Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Opening point cloud in SolidWorks

Status
Not open for further replies.
Replies continue below

Recommended for you

The middle coordinate in the csv file are all 526. Y is up in SW and Z is up in most survey equipment. It seems you may need to re-order your coordinates for it to appear as you would like in SW.
 
Seems like even if you re-order you are still going have one vector that is always the same, instead of a flattend plan view you will have a elevation view with no depth into the view.

I think your .csv file has some data issues.
 
If you keep scrolling down in the text file you will see that the y values are not all 526. The csv file doesn't look like it is messed up either.
 
Yes I see that now. I had only looked at the first couple of pages of points. So no data issues, agree. It does look like the coords in what would normally be the Y position is you elevation change and the first and last coords within a line are your plan view coordinates.
 
I looked at your data set in more detail. Are you sure its is all on one plane? Because it looks like you are looking at an area that cover a piece of land that is 243163 units (whatever they may be, feet inches, meters, etc) by 244956 units and is only 58 (edit: was 584) units deep.
 
However your data set does appear to be missing the coords for the 530, 568, 570, 578, & 580 elevations
 
I just realized what I need to do. I need to add the lowest elevation value to all the rest of the elevation values.
 
So your elevations cannot be relative to sea level? I guess you normalize the elevations so the top of the quarry is at elevation zero. ??? I've never done this type of work so I am learning also.
 
I had to import the points from a 2d sketch in another program. Unfortunately the contours lines at the different elevations where not on separate layers. I moved the contours lines to appropriate elevation layers. SolidWorks imported the sketch, putting the different contour lines into different sketches. I then used a macro to export the points in each sketch to Excel. I then just added in the elevation data to a third column.

ps. I still ran into problems. I've attached the text file with the corrected elevation values.
 
 http://files.engineering.com/getfile.aspx?folder=a774efed-abbc-4eaa-b5c5-fd4acccbdb9a&file=cloud.txt
I looked at that too quickly. I tried it a bit now, and I think your elevations are exceeding the solidworks limits. When I set the import units to meters (scan to 3D option), I keep getting an error that the model exceeds the -500m, 500m limits. Using mm as the units, it imported like in the picture. In a side view I see different elevations, but I am suspicious they are not correct. I have never tried working with our point clouds in SW before. I was told clouds exceeding 1million points are not easily manipulated in SW. 1million points is a small point cloud. You must have filtered this quite a lot. I am not sure this is a good problem for SW.
 
 http://files.engineering.com/getfile.aspx?folder=bd825032-ce6f-4eda-bc61-1711d460e92a&file=SW-pln.PNG
I'm not sure what else to do. I need a model of this quarry for work.
 
I checked the data in the picture I posted earlier. It appears to be importing correctly with the units set to mm, so maybe you can use a scale factor. I would include the SW file, but the point cloud disappears when I re-open the file. A bit of googling shows that is not uncommon. This would be very easy in Civil3D.

Hopefully one of the SW gurus has a better idea.
 
The elevation still does not look correct even if the units are set as mm. Here is a Google image of the quarry.


A little background information on how I got the points.

1. I was supplied a contour plot of the quarry. The drawing is a 2d sketch.
2. When I click on a contour line, I can see what elevation it has in the properties box.
3. I move each contour line to a new layer. Each layer represents a particular elevation.
4. I then explode all the contour lines (because they are polylines at this point).
5. I then open this dwg file in SolidWorks and set the option to have each layer in the dwg as a new sketch (hence the reason why I separated each different elevation contour to its own layer).
6. In each sketch I fit a spline to the points and simplified the spline geometry.
7. I then exported all the points in each sketch to Excel using a macro I found online.
8. Here is where I think I am messing up. Since exporting the points in a sketch only gives the planar coordinates, I have to manually insert the elevation data in Excel.

Example: I click on the sketch containing the contour lines for elevation 540ft. I export the points to Excel. I add a column of 540. I repeat this process for the other sketches.

The drawing was made in progeCAD and you can get a free trial download on Google.

I'm starting to think that maybe my elevation values I added are incorrect.

I have included the dwg file if anyone is willing to open it up in progeCAD to see the contours.
 
I still think it is a unit issue. Unfortunately I cannot import the cloud. But based on the cloud points that Brad posted, the Google map view and your 3D sketch it looks like the basic features are there.

Capture_qakovz.jpg


I can see the ridge (red), the water is kind of there (blue).

Are you sure of the contour intervals? As I stated earlier, it appears that you are working with a land area that is some 243,000 units by 245,000 units, yet your depth is only 58 units. That is a depth that is approx. 0.02% of the length and width of your plot. With this ratio the cloud is going to look pretty flat.

Are you sure your contour intervals are 2 units?

Based on the Google map the quarry looks to be about 2000ft across in the east/west direction. This seems to imply that your plan view units might be the issue. 2000ft times 12 inches per foot = 24000inches. This go so happens to be about 1/10 of you plan view coordinate size. Is it possible (doesn't seem like) but could your plan view coords be off by an order of magnitude?
 
I too was thinking about the contour intervals being off late last night. This would make sense since I put the intervals in manually. I am going to attempt to scale my interval offset and retry.
 
Another thing that makes me question units is that again based on the Google map (and again admitteldy this is just a guess) but it apperas as though the depth is about 1/10 of the width. There is our order of magnatude again. ???
 
This has taken you some time. I take it the surveyor would not provide the raw coordinate file? We have created TIN's or meshes from contours too. I see how you are doing this, and I would make sure you have not made any simple errors in Excel. If you have to do this in the future again, I suggest you convince your employer to provide you the correct software to do so, or sub this to someone who does.

I imported the same file into recap and then CAD. It looked about the same. I can see the outlines of the contours in the points, and the outline of the quarry walls are apparent, but I have no idea if the elevations make sense.

We have the scanto3D add-in, so importing points into SW is simply opening the text file.
 
That is interesting. What I have done is exported all the points from all the sketches and added in the appropriate elevation values for the contours. Instead of saving the Excel file as a txt file and opening it in SW, I am using a macro to import the x y z coordinates from the txt file into a 3D sketch. What I am noticing in the parameters dialog (while the points are importing) the y parameter value is not 526, 528, 530...etc. The values are in the 1000's. I'm thinking that SW is automatically scaling the y parameter value? This is still running but I will let you know if I get something looking good. The whole purpose of using SW is to possibly export the "new" 3D points and use them in another program to generate my surface. Then import this surface into SW to use in my model. I'm not sure if SW has an automatic point cloud to surface tool.

Edit: I asked for the survey data and they do not have the 3D point data. Hence my frustration of manually adding in the elevation data.
 
I keep coming back to the idea that on the Google map it appears that this quarry is in the range of 2000ft across. Maybe it is really 2400ft across in which case maybe your 243,000 by 245,000 unit area has the decimal off by two orders of magnitude (shift decimal two places to the left) and your plot area is now a 2430units (feet ???) by 2450units.

I still feel like there is a countour elevations unit issue relative to your plot plan coords as it seem like your cloud is just very flat, little elevation change. I know I am repeting myself but are you sure those are 2unit interval and not 5, 10, 20 ...

20 & 2, there is that order of magnitude again

And you did say that you had to add the elevation data. Who gave you the plot data, maybe you can contact them and ask what the countour increments are relative to the plot coords.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor