Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Working on a thing 5

Status
Not open for further replies.

Celt83

Structural
Sep 4, 2007
1,996
Some of you may have noticed by signature update. I am working on a passion project in the very little spare time that I have with the goal to keep this thing free and open for as long as I can.

I held off on making this post because there is one tool I wanted to have prior to letting the cat out of the bag so to speak, a continuous beam analysis tool.

I am at a point where the tool is in a useable state but still needs a lot of polish, and was wondering if some folks here wouldn't mind testing it out. The tool can be accessed from here, there is no direct link from just navigating the site since this thing isn't complete yet: [link thestructuraltoolbox.com/analysis/simplebeam]Link[/url]

Edit: if you end up getting an error hit your browsers back button.
I've performed a lot of validation checks against the steel manual and some other benchmarks I have and have been getting near exact agreements. I'd ask if you manage to break it which is very likely please drop a bug on the github tracker and I'll get it addressed.

Many many thanks to several of the regulars here for helping get this thing to this point either directly or indirectly.

Hoping this doesn't violate the Eng-Tips terms but if so please red flag it.



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

Recommended for you

Nice job. Thank you for going to all of this effort.

For most spreadsheets that I make for my own use, I like to include the reference material in screen shots. I saw your interpolation routine, and thought I would mention that it is nice to include the formulas, within reason, and in this case the graphical entry "matrix"

double_interpolation_x9x0qt.jpg
 
Thanks for the feedback dvd, I agree getting references in place is a good idea. I plan to clean some things up and add in the pretty math formulas similar to what shows towards the bottom of the concrete corbel tool.

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
Looks interesting. I'd like to play with it over the weekend.

BA
 
Thank you BAretired, the beautiful thing with this setup is You will always be accessing the latest version of the tool so you may see more detailed output as the weekend progresses and I add more features.

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

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik
 
Cool stuff Celt.

I'm sure you are on this but I would love some graphical feedback on the loads that are input. I will review more when I have time, looks like you have this one up on your github.
Side note, also get a lock on the website so it doesn't have the scary "not secure" note in the top left of the browser.


S&T -
 
Oh yeah I’m working on the graphical load feedback as we speak. That part is done in javascript which I am in learn by doing mode.

Also working on getting a security certificate for the https thing.

Edit: the plan is to make the code fully available aka the “enter the black box” link bottom middle of the page will take you to the github page containing all of the current source code

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
alrighty
The site now has a security certificate so the little padlock icon should be normal now
Implemented charting for the distributed load types
Caught a divide by 0 error for distributed loads that had equal end values with opposite sign

General Results Tab will show a very verbose log of the calculation process
ULS Results shows the ultimate Moment and Shear Envelope plots
SLS Results shows the service Deflection, Shear, Moment, and Slope envelope plots (in that order)
Basic Results will be empty for now

My solution method for the beam solves for the roots of the shear and slope equations so points of absolute max/min for Moments and Deflections are captured in the plots.

Capture_chovhh.png

Capture_o21lft.png


I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
Celt - I checked that out when it popped up in your signature a while back, and I've been excited to see progress. Thank you for putting it together. I'll put it to use next week and let you know what happens.
 
thanks phamENG!

FYI it charts point load and point moments now :), now I can focus on what result data to bring forward.

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
This is a helpful tool to complete quick calcs. I spend far too much time at the start of delegated projects figuring out dead and live loads so I can input them into different software. I have always wanted a solution like we did in Univ. Nice concise plots on a single sheet showing both shear and moments. I have done them in excel, but all of us engineers like nice pictures. This is great.

I noticed your tagline update a while ago. I do not think it quite gives yourself credit for the amount of effort. I have seen many of your posts with IDS discussing the coding reality and can only imagine the time involved. I am trying to learn python when I have time and it is not easy to find time. Some of your ideas remind me of the course material I am taking thru Degree Tutors Degree Tutors Youtube

If you have time you could add the load cases for Canada? The BCBC is available publicly, so I pasted the chart below.
bcbc-4.1.3.2_lbwgyd.png
 
Hi Celt83, Nice job! I'm also working on a similar thing - [link calcs.app]calcs.app[/url]. The underlying tech is different (React app, calcs are done on the front end in javascript), but feel free to reach out if you ever want to brainstorm or need help. On github I'm @jchatkinson.

edit - i should say that this tool is still heavily under devleopment and everything should be considered alpha stage - may be broken, not yet extensively verified.
 
Thanks Brad805:
"I have always wanted a solution like we did in Univ." this is my intent the overarching philosophy I'm trying to adhere to is to present a set of concise results along with the option to see more detailed breakdowns where possible and to format output close to a hand calc process.

I should be able to knock out the option for those load combinations this weekend, as a heads up I do plan to implement a toggle switch to go between imperial and metric units at some undetermined time in the future as that will take a bit more effort so while the combos will get added it will still be limited to imperial units.

Thanks Sonofatkins:
I stumbled on your calcs.app site recently great stuff there! I have a few items done purely in javascript namely the rebar development length, corbel, and interpolation tools.

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
@Brad805:

haven't got the load combos you asked for yet, but as a quick compromise I added in the option for user defined ultimate load combinations and a toggle to only consider the user combos:
Capture_id2z5s.png


The load charting now also labels the loadings for easier confirmation that they are applied:
Capture_g2atal.png


The detailed results for the main span for ULS, SLS, and Basic load combinations are done now as well:
Capture_nzyw3i.png


Combination results are in "drawers":
Capture_upesyv.png


The "drawers" expand to show each of the load pattern results:
Capture_dslsi4.png



I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
@Brad805:
The BCBC 2018 Non-Crane combos are now an option. F load types take the place of Live loads due to fluid in tanks from sentence 6 and Lr load types represent storage type loads per sentence 7.

Would love a back check on the combos which are built here (starts on line 345): Link

By default the webtool will do 0.9D factor for all load cases, plan to give you an option to say this isn't applicable to load combos including gravity load only.

An option for the inputs to be in Metric is in the works also.

How have you been liking the DegreeTutors content, some of the videos they have on youtube peak my interest?

I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
I will take a look on the weekend. I noticed your update. 99% of the time I am only interested in the D/L/S load cases for simple beam or joist calcs. US Imperial would be fine for my needs, but others might like metric.

I quite like the Degree Tutors site, and think it is very good value for the effort they put into the content. Lately I am not finding much time for the courses. There is not a great amount of focus on Python since Dr. Carrol believes in learning the coding you need as you go. He seems very engaged in the site, so I would think you would have good luck sending questions about some of the web programming things I expect you are working with. He is developing tools for members to use online based on his courses.
 
@Celt83:

One of the things that I find particularly interesting about your "thing" is the approach that you've taken to transparency. The universal software complaint in our world seems to be "too black boxy". The usual approach to addressing that is to create some great reporting. The trouble with great reporting, however, is that it represents a massive outlay of development work and, at the end of the day, it seems difficult to arrive at a solution that's satisfactory to a wide swath of people anyhow.

In a way, you've provided an extreme version of transparency in that you give out the actual code that drives your tools. So, in theory, a user can "look inside the box" to as great an extent as they choose to. At the same time, despite having that option, I suspect that 95% of your possible users won't ever look inside the box. And maybe that's okay so long as they have the option.

Have you had in feedback from your beta testers on that aspect of things?

 
Not much feedback so far. Have a coworker who has been using a few of the tools as part of their instruction at a local college. Have had some minor feedback from the same coworker on some presentation items and adding a possible sorting option to the steel database.

From the analytics that I have access to either:
- The tools that are available are incredibly efficient giving folks the answer they want in a timely manner.
- The tools that are available do not offer what folks are looking for so they get viewed but not used.
- A good number of visitors never went beyond the home page, so perhaps it's not obvious how to access the tools.

On the "Black Box" side your 95% is low, based on some limited analytics about 98% of folks aren't going beyond the Toolbox site. Of the 2% that did go to Github about 98% of them did not actually look at any of the code (perhaps a bit overwhelming looking at a series of folders).

Have a real lack of outside time currently so development is a bit stalled but I have a few additional pieces started. Long term I'd like to add an additional form of information for each tool which puts into blog or white paper form the methods and derivations of the methods used by the tools.



I'm making a thing: (It's no Kootware and it will probably break but it's alive!)
 
I'm hoping to put some time into this and a few other items next month when my schedule clears. I've poked around a little and been impressed. Having a web-based section properties calculator is definitely on my list of missing resources, I don't always want to pull it up in Python.

For the moment, I've been held back because my brain doesn't work well in imperial units any more. Any estimate on incorporating a metric toggle?

(Or I probably just need to get back in imperial mode before I return to NA)

----
just call me Lo.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor