One way to deal with the bench overlaps is to project them out at 0%. Then the overlaps don't matter and we can use them to a create reference surface for the next tier. Each tier could then be a separate alignment. By sampling the reference surface from the projections we can establish the profile for the next tier and then apply a 0.2' vertical offset from the profile to establish the bottom elevation for the tier (0.2/10=2%).
With this approach we can extract featurelines from the bottom and tops of the tiers and let the TIN handle the corners. There is still the problem of the truncated corners at the convex angle points, but if we project the benches far enough we can ensure a surface for the next tier. The one thing I'm not sure about is how to handle the portions of the tiers that do not have a reference surface from the previous teir.
Can this work?
I realize now that the zero % slope idea won't fly if the top of wall is not level so I'm back to the drawing board.
Just wanted to let you know that I'm pretty close to having my idea done. It is drawing walls and checking a bunch of stuff. No overlaps, it did turn out to be a bit more complicated than I thought regarding checking conditions. For example, when I get to the back of a wall I am looking out 10 feet at the 2% grade and then checking if there is enough vert height for the next wall, if not I am just doing a 2:1 from that point. Otherwise you could have a scenario where you go 10', get to the next spot and then go 2:1. I ran it this morning and found one bug, as soon as I get that fixed I'll send it your way. I'm working on it in the evenings so it probably won't be until tomorrow, just learned I have to do a topo at low tide today and that is at 6 pm..
Sorry for the delay,
Thanks for looking at this Mark. I'm looking forward to seeing a solution.
I have come to realize I need to correct one of the citeria. If the depth between the base of wall and EG is less than 5' we still need to add a wall. Otherwise the 2:1 slope will chase the EG slope well up the hill. So I'll need to add tiers until the walls go into a fill condition.
You are right Steve. I think we'll need to require a min 3' high wall with 1' min, 2' max. above EG at the top. I can model a swale behind it with gradings to convey any trapped surface runoff.
In cases where the min/max constraints are not met, then go hopefully it will catch grade in a reasonable distance if we project at 2:1 to EG.
I think I'm really close to getting this to work. I just need somone with expertise to help me fill in the gaps.
Here is what I've come up with:
1) Use offset alignments for each tier.
2) Create a bottom of wall profile for the first tier.
3) Create a feature line from the profile of the bottom wall
4) Create a stepped offset featureline from the bottom of wall feature line at a slope to correspond with the slope between the bottom of each tier. In this example the slope will be 52% (5' +10'x2%)/10'. Make sure the offset FL is beyond all the tiers so we can make a surface to use as profiles for the tiers. I found I needed to add featurelines that span the tiers across the angle points to get a nicely mitered surface.
5) Create a surface from the featurelines and sample it for all the tier alignments
6) Create an assembly with conditional branches to add a 5' vertical tier when in cut and no tier in fill
7) Create a corridor and add all the baselines and apply the assembly
I am able to get the corridor to build the bottoms and tops of the tiers but I don't know how to make it fill the gaps along a tier when it transitions from cut to fill. I'm sure it is an easy fix so I'm hoping one of you will have the answer. Once I get the corridor to generate all the featurelines I think we can build a surface from them.
To change the parameters of the walls we would change the slope of the reference surface by adjusting the slope between the bottom and top reference surface featurlines and/or adjust the horizontal offsets of the tiers. To aid in determining what the ideal slope will be we can create a grading from the bottom of the first tier that targets EG and tweak the slopes until we get a good fit.
If this works it will be a very simple solution and it doesn't require fillets on the baselines.
I am attaching my design file for reference.
Here is mine. It isn't working as good as I hoped but it does calculate the wall toes and tops plus daylights and will not overlap. It looks funny when it goes from a wall section to a cut section (because there is less than 3 vert feet to eg.) Another thing that really bugs me is that you can't do a corridor shrinkwrap with it, not sure why.
Anyway, here is the procedure to use it.
1. load the subassembly
2. stack a few of them back to back (see the pic.)
3. put some small curves in your alignment for outside angles (I used 0.02' R)
4. put in the "v" breaklines for the inside angles
5. build the corridor and target the ground, and your lines.
Then you can mess around with it as much as you want, it will update. But unfortunatly, in the end you will have to do some custom grading.
You'll see that it is still a work in progress, there is a feature line that is connecting that shouldn't. Open up the pkt file with subassembly composer and look around, pretty much everything has comments. I'm sure somebody could it it in 1/2 the steps, but hey, I had a fun time messing with it. I'll probably mess around with it this weekend. Let me know if it is handling something completely wrong and i'll make the change. I see that I have more benches than you but when I check the cross sections they appear to be correct.
The capture 3 image shows all of the variables you can change.
I've taken a few minutes to work with your assembly Mark.
I'm still having problems with overlaps on the benches. I've added small fillets to my baseline and tried adding sections to the mid points of the fillets but I can't get rid of the overlaps or get mitered corners. I don't understand your guidance to "...put in the "v" breaklines" and "target..your lines". Could you clarify?
I've put my approach aside until I figure out how to get the corridor to build a surface from the feature lines. So far I haven't been able to make it model the faces of the tiers.
Log into access your profile, ask and answer questions, share ideas and more. Haven't signed up yet? Register