Working on a concrete column design tool

Awesome!

Quick thoughts:
- Option to chamfer the column corners (minimal impact to capacity but matches as built better)
- Chart some the critical points sim. to SpColumn.

something like this output:


- Plot a capacity line from user load point to the capacity curve

Edit: for the biaxial stuff check out the circular section and TLR section spreadsheets here: Link and Also look at IDS's site for their concrete section tool and supporting documents.


I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

took a second to find the link but check out what these folks are doing:
[URL unfurl="true"]https://www.completecolumns.com/[/url]

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

For now you may want to limit the bar layout to symmetric layouts only as I assume you are keeping the neutral axis orientation constant and just varying the depth which means for configurations like the screen shot below the P-M curve is not correct. The section as shown has both M2 and M3 values with a horizontal neutral axis which means the plotted curve is the projection onto the strong axis of the actual P-M slice.



I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

Great suggestions, thanks Celt83. I agree the existing plotting needs some more work. I also like the idea of a detailed view showing the strain, stress, and resultants over the cross section for a selected load. Back to SVG land for me.

For bars in the compression zone, currently I am using their exact stress, but then subtracting their area off of the compression block area. I guess it should result in the same thing.

CompleteColumns looks cool. Lots of good ideas implemented there, like this visualization tool:


-JA
try [link calcs.app]Calcs.app[/url] and let me know what you think

Sonofatkins said:

For bars in the compression zone, currently I am using their exact stress, but then subtracting their area off of the compression block area. I guess it should result in the same thing.

Click to expand...

This works fine but I found once you get into the biaxial stuff it was much easier to modify the bar forces rather then mess with the math of included holes in the compression block.

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

one of my old threads where I figured out a method to handle odd bar layouts: Link
I did a lot of research into this at one point but had to put it on the back burner for awhile so feel free to reach out about anything.

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

Celt83 said:

I assume you are keeping the neutral axis orientation constant and just varying the depth which means for configurations like the screen shot below the P-M curve is not correct.

Click to expand...

Yeah, currently I am not iterating to find the NA orientation. The result matches spColumn for an analysis 'locked' about x-axis. That will need to be worked on to get biaxial bending implemented.

Celt83 said:

one of my old threads where I figured out a method to handle odd bar layouts...

Click to expand...

Nice! I will need to reference that, and may ask some questions once I get into the thick of it. It's a surprisingly challenging problem for such a routine calculation.

-JA
try [link calcs.app]Calcs.app[/url] and let me know what you think

Be ready to dust off your Calculus book

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

Agent666 also has a series of blog posts on the subject: Link

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)

Man Agent666 is the man. I love his stuff and what a read about solving this problem. Fascinating.

Nice stuff so far! I was just thinking to jump into this myself, but given how far ahead you all are I might just focus on cheering on your implementation.

I totally agree with Celt's feedback on plotting capacity lines as the most functional and practical output for practice. Although personally, visually showing that a load condition is inside the appropriate capacity curve is generally sufficient for my needs.

As far as biaxial is concerned, I'm also a big fan of IDS's implementation. In my recent practice (low-medium rise moment frame buildings), the controlling case has been biaxial flexure in the lower part of the curve. As such, I've just generated a bunch of charts from IDS using common column configurations showing the biaxial flexure curves for different levels of minimum reliably present axial force (horizontal slices through the 3D volume).


----
just call me Lo.

I think adding predefined rebars layout is important
since other software omit/has this problem
make the layout [ arrangement of rebars ] in according with ACI code limitations[cover & min bar spacing]
this would reduce time of implementations/repeating [increase efficiency]
Also [if you have the time]
-try adding inside circular/tubular hole [for drain purpose]
-try to make also double skin column with double shear reinforcement stirrups
Best of luck

Celt83 said:

Agent666 also has a series of blog posts on the subject

Click to expand...

Thats very helpful, and also a little terrifying. Thanks Celt and Agent666.

Lo and adn26, thanks for the feedback as well. That's a lot of good ideas for me to think about.



-JA
try [link calcs.app]Calcs.app[/url] and let me know what you think

It is certainly intimidating on the surface, here are what I considered milestones for myself:
[ul]
[li]Relearn Calc 2 and 3 especially vector operations and double integrals [great youtube series: Link][/li]
[li]General Line Integrals for your chosen stress block for a linear edge segment (the problem becomes a closed piecewise curve so you only need to solve the general intergral for a single line segment assuming your shapes will be defined by linear edges) [HINT: Utilize Green's Theorem Link][/li]
[li]Learn a root finding algorithm and apply it to solve for a neutral axis depth such that Sum T = Sum C [HINT: Start with Bisection Method][/li]
[li]Learn a better root finding algorithm and how to apply it to solve for the rotation of the neutral axis (this will not match the angle between the applied Mx and My vectors)[HINT: Newton-Raphson is very fast but can fly off to the moon, Brent's Method is nice but needs to bracket the root][/li]
[li]Find time to Code the above for a general polygon[/li]
[li]Learn about more accurate stress-blocks and repeat cycle, question your understanding of concrete when a parabolic stress block shows you some of your columns are under designed[/li]
[/ul]

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)