I have looked up some info about non-dimensionalizing. It could be worth the effort, but it only looks like a lot of work. Is there a way Maple can do this for me, or is it just a lot of hand solving?

I tplayed with the abserr command, but no result..

I have looked up some info about non-dimensionalizing. It could be worth the effort, but it only looks like a lot of work. Is there a way Maple can do this for me, or is it just a lot of hand solving?

I tplayed with the abserr command, but no result..

Thank you for your reply. For a plot from 0..1 they look quite smooth yes, but when I look at the total picture from 0..150 it looks like inaccuracies. 

What I would expect to be the result is smooth plots from 0..150. Especially the 2nd derivative, if you look at the whole picture, the highest (absolute) value is -0.04, at x=0. This is definately incorrect for my application. This means that a beam being clamped at a point also has an acceleration in the same point. Also this messes up all my further results.

I have a few options for what explains the error:

• The formulae have a inconsistency at x=0
• I don't apply the formulae in the right way -> I use the wrong commands in Maple
• Maple has inconsistencies in the solving.

Thank you for your reply. For a plot from 0..1 they look quite smooth yes, but when I look at the total picture from 0..150 it looks like inaccuracies. 

What I would expect to be the result is smooth plots from 0..150. Especially the 2nd derivative, if you look at the whole picture, the highest (absolute) value is -0.04, at x=0. This is definately incorrect for my application. This means that a beam being clamped at a point also has an acceleration in the same point. Also this messes up all my further results.

I have a few options for what explains the error:

• The formulae have a inconsistency at x=0
• I don't apply the formulae in the right way -> I use the wrong commands in Maple
• Maple has inconsistencies in the solving.

Small problem with the images, this time the correct links:

Urn:

d/dx Urn:

d^2/dx^2 Urn:

And the Maple file:

Simplysupportbeam.mw

Very clear explanation. Thank you very much!

Very clear explanation. Thank you very much!

That workes, thank you very much!

That workes, thank you very much!

@Carl Love I understand, it would be better if I would be more consistent in these things.

Thank you very much for your help.

@Carl Love I understand, it would be better if I would be more consistent in these things.

Thank you very much for your help.

I just build a 'simple' worksheet to post here to show my problem, and there it worked. So I went back to my original file and did exact the same and now it workes!

What you posted above does the trick. I don't really understand, because I tried that before, but then instead of 

unapply(sigvm(theta,x), [x,theta])

I did:

unapply(sigvm(theta,x), [theta,x])



I don't really understand the difference, but the important thing is that this workes.

A picture of the result which I was trying to get:

Thank you for your help, it is very much appreciated!

I just build a 'simple' worksheet to post here to show my problem, and there it worked. So I went back to my original file and did exact the same and now it workes!

What you posted above does the trick. I don't really understand, because I tried that before, but then instead of 

unapply(sigvm(theta,x), [x,theta])

I did:

unapply(sigvm(theta,x), [theta,x])



I don't really understand the difference, but the important thing is that this workes.

A picture of the result which I was trying to get:

Thank you for your help, it is very much appreciated!

@Carl Love You are right, it is a procedure (actually a combination of multiple procedures). For (x,theta) it returns a single value. 

I have tested the color on an undeformed pile and then it works, but it doesn't work on the deformed pile. I will post a worksheet in a few minutes to make myself clear. 

@Carl Love I'm sorry, the worksheet is really big and I think that will be even be more unclear. You are right with your transformations, only I took x=x. 

In my case I have deflection in 3 direction, so I can add them up in the cartesian coordinates, eg. xcart= x+ux, with ux the deflection. 

The plot of the deflection works fine, but I cannot plot the stress distribution onto the deflected pile.

The problem is that when I convert the stress to (undeformed) cartesian coordinates, it wants to plot the distribution on an undeformed pile. When I want to convert to deformed cartesian, the variables don't work anymore.

What I would like to do is use the 'normal' variables (x,theta) for the sigvm and project the output on the deformed pile.

I notice it is a quite complicated story and hard to explain. Tomorrow I will post a 'simplified' worksheet which explains the problem better.