subassembly that accounts for road slope to get true 2:1 slopes

[JamesMaeding]

Peter Funk and Chris Fugitt helped me on this a long time ago, and I can't find what they did.

So I have a road with big vertical curve so its slope along centerline is constantly changing.

I need to daylight the road to a surface, and I need the side slope to be true 2:1, which must account for the road slope.

If you do simple 2:1, you do not get 2:1slope, with respect to the downhill direction of the slope.

This is all because sections are cut perp to a road centerline, while a side slope downhill direction is never perp to the road unless at 0 slope.

thanks

[tdivittis]

I realize this is quite a late reply, but I never get time to peruse the forums.

The solution to this issue is tricky, at best, because in order to get a 'true 2:1' slope, you have to deflect the slope off of the perpendicular line generated by the corridor.  Short version:  It is geometrically impossible with a corridor.

You should be able to get a pretty good approximation by changing the slope based on the instantaneous road grade, although this will fall apart in certain circumstances, particularly steep road grades and tight curves.  If I ever fiddle with putting together a custom slope subassembly, I will try to remember to post it.  In circumstances where I need the slope, I have other tools that I use to generate the slopes, so I've been able to skirt this issue, for the most part.

Hope that helps anyone else looking at this, at least a little bit.  I spent more hours dealing with this issue than I care to admit.  It is a subject that does not come nearly as often as it should in the civil design world.

This comment is in no way intended to be critical of corridor functionality, they do exactly what they *should* do.  Our collective ignorance is another matter, entirely.  😉

[tdivittis]

Oops.........read the date wrong, I guess that is not late.

[JamesMaeding]

You are correct, as we are kind of "sampling" sections at stations, but the formula to adjust for road slope is actually simple as I recall.

You would think sin and cos would be involved, but it was much simpler, like terms cancelled out a bunch.

I don't think Autodesk ever internally defined what corridors were supposed to do, and I still do not think Dave's team has a clear vision.

Its one thing to design a software to be "dynamic", which really mean parametric, but another to forget what the end result is used for.

In particular, labeling in plans.

You cannot "label" a corridor or even share it (maybe you can share it now, does not matter), but only its surface.

Surfaces are not appropriate to use for road construction doc plan and profile, except for spot shots in detailed pavement areas.

In other words, you do not label top of curbs with surfaces unless you have some oddball situation that someone always comes up with, like a perfectly straight road.

You use alignments, but Autodesk misses that, as is obvious because they make it so rediculous to just make a style that labels station and elevation.

Come on, I have to pick my profile to tell what to use for the elevation?

They dropped the idea of a current align and profile, which could use data without having to "see" it on screen, and thus destroyed the chances of getting the land development world in their corner. And so it goes.

The upside of all this is people that can make their own tools look like heros, though they are just doing what should have been done by others.

[tdivittis]

From my perspective, corridors do exactly what the tried and true design process does.  We used to 'cut sections', drawing 2:1 slopes, in the section perspective, which will never result in a 'true 2:1 slope', unless the road grade is 0.

Geometrically, you simply cannot do what you want to do with road corridor geometry.  The geometry breaks down rather quickly.  I am not suggesting they shouldn't have some method of dealing with this, because I believe they should, but the reality is that it cannot be done with corridor based geometry and that is not their fault, it's "how we've always done it".

I'd be happy to share more information on what I have with you, but I'll caution you that it is not in any way, shape or form an elegant solution, it doesn't even pretend to be.  It does generate true slope lines from a 3dPolyline, however.

I hope I'm providing useful information, not trying to pontificate, or anything.  (This ends up being a pretty complicated conversation.)

[JamesMaeding]

I can tell you have your head on straight, good comments indeed.

I think it all depends what design process its being used for, as I agree it does automate the "hooking together sections" type design.

Its programmable too, so does it better than anything I have seen. The Bentley guys might argue that, and a few others.

I guess I just always thought the C3D team would come up with something more useful in plan production.

That arena is still ruled by alignment objects, and their management which I have yet to see a company do well at.

I'm guessing you have also encountered the "land is expensive" world like we deal with, where you draw contours at exact offsets using 2d drafting.

You waste land if you don't, but then have to be aware of other issues also when things are that tight. That is just land development.

The daylight lines are not exact, as they are only perfect where contours cross and so on.

Anyway, yes, a corridor type object can only approximate a decent true slope.

It has been done though, I have it somewhere in a custom assembly and its not that complex.

I'll page Mr Fugitt as he was involved. I think Peter Funk gave the formula though.

While on the subject, I also took a shot at computer generated "true" slopes. I call it the slope growth algorithm.

And I never got it right. There are really tricky cases where the slope folds back on itself which you must deal with.

My current need is not that demanding though, mush simpler cases involved.

[tdivittis]

I was dragged into the 'land is expensive' world kicking and screaming...lol.  I wanted no parts of it!

I would echo much of your sentiments, generally, I think.  To this day, I still like what Eagle Point's roadway design tools did compared to C3D's.  The overall package I would never want to go back to, but it is surprising to me some of the functionality that C3D 'missed'.  Some things are (in my humble opnion) unnecessarily over-complicated to account for scenarios that simply are not often encountered, at the expense of things that 'we' do in virtually every project.  (I like corridors overall.......love/hate relationship, maybe?)

I would love to see what they had come up with.  I have been convinced for some time now that it wasn't possible within the inherent constraints of a corridor.  I would love to be proven wrong on that, for sure.  (Not tooting my horn, but I've done a couple of 'written from scratch' subassemblies for stream mitigation, that change geometry based on distance from PCs/PTs/PVIs, etc., so I like to think I have a pretty fair grasp of possibilities.  I couldn't come up with anything better than an approximation that would still be unusable in tight horizontal curves.)

....sorry for getting long winded.  Like I said in my first post, I'm surprised how rarely this subject comes up.  In many cases it is far from a minor issue, not just with corridors, but grading/stepped offset, also.

[JamesMaeding]

You are correct, it is not minor at all, and is an elephant in the non-existent Civil BIM room.

I have major clients asking if we can come up with an interactive road-slope-pad model where things can be shifted around and things "adjust".

That all comes down to controlling slopes, and hooking them together to intersect each other.

Sometimes you want a pad edge to hold, and you want to see how far you can push a road towards it horizontally.

All that is so complex when you get down to the decisions, and there can be multiple solutions sometimes.

No one has come close to such a model, and I have seen the siteops and ASE civil progs through the years.

Its like no one even bother bringing it up anymore, its done by hand, which may explain what you observe.

[jeff.wilkerson]

I believe that the Autodesk University 2011 course "Create Subassemblies That Think Outside the Box With Subassembly Composer for AutoCAD® Civil 3D®" has an example of what you are looking for. Check out example #3.

[JamesMaeding]

yah, that was exactly what I was looking for. Good stuff!

I got a kick out of this statement in the pdf

"I kindly pointed out that the slope is drawn based on the
road cross section which scaled as 6.00‟, as expected. After
much discussion there was just no agreeing to be had and so I
did a quick and dirty fix..."

How in the world did the author, Kati L. Mercier, have pretty awesome civil3d subassembly skills, yet not know all slope max's are measured in the downhill direction?

Wish she was my co-worker though, what a great pdf and class she did on this.

I can't figure out here dg name or would @ her to say thanks.

[Jeff_M]

@JamesMaeding, @KMercier_C3D is probably the most knowledgeable person I know for working with the Subassembly Composer.

[KMercier_C3D]

It's funny to me reading back through old papers that I did and the things that I have learned myself since preparing them. Glad it was of help, even if the PDF is quite a few years old now.