Minor calculations as a structural engineer 9

[thomastheman]

Hi,

I'm a structural engineer working in Norway and have been working 6 years in oil & gas and 6 years with civil engineering. I've worked for different companies and what I see is that the level of hand calculations for minor calculations is quite different in each company. Someone have excel-spreadsheets for environmental loads, Mathcad sheets for the same, whilst others are making the minor calculations from scratch for each project. In oil & gas, the engineering companies are using the same one-span beam software when doing simple steel and column calculations, which also includes a large section library with section properties. These calculations can easily be done in the excel, mathcad or similar, but this particular software is still used across the industry for these particular calculations.

For civil engineering, its common to use a software package for environmental loads, but it also contains opportunity to calculate typical concrete beams, columns, consoles, foundations. The same software package is used across the industry and seems like everyone is doing it the same way.

So my question to the forum is, does this sound familiar with you coming from other countries than Norway?

 

[milkshakelake]

Every company has a different suite of spreadsheets and software for calculations. Even drafting software is different. One company I worked for had a spreadsheet for wind loads, another one did it by hand using approximate methods, and another one uses software.

 

[phamENG]

Here in the States, there's no 'typical' way to do it. Everyone does it the way it suits them best, so long as they meet the standard of care. So whether you do it by hand, home made calc sheets/spreadsheets, or licensed software you need to understand the inputs, the outputs, and be able to validate the results.

 

[DTS419]

This reminds me of a story I once heard of an experiment that was conducted at a large engineering conference. Engineers from several large firms who were present were given a simple design problem, something like designing a support system for a roof top air conditioner. The range of solutions that were submitted varied so significantly, I can't remember to what extent but the moral of the story was that in the engineering field there is a ton of variation in how even simple problems are solved, so imagine the complex ones. In the end I suppose this is an argument for very conservative factors of safety set by the codes.

 

[lexpatrie]

No. It's about having responsible design professionals with adequate technical background, passing licensure exams after a period of out of school practice and mentoring, who practice in their area of demonstrated competence, and who adhere to the ethical requirements and design standards and standard of care (that mercurial little pixie.

Factors of safety are NOT a substitute for competent engineers following design standards and the generally accepted principles of mechanics.

 

[JoshPlumSE]

Pham -

I'm with phamENG. There is no "right" or "typical" way of doing a problem. I used to use MathCAD spreadsheets a lot. I liked this because it allowed me to have a very nice looking organized calculation that I could put into a PDF file easily.

Then I worked for another company that had set up a lot of Excel Spreadsheets to do the same type of things that I had done in MathCAD. Neither was inherently better. You still have to know what you're doing and check your work (and the program) to make sure it's correct.

Note: For things that would not be good to do by Excel, we'd do hand calculations and scan them. This honestly worked quite well. We had a PDF editing software that made it easy to combine multiple PDFs together with a common header and footer to make it into a nice calc package.

 

[JStephen]

My impression: Some fields of engineering seem to be more cook-bookish than others so it kind of depends on what you're doing.
A quote comes to mind: "All values essential to the aerodynamic design analysis of towers and stacks...particularly the damping data, are subjects of wide differences of opinion among various investigators..."

 

[human909]

It varies alot. Older engineer often do hand calcs or excel spreadsheets. Younger ones might use more computer software. There is a bunch of good stuff out there that relying on Excel or hand calcs is steadily become the 'old ways'. Of course if you blindly use a software package then you can readily get yourself in trouble.

There is several pieces of software out there now that is no longer black box calculation, you can see the calculations there in black and white if you desire along with the specific code references they relate to. ClearCalcs and SkyCiv are two that spring to mind.

Probably 95% of my formal design calculations are performed using software. I was doing wind by hand for a long while but Checkwind does a decent amount of that for me now. Of course I still do "back of the envelope" calculation daily on scrap paper on my desk or on excel as either brief checks for other purposes in predesign.

 

[EngDM]

I've been trying to make the switch from hand-written calcs to a typed reporting system like SMath or some python repo's.

ClearCalcs and SkyCiv are two that spring to mind.

RISA appears to be implementing this as well, granted anything with a page number reference is not very helpful since they don't directly state what edition of the standard they are referenceing. For instance if a clause remains unchanged, or portion of commentary remains unchanges, the reference is of the book they originally used when creating the check.

 

[thomastheman]

Probably 95% of my formal design calculations are performed using software. I was doing wind by hand for a long while but Checkwind does a decent amount of that for me now. Of course I still do "back of the envelope" calculation daily on scrap paper on my desk or on excel as either brief checks for other purposes in predesign.

What about simple one-span beam and column calculations? I've worked with people that model everything in a FEM software, even minor access platforms. I would look at the structure and if I can't see a benefit of using a FEM software, I would do it by hand calculations. Usually, the hand calculations can be supported by software that can easily calculate beam capacity, both in strength and stability, giving reaction forces that can be implemented in the hand calculations. Do you use a similar software or do you select beams/columns from for example a table where all sizes have been calculated with their respective lengths and support conditions?

 

[wth]

I practice in Norway (buildings), and use a mix of hand calculations, excel, smath, "simple" software, python and FEA, depending on the task. And having worked for one of the largest companies it varies from team to team even within the companies how engineers solve their tasks.

For your simple beam example, I typically use a software like what you described (Colbeam EC3) and not hand calculations (except for loads). Mostly for efficiency, using software I can design it in a couple of minutes, try different section shapes and sized, and print a report for documentation with all checks required in EC3. And should something change in the input, I can just edit the input.

 

[human909]

What about simple one-span beam and column calculations? I've worked with people that model everything in a FEM software, even minor access platforms. I would look at the structure and if I can't see a benefit of using a FEM software, I would do it by hand calculations.

If it is super simple I wouldn't even calculate it. Minor access platforms are even more about simplicity of design than large strutctures. But I wouldn't calculate even a simple one span beam or column by hand because my muscle memory is faster doing it in under a minute in software.

I'll bring in a hand calculation for a simple span on depth calculation. If the span on depth is <25 for a lightly loaded platform then I'll 'approve' or 'specify' the member and move on.

or do you select beams/columns from for example a table where all sizes have been calculated with their respective lengths and support conditions?

The only time I use load tables is for quick determination of capacity of compression members based on effective length. Normally for pre-design or for a very simple design. I never use tables for beams.

 

[JLNJ]

I had a mentor who told me that calculations should be at least 50% words. The calc package should tell the story so someone can follow it. No one cares about 500 pages of output from 200 load combinations.

 

[271828]

My experience is about like most of the folks' above.

For framed structures, even small ones like little platforms, there's always the question of whether to build a model and perform all checks in it, or design components one-by-one with manual calcs, etc.

A lot of this decision comes down to how often you're using the modeling program. In my experience, IF I'm using the program enough to stay fluent, it's usually better to use it for all but the smallest and simplest designs.

Mostly, this is about changes and investigating other options. It's a lot faster and more organized to modify the model and run it again than the alternative -- revise the key plan, re-generate calcs for several individual members, and book-keep that process.

If I'm not using the program enough to stay fluent at it, then I'll use a combo of spreadsheets, Mathcad, commercial individual component design programs, etc.

 

[Tomfh]

Factors of safety are NOT a substitute for competent engineers following design standards and the generally accepted principles of mechanics.

Not in theory, but in practice safety factors catch many a design error.

 

[271828]

Not in theory, but in practice safety factors catch many a design error.

Just by sheer luck! LOL

 

[RPMG]

I prefer to show people hand calcs over excel or FEM because the best engineering calcs/reports that I've seen have all been primarily by hand, and I agree with JLNJ. Also, I would say that hand calcs are not an "old people" thing. I think you're far more likely to receive a hand calc from an experienced young person, who wants to demonstrate their competence and cares about the presentation of their calcs. With that said, I typically use excel and FEM, not a pocket calculator, and I don't reiterate tasks by hand.

 

[271828]

I had a mentor who told me that calculations should be at least 50% words. The calc package should tell the story so someone can follow it. No one cares about 500 pages of output from 200 load combinations.

That is awesome.

For unique calcs, it seems like an argument for Mathcad instead of manual calcs. I type a lot faster and a lot more comfortably than I write. I had dropped MC for about a decade, but picked it back up late last year and it has helped a lot. The downside is that it's harder to sketch in MC, but I can drop in screenshots from Bluebeam pretty quickly.

 

[lexpatrie]

I can't decide if this is a tangent I should just let fade into the distance or this is germane to the conversation. For future generations, I'll comment. I guess I got 3 gold stars for the previous comment so maybe I should expand....

In the interest of clarity, my reply above was responding to the comment from

In the end I suppose this is an argument for very conservative factors of safety set by the codes.

Safety factors are simply NOT there for design errors by the engineers, they are there for other things, variations in the materials, loading, and construction tolerances. I'm not sure I've seen any comprehensive listing of what exactly the phi factors are accounting for, but back when I was going through Reliability with Ted Galambos, the monte-carlo simulations we did all involved various coefficients of variation of the sections, yield stresses, etc, so even something like tolerance on rebar placement wasn't in that calculation (which, admittedly, was for structural steel, I mean it's Ted Galambos teaching it). I'd expect some measure of tolerance on rebar placement in a flexural member to be in that sort of reliability calculation behind the phi factor, but I can't confirm or deny it's existence as I just don't know the background on concrete design that deeply. They rolled out their LRFD before I was born.

Not in theory, but in practice safety factors catch many a design error.

Potentially, but I'm not aware of any situation where that's been identified as a factor, so I'd be curious to know if you have one specific circumstance of that actually happening. Beyond a minor overload, or designing say, a corridor for 40 psf and it's required for 100 psf (and never sees the design load because there's never been a sit-in protest in the hallway)... I do have one 100+ year old wood structure that doesn't calculate out to current standards...... but under no circumstance would I propose that was an "engineered" design. I don't think the NDS was even around then, let alone a building code.

All the (admittedly rather dramatic) failures I'm familiar with have design errors and fabrication errors and many of them have supervision issues, bad formwork practices, poor shoring or reshoring, threshold building inspector that never went to the site, etc. (I'm specifically thinking of Berkman Plaza 2, collapsing during construction in 2008, and of course, my personal "go-to" Harbour Cay, collapsing during construction in 1981, and Station Square had a whole raft of issues including major construction deviations that might have been enough to collapse the structure "on paper", even if it had been properly designed (which it most decidedly was NOT).

[ I'm speaking off recollection, here, regarding Berkman, I may have an item or two not quite remembered right], the Hard Rock in Louisiana seemed to share a lot of the same symptoms (at least, on the engineering competency side of things, not so much the contractor on that one).

Now, to reengage with the post directly above, 50% words in the calculation package is perhaps ... ah never mind, that's not going to make sense as a comment. Here's what I have to say, there should be sufficient words in the calculations to explain the "strange stuff" that another engineer won't be able to follow. I don't need a standard calculation that babbles on endlessly about the stress block origin and articles that reference the development going back 30 years (not that anyone besides me might be even tempted), but the basic things like bd2/6 don't need explanation, it's when the depth is wrong, that's when I need an explanation, so "rough sawn" would be helpful, the actual species "picked" in your various retroactive after the fact calculations on existing construction, sources for the allowable stresses, ... what I call "connective tissue" for the calculations, little explanatory notes.

As a side note, one of my classmates, who is now fairly "up there" in ASCE 7, used to write "and then the magic happens" when we were doing dynamics, (this class most of the assigned homeworks had solutions (answer only) in the back of the book. Fat lot of good the final answer did because half the points were for the "connective tissue" between - the free body diagram, the equations of motions, the known values, the answer, in a box, was worth 1 point out of ten. How I hated that class when we were in it, yet I invoke the mentality so often now. "How did this calculation come about"? Is a question I so frequently ask as I struggle through some engineer's forest of numbers that's cleverly camouflaged itself as a calculation package. Usually this is where the problem lies, a 5,000 lb point load somehow gets spread out into five or six studs and then the studs "work". The calculation skipped over the part where the point load gets distributed out. What's the mentality? It happens by "magic"? It's presumed to distribute through the sheathing? Except there's no sheathing, just metal panels and horizontal girts at 32" o.c., and the metal panel is already "busy" being the shearwall. The (single effective) top plate transfers the load? How is this accomplished? And it's one little line on a "standard" calculation sheet, i.e. given this is in-house software, this has been done on dozens of projects, if not hundreds. And fundamentally there's no load path.

and here's my personal gripe,

Discussion of brace locations on things deemed to be braced. "Braced by metal deck" for your various steel roof beams, and so forth. Done. A simple "Pin-fixed" notation on a column design where the explicit K factor is not disclosed (K=1 would work just fine, thank you, just disclose it), for example.

You have no idea how many times I look at a building and there's some garbage mid-span on a roof beam or girder that's one step up from vermicelli that looks like some numbnuts considered it a braced location for the roof beam in uplift. What perhaps saves these is while the unbraced length is wildly incorrect in the calculation, there's the 0.6 on the dead load (presuming an ASD Design), there's more load, perhaps 0.9D in reality, the design wind isn't there (yet), and the C[sub]b[/sub] for a continuously braced tension flange is 2.0, so there's two mistakes (at least) and they are somewhat offsetting each other. Much as one might be tempted to say "no harm no foul" the errors cancelled each other out, this is all the more troubling because there are two errors being made versus just one by a more thorough engineer, and the errors are errors they are totally unaware of the existence of. NOT GOOD. ((I may have just identified the situation I asked Tomfh to find)....)

A second personal gripe: No explicit consideration of ponding. I don't expect to see a full-on ponding analysis, what I mean here is "roof slope of 1/4" per foot provided by tapered insulation" or "roof slope of 1/4" per foot provided by sloped steel framing. Maybe even "roof designed for 2" of detained water at the drain, plus slope" or just the results: Roof designed for a rain load of 22 psf. "Structural notes on framing plan require roof drainage system, designed by other, to limit water depth to 2"..." something like that. (Very few roofs are free-draining, and a gutter is not "free-draining").

Generally:
A figure showing what the heck the calculation is doing (this you'd be surprised is almost always missing, and it's a major impediment to understanding the calculation during a peer review as well as working with the calculation file long after it's built). I just see a wall of numbers and an "OK".

You really want to provide enough explanatory text so YOU, ten years from know, know what you did. At least that. This is easier to conceptualize if you shelve a calculation package for 10 years then excavate it and you'll see how hard it is. But it takes ten years. (Last I checked New York requires files to be maintained indefinitely, there's no statute of repose), so if somebody complains about it, you can invoke that. I've had people purge my calculations and records in the past. Not that I'd ever have had access to them, but the company shouldn't have purged them. I've actually heard the mentality that detailed calculations can only hurt you. In addition to this being objectively false, (hypothetically an expert witness defending your work will be able to at least see things,), the overall quality of the people who "can't do, but critique" isn't universally fantastic, as the people who hire them (attorneys) aren't deep in the guts enough to know if an engineer they hire is any good. Further, the ones who specialize in these kind of issues are fairly rare, and there's the temptation to ignore the latest developments in the field so they can become out of date rather easily. (This is something I'm feeling quite acutely as I suffer through an ASCE 7-22 project where my "home state" is on ASCE 7-10, and I actually do design when none of the usual design engineers will consider it).

TLR, if you can't read the full post, that's fine, but it's not a "me" effort to summarize it. Sorry. (that's not meant to be snarky, I'm just too tired to summarize).

 

[271828]

I think we need to define "catch a design error." I assume this is referring to something like if 20% of the load was missed and everything else has been done correctly. The computed stress ratio is 1.2, but we all know there's a 1.67 factor of safety in the design. The factor of safety prevents an actual failure, so it "caught the design error" in that sense. However, the structure isn't designed per the Building Code, so it didn't "catch the design error" in that sense.

Design errors are of unknown size. If half the load was forgotten, the factor of safety won't cover that. If the wrong equation was used, it won't cover that. If a decimal is moved one position, it won't even be close to covering that.

I think it's better to utterly flush the idea that the factor of safety covers design errors. Especially when the group includes newer engineers who are trying to make sense out of the design process: load factors, phi factors, safety factors, loads that are for specific mean recurrence intervals, satisfying the standard of care, satisfying the Building Code, etc. etc. etc.