Topic: [Autocad 2010] Failure to.. completely... Sweep

[JCTER]

Can anyone sweep the magenta triangle along the green polyline?

I have tried it with the Alignment setting either On or Off, during the Sweep command, and both results are the same.  It appears to be something to do with the geometry of the contour, but I cannot figure out what it is.

I want to get some basic volume calculations to be able to ballpark a figure, and this is a rather... integral... part of the process.  As you can see in the attached screenshot, it's like the swept solid only partially gets swept, but then continues generating "mass" towards the end.

In the attached .DWG file, that I have since the screenshot edited, you'll find the swept solid on layer 0, gray, and you'll see that if I move one of the vertexes (at the center of the yellow circle I drew to indicate the point of deviation) away from the original green path... you'll see it will generate the swept solid further on down the path until it stops again.  I'm not sure why this problem exists.

I ran a "massprop" report on the solid before and after moving the vertex and it does generate different volumes consistent with the change I saw on screen.  This indicates that it is not simply an error in visualization or video card, but that the solid is actually failing to be generated... partially.

Anyone ran into this crap before?  Is there anything I can do to remain "on path" and still get this thing to generate?  I may have to end up "fudging" the path from the surveyed topo contour line and trying to account for it in my report.

[Bob Wahr]

Something screwy with it.  I killed the vertex it "ended" on and it made it farther befor stopping again.  Not sure why it's doing it though.

[JCTER]

Something screwy with it.  I killed the vertex it "ended" on and it made it farther befor stopping again.  Not sure why it's doing it though.

I'm thinking it has to do with the angle between segments, or possibly segments being too close.  If I move the vertex before the one it ends on, to change the angle between segments, it'll go about twice as far.  I think I'm just going to fudge it so that I can keep getting work done, and not get too hung up on this, but man, I'd like to find an alternative or solution to this that is better than altering the path.

[mjfarrell]

does altering the SURFTAB settings have any effect?

[JCTER]

Thanks for checking this out though.

I had never seen a PARTIALLY completed solid.  I found it pretty strange.  Before today, I've only ever seen a GO/NO-GO situation where it either did it or it failed.

does altering the SURFTAB settings have any effect?

Command: surftab
Unknown command "SURFTAB".  Press F1 for help.

I searched a bit, but dunno what you're referring to.

[JCTER]

SURFU
Sets the surface density for PEDIT Smooth in the M direction and the U isolines density on surface objects.

SURFV
Sets the surface density for PEDIT Smooth in the N direction and the V isolines density on surface objects.


Them?

[mjfarrell]

surftab1

and his little buddy

surftab2

[mjfarrell]

or the new ones in 2010, or maybey all fours?   

[mjfarrell]

also verify you have no zero length segments

[JCTER]

surftab1

and his little buddy

surftab2

Tried changing those, and the ones I mentioned, and at max and min values, no change was discernable between it.  I think they only apply to surfaces created, whereas I'm using solids.

[JCTER]

So in order to check for zero-length segments, I did a pretty roundabout way of doing it since I hate LSP right now.

Took the "LIst" output from the pline and copy/pasted it into notepad and saved that as "test.txt"  Then I created the following .php file, and uploaded them both to the webspace I have and opened the php file in a new tab, yielding the first 14 characters of each line.  Took that, pasted it into excel all in one column, then in the second column that had a conditional "=IF(A1=A2, TRUE, FALSE)" copied on down the line.  

End result... no zero length segments.  Every point has a unique "X" value, so there can be no zero length segments.

How's that for a "rigged" way of coming to that solution?

Code: [Select]

<html>
<head><title>I can't believe I am doing this</title></head>
<body>
<?php
$add = "</br>";
$fh = fopen("test.txt", "r");
while (!feof($fh)) {
$line = fgets($fh);
echo substr($line, 0, 14) . $add;
}

fclose($fh);
?>

</body>
</html>


[JCTER]

End result... just tweaking points until I get the whole thing to generate... man I wish my computer was faster.  Each point move requires 2 different times where I have to wait a good 4-5 seconds of annoyance.

[JCTER]

I have -no idea- why this solid failed to generate.

I took a couple Screenshots along the way, just to show...

It has -nothing discernible- to do with the angle between segments.  I'm flummoxed.  In the SSs below, you can see at one point it's very sharply turning acutely and in another case quite quickly turning obtusely, and it had no problem... however, it will stop at a point where it's almost a straight line through the next few segments!

I can move a point on this TWO THOUSAND FEET LONG polyline as little as ONE TENTH OF ONE INCH in the X or Y direction, and the solid will generate the next segments fine, until it decides to stop again.

I just wish I had an explanation for what's going on.  This is the strangest behavior I've ever seen of it's kind.

[Strucmad]

I don't have the sweep command available to me.. (version 2006).

I tried extruding along the path but it crashed as well.

However after editing the Polyline to a Spline, the extrusion took.

Don't ask me why..I dont have a need for 3D solids much anymore..

Here is a screenshot, Note I was playing around with a very low Isoline setting.

[JCTER]

I don't have the sweep command available to me.. (version 2006).

I tried extruding along the path but it crashed as well.

However after editing the Polyline to a Spline, the extrusion took.

Don't ask me why..I dont have a need for 3D solids much anymore..

Here is a screenshot, Note I was playing around with a very low Isoline setting.

Thanks for that!

I didn't think to try "Extrude/Path" because I got so used to sweep... I will have to see if it behaves differently when I get back to the office in the morning.

I actually -did- convert the pline to a spline and try 'sweep' though, and it DIDN'T work.  Very weird.  I'm thinking about submitting the file and description to Subscription Tech Support, or whatever they call it, just to see what they say/do about it.  It ain't gonna help me/us any, as I need the results by 8am tomorrow (gonna be an early morning tomorrow... ) and this solid was just a stepping stone.

I did get the solid completed, just about, via fudging the points here and there.

Thanks for the alternate PS_Port.  Something to consider when trying to better understand how this beast works/thinks.

[Shinyhead]

I gotta ask, wtf is that particular object for?  looks like a drunk triangular exhaust pipe.

[JCTER]

I gotta ask, wtf is that particular object for?  looks like a drunk triangular exhaust pipe.

It actually -is- a drunk triangular exhaust pipe!

Seriously, it's a river bank that a company is regrading and fortifying so that they can build a casino atop it in Baton Rouge.  I need to provide some estimated volume calculations based on a few design options.

[Shinyhead]

Ah ok. heh.

[mjfarrell]

I gotta ask, wtf is that particular object for?  looks like a drunk triangular exhaust pipe.

It actually -is- a drunk triangular exhaust pipe!

Seriously, it's a river bank that a company is regrading and fortifying so that they can build a casino atop it in Baton Rouge.  I need to provide some estimated volume calculations based on a few design options.

you know; somehow I thought you were using PAINT for your 'CIVIL' design needs ?!?