Frequently-Asked Questions about Laser Scanning

Laser scanning offers a tremendous value to both new construction projects as well as retrofits and renovations.  So much data can be extracted from the colorized point clouds including measurements, 2D drawings, 3D models, and specialized applications for flatness and verticality.  What are Owners in your area asking for and do you have the equipment and expertise ready to execute?

 

Read through the questions we've received from other GCs and let us know what questions you have.

 

Q:  How large was this area and how many scans were taken?

A:  For this particular area there were nine laser scanner set ups. And dimensions were an area of about 25 by 25 meters.

 

Q:  How do you scan above ceiling areas?

A:  You would definitely have to take down at least a couple of the tiles. And there is a tripod that you can basically jack up that will push the scanner up above the ceiling and you can scan into that area. The scanner is small enough so it will fit in that area quite easily. There is a special tripod where you can hoist it up above the area.

 

Here's a video about this tripod...

 

 

Q:  Do you have to use RealWorks to act as the hand off from the scanner to AutoCAD?

A:  You can extract the data directly into AutoCAD if you want to. However, there is one issue - you don't have some of the tools that you might need like registering the data in a very quick and concise mode to give you enough information to make sure it is good.  So it is possible to do it that way, but I would recommend RealWorks because it would start to segment out some of the data and clean up your dataset a little bit because you have a lot of information in there. So you have a much cleaner dataset before you get in and start using it within AutoCAD.

 

Q:  How many stations were set up for the model being demonstrated?

A:  On the one floor there were forty different scans.

 

Q:  How did you establish a coordinate system in the first example?

A:  We brought in a robotic Total Station and we laid out three different targets and shot them into the Total Station. Because it was an existing building and we didn't have any existing control, we created our own control for that particular example. When it came to the building under construction, there was an on sight coordination system that was used for that example.

 

Q:  Do you recommend a scanner by itself or do you think there is a benefit to having both the scanner and the Robotic Total Station?

A:  Definitely both. The reason being is if you have the Total Station you then have control of where you can place control. You'll need control if you want to report back exact information about the project so you need the coordinate information for that job site. You could go out there and scan but you would then have to identify some locations within the scan and receive some coordinate information from somebody else on the job site. So then you start to lose a little bit of control and you need to make sure that you have good communication and know where those points are that you can then bring in that control and geo reference the point cloud. I would recommend both if
you have the chance.

 

Q:  How complex is the registration process to stitch the point clouds together?

A:  We have a lot of options when it comes to registration and most of it is all automated today. So we have flat targets we can put on flat surfaces and we have spheres that can be recognized in the software. And automatically when you bring it in you can automatically register with those targets and it registers the data. The other choice is a function where you can register by planes. The software will look into the point cloud, identify where there is some vertical plane, and it can use planes to register the data as well.

 

So they are the two main methods used and all you do is bring in the data, click in the software, pick one of those methods, and the software will automatically go through and process the data and pick out the targets or the planes and stitch the data together for you. And depending on how large the number of scans are it can take between a few minutes or it could take 1/2 hour to 45 minutes depending on how many scans you have and the power of your computer. It is all automated and just requires a few clicks in the software which does the work for you and will generate a quality control report on how well that's been done.

 

Q:  What is the margin of error on these scans?

A:  Generally the range in the example I've shown is just two to five millimeters. So very tight tolerances.

 

Q:  When you edit the poly line what precision are you maintaining?

A:  From millions of points we're cutting a section through that point cloud and getting thousands upon thousands of points to define where the line of that wall is. So looking through that wall there's a slight deviation in the measurement in each point and it's taking an average of that. So we're averaging between just a few millimeters. If you compare that to a hand measurement or if you did it with a Total Station you would only take a few points. You might take one point at the end and one point at the other end.

What happens if there's a bow in the wall? You wouldn't necessarily see that and are making an assumption that the wall is perfectly flat and in some cases it may not be. So we have a lot more information and when we move it in the software we're moving it to those points. If there are some missing points we can ask it to interpret from one line to another and intersect the two. So in this case you can get the 3D points and manipulate it and fit it perfectly to line up with the millions or thousands of points depending on what you have.

 

Q:  On metal surfaces, can the laser scanner pick up areas of rust or corrosion?

A:  Certainly you can see the corrosion based on the visual colored representation on the point cloud.  To measure the corrosion you would need to have significant build up or degradation maybe more than 3mm.I have attached a couple of images to demonstrate this.  One is on a metal surface (beam) and the other is on a concrete
support column.

 

can laser scanning be used to measure corrosion

 

Q:  Once the polylines are created on a sections does it automatically apply these polylines throughout the height volume of the area to create solid surfaces or do we have to do that with another tool?

A:  From that polyline you could generate surface up and down and you can control the height of that. We've had some customers who have taken the 3D cross section and have brought it in to CAD and generate the model just from the cross sections themselves. So you could definitely do that.

 

Q;  Does the cloud support snaps and ortho type controls for accuracy?

A:  There is a control down in the bottom left hand corner of the software which allowed me to actually control the angle at which I created the polylines. So if I go from one line up to the corner I can actually say, "I'm going to draw it 90 degrees," or "I'm going to draw it 45 degrees," etc. So I can set the angle and I can also just set it to snap to
a point as well.

 

Q:  How long does this take start to finish to scan the area?

A:  The scanning time for each station depending on the resolution was between 3-6 minutes to set up. And then the processing of importing and registering the data probably about ten minutes in total for that small example. And then I did the editing and it took between 10-15 minutes to do the drawing of the cross section.

 

Q:  How does this compare to using a Total Station?

A:  For instance, with a Total Station you would have less data. So you would decide where you captured the points particularly around the section that has an arc. You would have to make sure you collect enough points to be able to collect the arc. So to create enough points to be able to create cross sectional information it would take a little bit longer. Because you have to collect each little corner of it... every piece of the building. You could definitely do it but it would take a little longer but you would still get a comparative result. The big difference is I also had much more information that I didn't use at that particular time about the location of all of the sprinkler heads, light switches, any finishes, any sensors that were there, light fixtures... All that information that I didn't create a deliverable from would take a lot longer to capture with a Total Station because you are only catching every little single point. So you could do the job but it would take a little longer to produce the results.

 

Q:  Does this workflow for 2D work with Revit as well?

A:  You could bring in sections of point clouds into Revit and could do a drawing in Revit. So you could export out the point cloud and do your 2D drawings in Revit and extract further 3D information. You could maybe input some point cloud and do some modeling in Revit. So you could export out the point to a DXF or DWG and then import them in to Revit as long as you don't do too many of them. And you could create your drawing in Revit if you wanted to.

 

Q:  Can you change to other units besides Metric?

A:  You can change your measurements to international feet or U.S. survey feet if you want to. You have many different settings available. You can have it in meters, millimeters, feet, inches, etc. So they are all available. It will dynamically change that. So if I have it in Metric and in the software dynamically change it to feet it would change on the fly to feet.

 

Q:  Was the original point cloud created in RealWorks?

A:  The original point cloud is collected in the 3D scanner and it stores the data onto an SD card. You then take the SD card, put it into your computer and the software will grab that information from the project file download it into Trimble RealWorks and then you can start to do all of your registration and creation of your deliverables.

 

Q:  How are the different stations shots registered?

A:  You have several methods to be able to merge different station cells together. One which you can set out flat target and spherical targets and the software will scan as you are scanning and will pick those up. And then the software will automatically identify those and by geometry put them together. So it figures out the geometry between different targets and slots it together. Or you could use plane-based registration and select planes throughout the point cloud and match those up and also register the data
together. So it's an automatic process but you have to make sure that if you want to use targets you have to physically put them out there and you have to select your method. So that comes with a bit of training and experience on where to place those, but it's just a matter of understanding how it's done and it's quite easy to do.

 

Q:  I see in the station explorer it looks like it's in color, but I've only seen grey scale or white. Did you have the option of color and just chose not to use it?

A:  In the scan explorer you have a button you can click to display in color or display in grey scale (if you have scanned in color, that is). What you have to do is when you scan you have to also collect images so that means you scan and then the scanner will go around and collect images. And as you import the data it will automatically color the point cloud as you bring it in and then you are able to visualize it in color.

 

Q:  What type of software will this pipe export to?

A:  You can export that in various different CAD formats. So Autodesk, MicroStation, Tekla, Vico Office, SketchUp, etc., and it will read the data.

 

Q:  Could the floor flatness tools be applied to a low sloped roof?

A:  Yes, you could definitely use it for that. We've actually had some customers use it for analyzing slopes of roofs and also parking decks, for instance. So you could do if for the roof or also for a deck or similar as well.

 

Q:  What are the hardware requirements necessary to run RealWorks?

A:  With today's 3D scanners where you're collecting millions of points per second you need a decent computer. So you want at least a quad core processor with probably a minimum of 8GB of RAM. But it is certainly better if you had 16GB or even higher. Then you need a good graphics driver (at least 1GB of dedicated graphics driver). Those are the main components you would need in your PC... and certainly lots of storage space.

 

Q:  For the floor flatness comparison result, how can that be transferred to the slab in the field to show or project on the actual slab where work may need to be ground or filled?

A:  You would select multiple points around that area to define it. You can export that out as a file that can go into your Total Station and then you could actually lay out those points. Or you could get a reference where you could measure off some beams but you are going to get a coordinate that's related to the job coordinate system. So you could use your Total Station to lay it out which I would recommend for accuracy, or you could certainly measure off some known locations and maybe some structures within an environment where you could measure out with a tape, as well.

 

Q:  Does the RealWorks software have network license capability?

A:  Yes. You can set up a license where it can say the number of users you can have at one time. And then what you'd have to do is divide up the project with different people working on different parts of the project. 

 

Q:  Sometimes it seems to me that scanning is almost as much work as surveying. What do you think about this?

A:  I have been surveying for close to 28 years and have had 12 years of 3D scanning experience and it has evolved a long way since I became involved in it. So today 3D scanning is quite simple. The instruments are light, easy to use, and quick and easy to capture the data and get it back into your computers. So what I find with 3D scanning is it is much faster in the field than surveying, but it takes a little bit more office work. So you have these big 3D point clouds and you've got to decide how you want to produce and what type of results you want to produce. I would say there's a lot more office
surveying compared to field surveying when it comes to 3D laser scanning and you really need to go into it just like any survey...what is the deliverable you want to create in the end? That will determine how much information you collect in the field. That's nothing new - it's just common sense when it comes to survey. The big advantage of 3D laser scanning is you get a lot more information. Sometimes you don't use that information right at the beginning, but when somebody comes back to you and says, "Hey, we really needed to get all of those beams captured as well..." Well, if you went out to survey and just surveyed the deck where the ultimate aim was to get say a 2D plan, but now they want the 3D structure of the building, you already have the data so you don't have to go back to the field.

 

Q:  What about other markets where there isn't a scanner that you can rent for a project?

A:  We do have a dealer network and people can rent scanners from those dealers. So even if they are not based in your country (maybe they are on the mainland), you could contact one of them and rent the scanner from them for a project. If you decided to invest in a scanner you could see a return on investment just from the sheer amount of BIM deliverables you would be able to create. The cost of a 3D laser scanner today is very affordable and the return on investment is much faster than before. It's very similar to the Total Station. It's a very similar level of investment so I think if you have a need for it today then you could certainly make it into a very profitable business in a short period of time.

 

Q:  Could we also program material composition by layer or thickness?

A:  In the Trimble RealWorks software we do not have the ability to add annotation to the piece that's being modeled. We can, though, export it in different layers. So if you model the first floor you could export all of those geometries out as say the first floor. Or if you wanted to just export all of the beams you could export all of those. You just have to put them in different layers. But it doesn't add information about it. It doesn't tell you, for instance, if there's some sort of fire resistance on the beam you wouldn't be able to annotate that in the Trimble RealWorks software. That would be something you would do in your BIM software.

 

Q:  How much does a point cloud weigh for an entire building say 10,000 square Meters when you export it to another format to analyze it in a software like Revit, Tekla or Digital Project?

A:  It depends on what you export. If you export a 3D model it's quite light. You're only talking about MBs of data. If you want to export the whole 3D point cloud the potential is that you have GBs of data. If you only export small sections then it is MBs of data which is very manageable. It comes down to what you want to take across.

 

Q:  Is the software intelligent to the point of selecting the slice so it obtains the height and depression on the floor? In other words, how do I know everything is not above where it should be or everything a depression, or do I just set coordinates?

A:  When we were doing the floor flatness it took the level plane or the design model depending on what you want to create. It compares that to the 3D point cloud. So it's all based off what's in the same coordinate system or in the same reference. And when it did the comparison it knew where level should be so it knew "zero is here" and it automatically was able to compare what is below and above where the flat plane should be. It knew where the depressions were and where the high points were.

 

Q:  How precise can I place columns in a point cloud? Do we take the same tolerance you spoke of earlier? The 0-5 mm tolerance?

A:  The 3D point cloud is accurate in its position. When I model through the point cloud obviously I'm fitting something to some points. So if I just have survey points I'm fitting it to some very small amount of points. With a 3D point cloud I can see the full structure there. So when I fit a beam through the point cloud it gave me an RMS value and told me how well it fit the beam to the actual point cloud itself. So it gave me a variation report. Because you are actually fitting something. And the reason for that is the point cloud actually could represent a beam that could be sagging for instance. So when we model the beam it actually models it straight. So then it would show us in the RMS value the point deviation from the model to the point cloud and highlight to you if the
beam was actually deflecting in that specific example. So you look at that RMS
value to make sure that it is fitting within the tolerance that you're asking
it to fit in.

 

Q:  I work in the residential design field in the town of Sun Valley, Idaho and use SketchUp Pro along with ArchiCAD.  Would laser scanning work for an existing stone farm house or log cabin, or is it scaled only for larger projects?  Please see photo examples below and advise.

 

can laser scanning be used on small construction projects

 

A:  To answer your question about 3D laser scanning for an existing stone farm house or log lodge, the answer is yes.  The Trimble TX5 is small and light weight and is ideal for small and large projects. The laser that is used in the system is tuned to pick a wide variety of different surface types, rough exterior like stone or smooth like the log
interior in the picture you provided.

 

 

For more information on Trimble Laser Scanning hardware, software, and services, please utilize these resources: 

Website: Trimble Laser Scanning

Webpage: 4 Applications for Laser Scanning

Webinar: Best Practices in Laser Scanning

Dealer Locator: Name the hardware you're interested in and your location to find our authorized dealers in your area