How to represent a slope wrapping around an object

[Anonymous]

Hi everyone. I'm trying to model a concrete caisson for land reclamation purposes. The thing is I want to represent its backfill, which eventually wraps around the caisson and ends touching its front face. I've tried to recreate this using the grading tool, but have not been successful. I'm attaching a simple image of the caisson with the contact areas of the slope, which by the way, is equal along all of the faces.

I am not a native English speaker, so if you need any clarification on the issue, feel free to ask.

Thank you very much in advance!

[Anonymous]

No worries, Joe. That is correct. The backfill platform is fixed at Z=+2 m. From there, there should be a constant slope of 3:2, or 1.5:1, however you want to put it. So in this exercise, due to this condition, the meeting point of the slope with the foundation should be located 15 m away from each corner of the caisson.

This is driving me mad! I find this to be a feature in which Autodesk should make C3D smarter, mostly because BIM is becoming so important. This is real BIM 

[Joe-Bouza]

Good morning E

Modifying the Feature line of BFP should be fairly straight forward.

If you find the solution to the original post solved (cant project a draped line to the same surface) Please select "accept solution" so other readers may be aware of your solution

[Anonymous]

As I stated before, I'm finding trouble when trying to solve this. The feature line causes the grading slopes to jump. Am I doing something wrong? I attached some screenshots on my previous post. I apologize if I'm not understanding your solution at the moment.

[Joe-Bouza]

No worries. I guess I may still be missing what exatly you want. what exactly is Z+2m at the back? And what is the distance of the flat area you wrapped around the side. Where is the jump you do not like?

[Anonymous]

Thanks. The flat area is only required on the back face of the caisson, to guarantee there is no overpressure on the structure. The Z=+2m refers to the elevation, which is located 2 metres over the reference level (still water level = +-0 m. The width of the flat area = 5 metres. The grading must be done with those conditions, making the slope wrap around with the 3:2 slope, which is the material's internal friction angle. Does this make sense to you? Thanks again

[Joe-Bouza]

So the top of back fill = 2 m, yes?

this is 2m top BF,

5m flat wrp on side

3:2 from 5m flat elev 2m @ 3:2

that puts you 4m  into the cajones  

[Anonymous]

I'm attaching a new image on how it should look, partially... I see you inserted a vertex on the feature line around the centre of the sides of the caisson. That shouldn't be there; instead, the slope comes down directly from the back of the caisson @3:2 until it meets the front face of the caisson. I've marked with a red line the path it should follow. Comparing this with your latest screenshot, you'll note that there is a jump on your slopes.

[Joe-Bouza]

Its not a jump. The grading runs the slope projection perpendicular to the feature line. and the contours are calculated from the resulting triangulation. As you are decending and turning at 3:2 and projecting away at 3:2 it going to get steeper. If you calc the contours on paper you'll get steeper as you turn.

IMHO what it tells me is the design  has to adjust the maximum allowable slope to account for the geometry. Think about it; If you profile a road at the center line and the road curves the edge slope of the interior curve is steeper and the exterior curve is less steep. You have to acount for that. 

[Anonymous]

Hello again, Joe.

I don't really agree with this. As I said before, the slope is set during construction, but it ultimately depends on the material's internal friction angle. After grading, sooner or later, the material will adopt its natural slope for stability purposes. So no, there can't be steeper slopes than 3:2.

I've done a rudimentary approach to the desired solution so you can see what I'm talking about. To do so, I had to grade a second time, but on this occasion, along one of the shorter sides of the caisson, so that the slope wraps around the edge and down to the foundation. The idea is to find a simple way of solving this to ensure accurate measurements and drawings.

BTW, I've only done it on one side.

Thank you.

[Neilw_05]

I wish I had time to put a video together but for now all I can do is try to explain. My approach is as follows:

1) Limit the grading that projects from behind the caison down to the ground so it does not pass through it. Instead make a gap in the grading so it passes along the sides.

2) Drape feature lines on the grading along each side of the caison so they match the slope passing by. These featurelines are perpendicular to the downhill face of the caison.

3) Create gradings from these 2 featurelines using the desired slope (i.e match the slope running past the sides) so they project across the downhill face of the caison. The target for these gradings can be an elevation or a surface depending on your goal.

Additional gradings and infills can be used to fill in any gaps behind and in front of the caison. 

[Joe-Bouza]

I never said steeper and would never suggest that.

Either way you seem to figured a way to maintain 3:2 around the sideslope. Excellent. Its not clear to me how you did it?

Back to your original post. Your problem was that the grading would not project, not that the projected slope was uneven (I realize we got side tracked after the projection issue was solved). So your orignial issue has been solved by removing the portion of the target surface from the area the FL is draped, yes?

Please select "accept solution" so other readers may be aware of your solution

[Anonymous]

Sorry for the delay, but I've been very busy these past days. I will try to put in practice Neil's solution first and then I'll bring my conclusions with the acceptance of the solution.

Thanks

[Joe-Bouza]

No worries. I hope @Neilw_05 can do a screen cast to help. I didn't fully understand his approach

[Neilw_05]

Will do Joe.

[Neilw_05]

Here is a screen recording showing the concept. The link to the live dataset is expired so I created my own for illustrative purposes.

Forgive the crude screen recording. I didn't put a lot of time into it. Hopefully it gets the concept across.

http://screencast.com/t/qWBEloKxcQSZ

[Joe-Bouza]

Very impressive Neil, but I still don't know how you got the grading to wrap and make 2:1 perp to the contours?

[Neilw_05]

t is not done with a single grading Joe. The 2 oieces across the downhill face of the caisson are idividual gradings attached to the magenta featurelines in the video.

[Anonymous]

Thank you very much, Neil. Nevertheless, have you seen the 29th post? I attached some screenshots with the results I'm willing to get. I'm reattaching the dwg file with my rudimentary approach, so you can access it. Perhaps you can tell me a way to reproduce such solution in a more efficient way.

Thank you

https://goo.gl/dEf6eW

[Joe-Bouza]

I'm confused. Did you do it or not? your screen shot appears that you have achieved your goal. I missed something along the way

[Anonymous]

Unfortunately, I did not manage to do it. Note that I had to grade two times. I'm attaching a screenshot so you can see there are two surfaces. The idea is to have a single surface.