Geotechnical Models - Seequent / Leapfrog

Does anybody have experience in importing files (surfaces) from Leapfrog?
The current outputs I’ve got are *.dgn but they import as Shell3D - I presume I’ll have to use RPS Explode Solids?

I’ve also tried 3d DXF with top surface & bottom surface exported seperatrely but no boundarys come through some I’m going to have to create a lot of boundarys (there are 30+ regions)

Any suggestions?
BTW its for a 12km Roading project through about 6 differing soil types with minimal Geotech at the moment and I’m going to have a lot of fun with unsuitable material!

Exploding the dgn shells is the way to go and rename them

RPS has a specific explode for shells. “EXPLODE IFC SHELLS”

Thanks Gavin & Francisco - Exploding Shell was still running after 1/2 hour with no feedback so I’ve stopped it! I’ll try again…

My thoughts, if you didn’t want to go the way of Explode Shell, would be to use the 3D DXF information to use that information to create bndy’s. You could bring in the 1 you need and use the change elevation command to make that linework 2D. This will help speed up the process of creating them.

Open a new project and import the data you want to use. Use the Track Region command to create the bndy’s you need. Use the Copy / Paste to bring those into your project. Then you would be able to create surfaces from the individual areas you need from your 3D data.

Track region should cut down the time needed to create your bndy’s.

I would use RPS Explode Solids and use that to separate the Top Bottom and Sides, my experience with Leapfrog is limited but I have manipulated one data set using Explode Solids and it worked pretty well to create the Strata Layers for the Volumes needed.

If you want to share the DGN / IFC file we can look at it also

Alan

Use Explode Solids v Explode IFC Shells - we have not really worked on IFC Shell command we only focus on Explode Solids so that should work better - if not send us the file and we can see what is going on

Alan

Hi Alan et al,
The Explode Solid is “hanging” after 2 hours of processing so I’ve crashed TBC,
3d DXF is working (just) but very tedious process creating boundaries

The Shell3D has about 892000 Triangles (2676000 vertices)

GM_-_Q1a_Holocene_Alluvium.dgn (8.8 MB)

Hey John

took about 20min on my computer to explode here is the file
john mills.vce (39.6 MB)

1 Like

also use the shrink wrap command for internal / external boundaries

Thanks Gavin,

BTW - I started with the largest file
Loess dgn 17.2mb - ID10T error to jump into the deep end.

I’ll need to explore how the Designers create models in LeapFrog and if possible reduce the size of their outputs - I suspect it’s because their Original Ground model is large and its replicating that.
There are only 20 or so Geotech sites at the moment so it does not need to be so detailed would be my thoughts

I’ve worked out a solution,

Using 3d DXF from LeapFrog for each surface
I’ve imported that into GlobalMapper (it handles large datasets a lot quicker)

Then right clicked the that data set (Layer) and
Layer - EXPORT
Selected LandXML format
Imported into TBC - it respects Regions and creates surface (there are some Surface error flags but I’m ignoring those)

2 Likes

Good to hear you solved it - when we explode the solids we then take time to de duplicate the linework (each triangle has 3 lines, side by side triangles create 2 lines along the common face - the De Dup process is slow, and slower with more data that you throw at it - I do it piece by piece because it is faster because it doesn’t have to compare every object in the file with every other object in the file. Appreciate that we all just want the “Easy Button” i.e. select all and fix it for me - however you have yo be cognisant of how the processes work and what they actually do - in this case the Explode doesn’t take that long, it is all the DeDup that takes the time. On big data sets that takes a lot of GPU and RAM as well as process horsepower to get er done - doing a lot of data on a lighter weight machine is going to take some time to sort through all the data - draw some rectangles and select by polygon and process each area separately - you will find although somewhat manual it takes less time - we could do that programmatically and multicore this piece I guess also - will put it on the Chrstmas List!

Alan