in the same way that the concrete in your driveway or your sidewalk cracks, well, same thing happens to a dam. And we have to be concerned about that because it's holding back all of the water.
And the question was, are these cracks that we're seeing, do they connect anywhere? And so in order to do that, I went through all of-- well, I didn't go through all those drawings. I went through quite a few of those drawings to find the original design.
From that, I created this model in-- back then, it was Map 3D, and with the idea of making it so that all of those green lines that you see there, those are where the cracks are inside in the galleries and the adits.
And this helped give us kind of an idea of how the facility works together. But it was very-- it was too rudimentary. It just didn't do enough for us.
I've been trying to find a better way to make this model work for us for quite a number of years. And about four years ago, I went to our R&D office, and I said, this is what I want to do. I want to create a comprehensive model of this facility so that we can manage it, operate it. We can do our redesign-replacement with a state of the art model and tools.
In order to do that, I went around to find all of the survey information I could find. These brass caps, notice here this one was put in in 1957. So it's been there a long time.
I found them up on the edge of the concrete-- or up on the edge of, actually, the canyon walls. I found them in lots of different places. And then I actually found more as we went along.
So that helped give us the information we needed. It also provided us with a GPS location. I brought in a surveying team, and they spent a week going through that and establishing control for the work.
The types of data we collected were we collected LIDAR data, SONAR data. We collected photos, and photogrammetry, and also survey data. We collected that data on the exterior of the dam, throughout the power plant interior and exterior, upstream, downstream, on the canyon walls.
So at that point in time, this was August of 2015. We had a group of Reclamation employees come together with Carlos and some other Autodesk people. We had three main scanners that we used. We scanned all of those places that I talked about earlier.
But as you can see from this photo-- and this is a pretty big job. That distance from where that scanner sits to the crest of the dam is over 500 feet. So over 700 different locations were used to collect LIDAR data-- upon the walls, downstream, on top of the power plant, inside the power plant.
One of our units was actually undergoing maintenance at the time. And so it was partially taken apart. This was a confined space. But we had one of the facility employees take a scanner inside there. And he scanned where the water actually comes into the generator, the penstock and into the scroll case.
There's a false floor here. The top there is the turbine runner. And that's what the water pushes to spin it, the generator.
Now, all of those scans were taken, and they were merged together using ReCap Pro. If you see there on the right side, and all of those circles, those are locations where scans were taken. Since this is a plan view of that, those scans were actually taken in four different levels of the power plant.
From that was created a point cloud. This is not a photo. This is the point cloud. This provides us with a lot of information just as it is. And it's just the start.
This is the main generating room, the point cloud. What we have never seen before though is this. This is that generator-- or generating unit. And we can see it on different levels. We see how it all works together.
I've wandered down those corridors for over 20 years. And I've never seen something like this. This really helps with our understanding.
We did both of the spillway inlets. And these are both set into the canyon walls on either side. We scanned not only from the concrete but also from the radial gates themselves. We never did get down there into the inlet because it takes a rope to get down there. And you don't want to be on a rope with a really expensive piece of equipment.
But from all of the locations where we took scans, we were able to basically recreate that in the point cloud. Now, I can take this information. I can send it to our designers who are remote from the site. And then they have enough information that they can do whatever work they need to do.
If it's rehab efforts, replacement, they don't have to be there. They don't have to come to the site. I can give them all of the dimensional information. And they also get visuals. They can move around this data. They can flip it. They can look upside down, whatever they need to do.
And you'll notice there that there is a big, dark hole. That's a spillway. And it goes way down, and it's dark. This information helps us really to understand the facility and how it interacts with the canyon itself.
This is the data that we collected, a massive amount of data. From that, we're able to create these renditions. OK? I told you that we had LIDAR data. We have data from the canyon walls. We had SONAR data. We can take all of that, bring it together in Navisworks, creating any type of visual information that we need.
This is good for visitors. It's also good for our employees. Many of them don't get the chance to ever see this. It helps them with their understanding of who we are and what we do. And if they're working on this facility, it helps them understand how everything fits together. Same with this, you can readily see how the power plant is put together.
Now in addition to that, we took lots and lots of photos. Over 2,000 photos were captured and probably even more. I've gathered a lot of them. We were able to actually even get on the river down below the dam, take photos there and video.
In order to get some of the aerial information, Autodesk hired a helicopter. There's only one helicopter there. It's the medevac. They weren't being used at that point in time, so they purchased a few hours of time and got quite a number of spectacular photos and videos.
OK, from that, then we created a photo model. Now, at this point in time, because of requirements for cloud applications, the government requires that those providers and the people that host those cloud services have to have certification.
At that point in time, Autodesk didn't have it. And so rather than doing this, like would normally done in the cloud very quickly, this was just done on a single PC. It took almost a day to complete that.
AUDIENCE: Except, David, you made it sound like Autodesk has that certification.
DAVID WINSLOW: They haven't.
AUDIENCE: It has not.
DAVID WINSLOW: They still are working on that. Hopefully, they'll get it one of these days soon. This is the photo model. And you can switch it from-- and you can see the color, different versions. You can see it just in the tin. So and this is just created using that photo information.
Originally, that was-- I don't know if you remember the, a number-- a few years back, it was called Memento, I think. Then it became ReMake. And this product is now ReCap Photo. That's what was used to create that.
OK, other requirements that we had were-- we wanted to try as many different technology as we could. So we thought, well, we could gather some of this information with a UAB. But there were requirements on us that really would have put us way behind. It would have taken a number of months to get those approvals. That's why Autodesk hired the helicopter.
But the interior space did not have to have those FAA approvals at that point in time in 2015. So, and it still doesn't. But Autodesk asked, and we gave them permission to use this drone inside the power plant.
So here, you can see pilots. And there's the drone up there. The drone is just taking photos, video. And you can see that generating unit that's partially open. And from those photos, they created this video of that. This is what's called the stator.
Now, one of the other things that we were looking at was we want to get some SONAR data. We want to find out about the upstream area of the reservoir, just upstream from the dam. And the people there at the facility, they said, one of the questions we have is, is sediment reaching the dam?
There are some locations, some places where sediment inflow into the reservoirs is really a big problem. And there are facilities that are nearly useless because their reservoir's been filled in with sediment.
The water here is nearly 500 feet deep, and it can be higher. So getting a good understanding of how much space we still have that's available for sediment is something that they worried about.
So another partner came in. They had that little ROV that you saw. They established the base station on the GPS point. We used the facility's boats. And they had a driver there in that boat. There was power moving this, this little ROV, where it needed to go.
All of the information that was collected by that ROV was sent wirelessly up to this base station. And the operator there could see in real time what was being collected, if it was meeting the needs, and tell them where to go down for the next pass.
We did the area between the face of the dam and the buoy line, which is about a quarter mile. And they did that in about two and a half, three hours. It took them longer to set up and prepare it than actually do the SONAR collection.
And from that, we got this information. And this has been colored thematically to give us a better understanding. Here, we get a pretty good idea of the inlet channel there, what it looks like.
And in discussions I've had with others, that little ridge that you can see right here, we believe that's the original cofferdam from the construction in the '50s and '60s, the upstream cofferdam, which really tells us there is very little sediment that has reached this point so far.
That information can be utilized inside of ReCap also. We can move it around. You can see here it's colored thematically also. It tells us we have about 270 feet before sediment reaches the bottom of the intakes. So there still a lot of life in this facility, in that reservoir. Are you next? OK, kill 30 minutes.
CARLOS ORONA: So the process of--
[MICROPHONE FEEDBACK]
Hence why I love miking myself. Is that better? There we go. OK. So the process here that we came into, the modeling part of using the scan data, one of the things that was asked was, well, let's model this. And my first response to that was, I can model anything. Let's model. Let's model the dam. And one of the things is getting into the complexity of it and not just the level of effort involved.
Taking the scan data and using the scan data to get proper dimensional information and being able to model it, so that that was a huge level of effort, just taking dimensions and coming up to this type of a model with a structure. That was the first part of it-- not too complicated.
What got complicated was the custom connections in the steel. So there is a manual effort when it comes to modeling. And then there was feature extraction involved in this and then custom Revit families to accommodate the detail of the model.
So these are some of the custom connections you see on the bottom. So it was a lot of repetitiveness here when it comes to the structure. But still, it was complex-- is making sure that the connections were right.
And there's some conceptual information when it comes to the generators. And you'll see those modeled to a more higher level. For those of you who know what LOD is, or Level of Detail, this would be in the ballpark of a 450, 500.
And same thing here using feature extraction, there's two softwares that we used for that. One was Kubit, and the other one was PointSense. I think it's called As-Built now.
Taking the point cloud data, setting it up, there is some setup that needs to happen for that-- but setting up to where you can select the point cloud and then it map to Revit objects to get the families correct. So that was all the structure part of it.
I rendered this just to give it an image of what it looks like on the inside, adding some lights. And those generators are now the Revit model generators-- and using feature extraction, custom parameters in the Family Editor to make that look how it is now as in As-Built
And you can see, using the model, we can animate. And this is showing the maintenance of the stator-- so model, point cloud, together, and then animating the information all in one.
So in the advent of the information that we had in the work that we've done so far, using the point cloud data, what could we do? Up to this point this year, we have this project that came about where it was to use existing information that we had modeled, model a system, and see if we cannot get clear judgment calls on how things are going to get installed in the field and for those who are going to be doing the work.
So this was a UV system, and you can see that over here on this drawing, to be installed in this fire protection line. So existing condition modeling, again, using piping systems, this particular model didn't have any.
So setting up the routing preference for the pipe system, making sure that it's correct, making sure it's reading the right tables for welded pipe and also for flanged connections, and feature extraction for the cylindrical shape of the pipe, making sure that the slopes were correct, and that's what you will see up in the point cloud up here in the top.
And getting all this information together, then we can come up with, what does the existing look like? And how is it going to look like when it's completed? And making sure that it fits, because that was one of the main concerns is, is the equipment going to fit and accommodate the space?
The particular unit you see here, this UV unit, that was a challenge because I didn't have it. And modeling it wouldn't model to the level that we needed to accommodate that LOD 500. The manufacturer did have files for it, but they were done in SolidWorks.
So I needed to bring that information in. And it was just too heavy. It would just bog down the system. It was like 400 megs or something like that, some crazy number.
So taking that into Inventor and cleaning it, shrink-wrapping it, and then bringing it into the Family Editor, adding the proper connectors to it so it can flex accordingly-- non-parametric, but the connectors are. So then connected to the end, it will detect a pipe, and then you can continue on with the run.
So once that was completed-- whoops. Let me go back. And once that was completed, then bringing that into all the different generators and being able to have an As-Built model of that particular unit-- and that is one unit. There's four that were done this way with different systems across the piping gallery and the underwater gallery.
So you can see, leveraging the ReCap model, there's just so many things that can happen on the back end. So big level of effort doing the scanning, very challenging work, but very rewarding at the same time.
DAVID WINSLOW: All right, so this really proves out the model for us. In a way, it allows us to say, we've got these pipes. We've got this piping system. It's throughout this facility.
We even have personnel there who they've said, if you could provide me with something that tells me what that pipe is because, even though I walk through here every single day, I don't know exactly which system is which. That's part of what we're looking to do.
But in order to prove the value of the model, it was, we've got to replace this. We've got to bring in new straining systems. The strainer isn't good enough for the clog of mussel infestation that the facility is now having. It's a problem throughout the US, especially here in the West.
So we've got to get this new piece of equipment in there. We were also looking at adding that UVA system to disinfect to try and take care of those. The combination of those two, we hope, will keep this problem at bay for a while.
But specifically, what it did was it told us that there is room there in that concrete for this. We had originally thought of putting the UV system somewhere in here. It wouldn't fit.
Carlos and the design engineer looked at and said, we can do it around the corner here. That helps us out. That helps us to do designs better.
Our craft people there at the facility, they are going to be installing these pieces of equipment. So we want to give them as much information as we can to help make their job better, faster, safer.
This visual, this design helps them to understand exactly what they need to do, what pipes need to be cut, what needs to be added, what needs to be taken away. That will pay dividends for us in the future.
This project was such an innovative project that it was awarded the Achievement Award from USGIF, which is the US Geospatial Information-- excuse me-- Intelligence Foundation. That's why I put it up there because I couldn't remember what that acronym is.
It basically shows what can be done when you put together a good team. Autodesk was a great team. They really helped to make sure that this was a successful project. If you've got something like this, I would consider working with Autodesk, getting their input, getting their help. It was a big, big part of the reason why this was successful.
OK, so where do we go from here? Well, we proved that those models can help us with our design work. In order to help those facility people, we're going to be adding all of those different piping systems and labeling them so that as they walk through the facility, either with a laptop or something like that, tablet, and my idea is they just kind of point it at what it is, and it will tell them what that pipe is.
It will tell them where that pipe goes, where it's coming from, where it's going to, and whether it goes up or down so that they have a better understanding of what's there in facility, so they can do their job better, so they can be safer.
The electrical systems in that facility are very dangerous. We want to give as much information to our employees as we can get to them. We want to make sure that the drawings are up to date, that they have the latest documents, the user manuals, the repair manuals, the inspection reports. Anything that we have has been digitized and is available to them when they need it at that location so they can do their job.
In the future, the electrical systems in the facility normally are about 10 to 15 years before they're replaced. Mechanical systems, that pipe is good for around 50 years. We rewrite-- excuse me. I can't talk today.
We rewind generators. They're taken apart, rewound about 30 years, every 30 years. So there's always something going on at the facility. We want to make sure that new designs are added to these models so that, over time, it will become fully featured.
You'll have all of the information and that can be used not only by technical and craftspeople, but also by management to better understand the facility, to better operate, to better maintain, to better keep it going forward.
The concrete in these facilities will be around for over 1,000 years. As a federal agency, our mission is to make sure that that facility is operable for as long as possible, that it's a benefit to the people of the nation, that it continues to produce water and power to make sure our economy keeps moving.
We will also be looking to add our Maximo system into this, where the work orders we have, where with spare parts, so that our inventory and our jobs are tracked, and managed, and all of this information comes together.
We'll be looking at moving this into InfraWorks, make it easy to use for non-technical people. We'll probably even take a stab at VR or AR. We have the ability to really look at anything that comes about.
Tying this all together, they had some sessions yesterday on Forge. I don't know if any of you went to the DefCon yesterday. But they were actually underpinning with Forge. There are other applications that can help us do this work. So as we go over the next few years, be continuing to add to the model to refine it, to make it more useful, more complete.
This is what we have today. We don't just have those hand-drafted drawings. We have a point cloud. We have Revit models. As you can see here, that point cloud matches up very well with our original drawings. All of that can be used to help us do a better job.
OK, so I've kind of talked about what those benefits are to us. Hopefully, this will be a prototype for some of our other facilities. We all need to do a better job.
Those of you who have dams, it's aging infrastructure. We need to make sure that they are well taken care of. Because if they're not, they can be a hazard to any population below them.
This is the machine shop. There's the main generating room. There's unit number 6 that was open. This is the scroll case where the water pushes the turbine runner.
This is the lowest gallery. This is the pipe gallery. And you see a number of pipes there-- generator gallery. And this is the transformer deck, and you can see the transformers there. Transformers are going to be replaced next couple of years. So we'll have new designs for that-- will be added into these models.
It's just a continuous job to make sure that this is done in the right way and it's useful for the future. OK, we've got about 15 minutes left. If you've got questions, we'll take those first. And I've got three gift cards to give away. Sir.
AUDIENCE: If I ask, can I get a gift card?
DAVID WINSLOW: You can ask.
AUDIENCE: What is the average cost [INAUDIBLE] for [INAUDIBLE]? So just to have an idea of how much it would cost for all that [INAUDIBLE]?
DAVID WINSLOW: You know, from the reclamation side, in terms of just dollars that we put into a contract with Autodesk, it was about a third of a million dollars. But that doesn't include our time.
And Autodesk donated a lot of time also. They wanted to see a big infrastructure project. I'm assuming that their costs would be quite a bit more than what ours were, a million or two.
AUDIENCE: And it's all federally funded, or--
DAVID WINSLOW: Yeah, our side was all federal funded.
AUDIENCE: Yes. Obviously an incredible level of detail goes into [INAUDIBLE]. I saw that there was some spots inside [INAUDIBLE] very impressive. And you guys have gone through a lot of effort to make this very real, so kind of some questions to your rewinding generators on the cycle, replacing equipment on the cycle. Are you planning on people verifying and re-utilizing all of the time and set-up that you've put in to making this real in your [? wheel ?] verification. So you have, let's say once every 10 years, you've got a fresh scan that you backdate all of your installs with?
DAVID WINSLOW: It's my belief that, as time goes by, that we will see the cost of the equipment going down. The equipment will become more autonomous. It may be that instead of having a car with GPS or a LIDAR scanner mounted on top, it might just be a little tracked robot that goes around and does it all by itself.
Yes, this model we have right now is basically the As-Built for the facility. It's only 50, 60 years late. This will be used as the base for everything we do in the future. Sir.
AUDIENCE: One of the slides you showed using AutoCAD features in Map, that cracking--
DAVID WINSLOW: Yep.
AUDIENCE: --on the dam face, you get other technology, like point cloud, that you use to keep monitoring the crack situation, or every time you have to do it kind of like semi-manually. I'm interested in how you guys do that.
DAVID WINSLOW: So the crack monitoring-- and because of the chemical nature of the water, when the water comes through the crack, it deposits minerals. So it's not only a crack, but we also have mineral build-up.
It's almost like stalactites and stalagmites. So those get created over time. And every once in a while, they clean them off. When I originally did this, I went through every bit of that facility, photographed every single crack.
AUDIENCE: And instead of manually, [INAUDIBLE] just guides us to more--
DAVID WINSLOW: Yeah, I did that manually. Now, in the future, what I would do is I would take that little tracked LIDAR robot, send it through, have it collect all of that data, process that into a model.
And then I could take those. And every couple of years, I could compare them and have it show me, highlight the differences, tell me what's happening here. So that's really where it's headed. It's as the capability goes up, the cost goes down, I can do this much more often.
AUDIENCE: And I know [INAUDIBLE] in I believe it was Sweden-- it might be Norway. But somewhere in that area, they actually use drones to photograph the outside surface of a dam to do exactly that, to monitor the cracks on the outside surface of the dam.
And so that could be done here as well for the exterior. And then you can start making connections of the interior graphics and those on the inside.
CARLOS ORONA: Well, that would be great. Just one of the limitations is that the drone at a high elevation, wherever we are allowed to fly, may not be able to see that cracked dam. A crack is rather small compared to a large structure.
AUDIENCE: Well, and this is--
[INTERPOSING VOICES]
AUDIENCE: --dam and cover the entire surface.
CARLOS ORONA: Because I'm interested in not just cracks, but how big the crack is and in its size, what monitoring does [INAUDIBLE]?
DAVID WINSLOW: Yeah, and one of the problems there--
AUDIENCE: I'm not saying--
DAVID WINSLOW: Yeah. One of the problems there is that when you're on a rope that's 500 feet above--
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: --certain death, you don't want to get out there very often.
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: If we have some kind of a drone or something that can get out there and do that on a regular basis, we're much better off. There was a question.
AUDIENCE: Yeah, I was just noticing here-- in the winter time, there's going to be water gushing in through all those cracks everywhere. But in the summertime, they're all sealed up.
DAVID WINSLOW: Oh, it's not only that, but it's also depending upon the level of the reservoir.
AUDIENCE: The head?
DAVID WINSLOW: The head, yep. Sir.
AUDIENCE: With such large spaces, how do you control your target? Using spheres or flat targets? I mean, a dam's 500 feet tall with a very smooth interior space there.
DAVID WINSLOW: Well, ReCap Pro is able to really do that itself. We really didn't have targets per se that we put out there for that. ReCap now has the ability to actually go in there. And it knows how to stitch those scans together.
AUDIENCE: All right.
DAVID WINSLOW: Sir.
AUDIENCE: You didn't use any targets at all?
CARLOS ORONA: No. There was enough detail within the space that we had that we can accommodate the parallax between the percentages. I would say that what was challenging was the roof because the roof was all white where we could have used registration markers for it.
Because we did have a challenge to, how do you blend white on white without having a point to reference? But other than that, everything else we just scanned without any markers in it.
AUDIENCE: Given the same data sets, the SONAR, the LIDAR, the point cloud were able to register all those?
CARLOS ORONA: Yeah, we federated everything as one model. And, yeah, we didn't use any-- I mean, there was just enough. There's kilometers and kilometers and kilometers of piping and rooms that we could pick from that gave us the right stitching.
AUDIENCE: By using the control points?
CARLOS ORONA: Yeah, mhm. Yeah, using the three control points.
DAVID WINSLOW: Sir.
AUDIENCE: Kind of following on these points, I think it was probably in the FARO software [INAUDIBLE] cloud registration, not in the [INAUDIBLE]. But because I'm wondering what the target is for FARO to see out of. Because FARO needs a 500 foot distance, or 500 yard?
DAVID WINSLOW: 500 feet.
AUDIENCE: 500 feet. FARO is only accurate at that distance. I'm just wondering what FARO used to--
DAVID WINSLOW: So the scanners were FARO 330s. And they're good for basically about 1,000 feet, 330 meters.
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: Yeah. But, now, the areas where we had the longest distances are probably the ones that we are the least concerned about in terms of accuracy. Just the middle of the concrete on the downstream face, is it important? Yeah. Does it need to be submillimeter accuracy? No.
The idea was get the facility. We will add extra scans in the places where we need it. There just wasn't a way to get the middle level of the downstream face.
AUDIENCE: Do you have [INAUDIBLE] people who give you very accurate data that also works very well with [INAUDIBLE]?
DAVID WINSLOW: OK. You know, there are still a number of things that I need to do. I need to figure out a way to get something down that 700-foot-long spillway tunnel, which is basically, I mean, it's total darkness. It's about 40 feet in diameter, runs about a 45 degree angle.
I need to have a drone that can do that and make sure it doesn't hit the sides or some kind of a sled that I can lower down, as well as I've got penstocks inside that are 15 foot diameter covered in coal-tar coating, so they're black. And they are almost vertical. So we still have a number of things that we want to try and scan. Over time, hopefully, we'll get them.
CARLOS ORONA: Just to add on that, that was the equipment and the technology that we had at that time. In just the last year, it's been crazy. I mean, the BLK360 came out. And when you're lugging around a 330 at 18-, 20-some pounds for 15 hours, and I saw the BLK, it just brought it-- like, you just went, where were you? You know?
So a lot of things have changed. And there's just so much technology now, so much competitiveness. Even on the drone side, when you're flying, and collision detection wasn't available six months ago, and now all the drones have collision detection on them.
AUDIENCE: How long are these RTC--
CARLOS ORONA: I have used the RTC, yeah.
AUDIENCE: I've got a question.
DAVID WINSLOW: Sir.
AUDIENCE: I know you [INAUDIBLE].
DAVID WINSLOW: Say that again. I'm sorry.
AUDIENCE: [INAUDIBLE] that the point cloud [INAUDIBLE]?
DAVID WINSLOW: So the-- did you get that?
CARLOS ORONA: No. I'm sorry.
AUDIENCE: [INAUDIBLE]
CARLOS ORONA: Yes.
AUDIENCE: And you have no rapid model. So geological elements. How did you do that?
CARLOS ORONA: The feature extraction, most of it, was done in PointSense. It's PointSense.
AUDIENCE: PointSense?
CARLOS ORONA: Mhm. So you set up the feature extraction by shape, not necessarily by element, so squares, circles. So squares are extruded. That would be for more the structure.
Circles would be extruded for piping. And then you map that data to Revit families. So the circles become pipe, the rectangles become structure.
AUDIENCE: So it's [INAUDIBLE]
CARLOS ORONA: They're masses. Correct, air masses. Yeah, yeah. You still have to map the information, and then you have to set it up. I would say the setup part of it is very in-depth.
But the mapping part of it is quick. Once you have the setup done, the mapping part is quick. You're just picking points, and it fills in the shape. [INAUDIBLE]
DAVID WINSLOW: OK, we're about two minutes left. If you've got other questions, please come up afterwards. I need to get rid of these right now though.
OK, first one, I had some comparison slides in there. Tell me which of those things that I showed you was the tallest.
AUDIENCE: Hoover Dam.
DAVID WINSLOW: Sir. Hoover Dam. OK. All right, the second one, how much concrete is in Glen Canyon Dam?
AUDIENCE: Oh, 5-- 700 and something?
[INTERPOSING VOICES]
DAVID WINSLOW: OK, 4.9.
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: 4.9 million. OK, I've got one last one. And this one was not part of the information we gave you.
AUDIENCE: Is it the gift card?
DAVID WINSLOW: No, no, no, no, no.
AUDIENCE: I have a question. Is Transformers buried in this one?
DAVID WINSLOW: [CHUCKLES] No. There are no Transformers buried in any of our dams.
[LAUGHTER]
AUDIENCE: Never allowed to skip.
AUDIENCE: Neither confirm nor deny.
CARLOS ORONA: There's some areas that I wasn't allowed to go into.
[LAUGHTER]
DAVID WINSLOW: You're killing me. You're killing me. You're talking heresy now. All right, the last one, who can tell me what year Hoover Dam was dedicated? It's not Glen Canyon. This is Hoover. Hoover was built during the Depression.
AUDIENCE: '32? '34?
AUDIENCE: '39.
DAVID WINSLOW: You're close, but--
AUDIENCE: '38.
AUDIENCE: '34 or '35.
DAVID WINSLOW: Who said '35?
AUDIENCE: [INAUDIBLE]
DAVID WINSLOW: All right.
CARLOS ORONA: They all got it right.
DAVID WINSLOW: Thank you so much. Appreciate your attendance.
[APPLAUSE]
Oh.