Deflection and moment diagrams using AutoCAD

[cve60069]

Dear all

Am I able to calculate the final deflection of a beam from the bending-moment diagram graphically using AutoCAD, please?

 

[rb1957]

yes, but you would you want to ? it'd be much easier using integration ... deflection = 2nd integral (M/EI) dx

 

[MiketheEngineer]

Didn't know Auto Cad would do engineering??

 

[rb1957]

i'm sure there are olde school methods that are graphically based, that you could determine areas and such from AutoCad. just seems to me to be possibly the most inappropriate tool for doing it. i'd use a pencil and paper if i didn't have a s/sheet or a calculator ... i'd use a slide rule (if i could).

 

[paddingtongreen]

There is a graphic method for shear and bending moment, but I don't remember one for slope and deflection.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[paddingtongreen]

I changed my memory, there may be a method using a funicular polygon. I don't know if it can be found, it isn't in any of the books that I retained.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[Once20036]

RB1957 - it's possible?
Can you provide some details?

To answer your second question, no, i wouldn't want to. But I`m curious about the procedure just to learn more about CAD.

I recently learned it can calculate properties of a cross section (Ixx, Iyy, rx, ry, Sx, Sy, y-bar, etc) and that's saved me a huge amount of time

 

[paddingtongreen]

A graphic method of calculating deflection will not save you work.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[IDS]

Doing engineering calculations in a CAD package seems to me like doing engineering drawings in a spreadsheet. It can be done, but it's not the most efficient way to do it.

But that said, you could generate a slope diagram by finding the area under your moment diagram, then generate a deflection diagram by finding the area under your slope diagram, then rotate the deflection diagram so the deflection at the supports was zero.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[BAretired]

If your CAD program can determine areas and area moments, then it should be no problem to find slopes and deflections at any point using Conjugate Beam principles.

The conjugate beam is an imaginary beam loaded with the M/EI values of the real beam. The shear at any point in the conjugate beam is equal to the slope of the real beam at the same point. The bending moment at any point in the conjugate beam is equal to the deflection at the same point in the real beam.

If you are not familiar with the conjugate beam method, Google "conjugate beam method". It really is a very useful way to find deflections, particularly for simple span beams.

BA

 

[ishvaaag]

One could do it by programming to interface with Autocad, be it programs in autolisp or by producing ARX modules.

In 2010 I think Autocad introduced the parametric design feature that allows to make curious things within the program. I made about 50 dwgs to explore the tool, mainly some math (finding roots, euclidean geometry, and graphic statics, such push on walls, funicular determination of moments etc). Unfortunately I don't find as of now where I have the examples if I have managed to preserve them, so I can't post any as of now. But through funicular determination of the moments certainly it can be done as BAretired says by making use of the parametric tool inside Autocad.

 

[rb1957]

AutoCad is an excellent tool for calculating section properties. it's not the tool i'd use to caluclate deflections, but it looks like the method you want is "conjugate beam" described above, or (probably) wiki, or olde text books.

 

[cve60069]

Dear all

The problem with integration and mohrs area theorems, the point of zero-slope has to be determined for the theorems to work and this is not always possible to calculate. Plotting any bending moment using AutoCAD is easy and accurate and I do not mind spending the time plotting the slope and deflection diagrams providing I get the right result.

I suppose I had better read-up on how slope diagrams are drawn and conjugate beams.

Thanks

 

[PMR06]

Unless your moment diagrams are always linear, how do you accurately draw a curve from an equation in AutoCAD?

 

[rb1957]

if the immediate issue is determining the zero slope point, does the method allow you to iterate on an initial guess ?

 

[cve60069]

To write an equation for the bending-moment diagram is very easy, and AutoLisp will plot the curve, its the subsequent iterations and the two unknowns Ax + B that are the problem. With a simply supported beam, the defection could be assumed to be a maximum at mid-span, whatever the load combination, and any error would be minor; but with a 3-span indeterminant beam say, the point of zero-slope may not be as apparent.

I have been reading Gere and Timoshenko and I attach one of their diagrams. I suppose I am trying to reverse the procedure shown in the figure.

Regards

 

[PMR06]

Is the drawing of the moment curve done in discrete line segments or a truly smooth curve? I would think it would have to be a pline of x and y coordinates at very small increments of x.

How are you calculating the moments of your multi-span beam? Moment distribution?

If you can program in AutoLISP to draw a line from a function, why can't you write a subroutine that internally handles the integration of slope and deflection then just graphs them?

I find this problem fascinating from a purely academic perspective. In reality, like others have said, I would of course rather use Excel, RISA, paper.