pencil & paper vs. software for calcs 2

[bjb]

I'm wondering, how many structural engineers mainly rely on software like mathcad for their calcs, versus how many mainly rely on a calculator with pencil and paper?

 

[MotorCity]

I use software for preliminary sizing and verifying final design. I use pencil and paper for design, regardless of project size.

 

[mrMikee]

bjb,

An interesting question. I brought up a similar topic in thread507-126479

I do most of my calcs with software now. Some of the reason for this is that I work on standard products where my work is routine enough that I can automate some of it. I do hand calcs for something unusual or out of the ordinary. Also do some hand calcs when doing a quick check of my output or when taking data from one program to another.

I mostly use CalculationCenter which is a lite version of Mathematica. It's a handy program that certainly is capable of just about anything a structural engineer will need. A disadvantage is that it's output is relatively unattractive especially when compared to Mathcad. I currently use it for ASD code checking of FEA output.

Regards,
-Mike

 

[SacreBleu]

Overall, we use software for about 75% of the calcs. Personally, I use software probably at least 95% of the time, especially Excel. ( I am an "old" guy but am kind of burned out by punching calculator and writing numbers all day). We don't use Mathcad at all.

 

[bjb]

Most of my calc's are still by hand. We do alot of alterations to existing structures and additions, and run into stuff that can be unusual. I have some spreadsheets that I made for things like shearwall analysis for rigid and flexible diaphragms, and I also have some things programmed into my HP, like alowable stress for wood columns and unbraced wood beams. I have checked out enercalc, but so far the powers that be are not into it.

 

[mrMikee]

I had considered buying Mathcad several times over the years, but each time I got interested it had drastically increased in price (currently it is $1200). Finally I bought CalculationCenter because it was relatively inexpensive at $295. It to has gone up in price however, currently at $595. I like it but it's not pretty. Mathcad seems to be the standard for engineering.

-Mike

 

[jike]

I am still patiently waiting for the day when I can have all computer generated calcs on a project (I will probably be retired before that day comes). We have over 100+ spreadsheets that we use for specific tasks in addition to RAM, Visual Analysis, Enercalc, etc. No Mathcad.

 

[bjb]

Jike, how do you like Enercalc?

 

[SacreBleu]

Unlike large steel or concrete buildings, wood-framed buildings are very tedious to calculate without software, and Enercalc doesn't really provide a total system approach. (Woodworks is somwhat better in that regard). Therefore we have developed an Excel workbook that greatly automates the procedure for wood-framed buildings, especially multi-family residential.

 

[archeng59]

I use both pencil & paper and Enercalc for component design. Primarily, I use Enercalc for steel and concrete member design, retaining wall design and spread footing design. I don't like to use the masonry design modules in Enercalc, but my business partner uses those modules. Use some of the Enercalc analysis modules occasionally. I use pencil and paper for masonry, wood, most cold-formed framing and shear wall design. I use LGBeamer for some cold-formed metal framing design. I use STAAD for frame analysis. I use D-Coder for wind load calculations. we have a copy of Woodworks that has wood framing and shear wall design that my partner says works well, but I have not used it.

 

[SlideRuleEra]

I use paper and pencil for most situations. Sometimes compose Excel spreadsheets to do repetitive operations. Use basic software for steel or wood beam continuous span calcs for shear, moment, & deflection. However, I no longer do very many design projects, and they are normally one-of-a-kind.

http://www.SlideRuleEra.net

 

[whyun]

I use pencil and paper for graphically presenting my thought process that may be lost few years down the road unless it is on paper. Typically include sketches, load directions on a connection, etc.

In lieu of using a calculator, I find it easier to make a mini-spreadsheet tool for complex formulas and many variables. If it is worth saving, I save it first and develop it over time into my official spreadsheet.

There are many software available to perform a particular task, but there are limitations on how the results are presented. I always add a written design criteria before any series of computer printouts stating the purpose and conclusions from the analysis.

MathCad and Excel spreadsheets developed for specific calculation task definitely reduce design time, but I always make my engineers perform a hand calc version at least ONCE before "trusting" and reusing the spreadsheet/MathCAD sheet.

 

[JedClampett]

Even though I wasn't asked, I'll give you my opinion of Enercalc:
It seems as if it was developed by software people with only a passing familiarity of design. For instance, for concrete beam design, it asks you for the distance from the top of beam to the center line of the reinforcing. There's no reason that with the bar size and the cover that this couldn't be calculated. I've programmed it into my HP. This seems like a minor quibble until you've had to do it a couple of hundred times.
Or take the case of updates. Their website looks like it hasn't been updated in years. They've been promising a version 6.0 for about two years now. This doesn't give me a lot of confidence in their business.
They seem to be the only game in town regarding individual element design and they act like it.

 

[Dinosaur]

I find the original question perplexing. In my experience, Mathcad is an elegant spreadsheet environment that merely performs the arithmetic for you. Since the user is entering all the equations, I don’t consider this a situation where the software is doing all the design. As you may have guessed, I am very pleased with Mathcad and use it often. My only regret is that it does not easily permit me to place diagrams into the spreadsheets. I know that I can scan a sketch into the computer and then place it into the file, but that is rather time consuming and does not lend itself to easy modification.

I do fear engineers that rely on software too heavily. A recent experience I had when I was asked to review another engineer’s work for my client is a good example. He had a major software vendor’s product that promised to design a cantilever retaining wall. There was a bug in the program, and the solution to the 20+ ft high retaining wall adjacent to a RR track was to supply minimum steel (in this case #7 @ 12 inches) rather than #9 @ 6”, which he detailed after I brought the problem to his attention. In my opinion, he should have taken out a pencil and checked it himself when he saw it on the drawing because #7 @ 12” just looks too light. But either because he was in a terrible hurry, was out of budget or possibly because he didn’t know any better, he did not do this. Fortunately, I caught the mistake, but his client suffered some delay and don’t know that he will work for that client again.

Good question, but I don’t think Mathcad, Excel or any other spreadsheet is the real problem.

 

[bjb]

Jed, I have the equation for the direct calculation for area of steel from the CRSI entered into my HP. It's a great time saver. In my calcs, I do the calculation for Moment strength to back it up, and also so it doen't seem like I pulled As out of the air (or some other place).

Thanks to all for the replies. Personally, I think that a combination of hand calcs and software is right for me. I think that I am biased more towards the hand calc end of the spectrum though. If I had more projects that involved cranking out lots of memebers, I would adjust for that and use more software. I only trust the software or my use of it if I can also do it by hand. For documentation purposes in my calcs, I include all relevant design criteria, assumptions, and descriptions of what I'm trying to do so that if I have to go back to these calcs later, it will hopefully make sense.

One thing that scares me though, is people who use all this software without a good grasp of the priciples, who actually are more like computer operators than engineers. I think that at some point in the future there will be a significant collapse that will be traced back to that kind of person using software with the garbage-in garbage-out syndrome in play. This last comment is not intended to be an attack on qualified engineers who choose to use software for most or all of their calcs. I have no problem with that at all.

 

[archeng59]

Jed, according to the Enercalc website, Enercalc was developed and is maintained by a structural engineer.

 

[jike]

bjb:

I use Enercalc for steel beams with partial drifted loads, steel cols with uniaxial or biaxial bending, base plates, combined footing, etc. I haven't been totally pleased with their service or frequency of their updates but I continue to use it until a better option comes along. I like the ability to store all the calcs in one project file.

I have thought about replacing these with Excel spreadsheets when I get some more time.

 

[archeng59]

I have two engineer interns who rely too much on software only. they are frustrated with me demanding they check things by hand, not to trust the software answers. one told me, "but the computer is always correct." my response, "how do you know?", didn't please him and I asked for hand calcs to prove his assumption the computer is always correct. He made some input errors and had an incorrect solution using the software. His hand calcs showed the correct solution. the same intern developed some a spreadsheet for steel beam design that turned out to be incorrect, also, because he didn't use hand calcs to verify his Excel equations. Seems like the young folks think that because they get an answer, not an error, when using software, the answer must be correct.

 

[whyun]

For designing simply supported steel beams or girders, I request my engineers to supply hand drawn load diagrams which shows the breakdown of dead/live loads, derivation of loads such as tributary area/width and surface load, identify the NAME of any unusual point load, etc. Only then I allow them to use a software such as RAMSBeam, Enercalc and so on.

No software I am aware of provide a comprehensive breakdown and identification of loads that can be understood with ease after six months (or by someone other than the one who performed the computer design).

Computer is available now as a tool for obtaining quick solutions but it is not yet read for prime time to fully replace hand calcs for many applications.

"Computer is always correct" argument is a scary one and all too ubiquitous among the new generation of engineers. It is imperative that the experienced engineers train and guide the young engineers to avoid catastrophe.

 

[Ussuri]

Dinosaur

I use Microsoft Visio for putting sketches into Mathcad. Visio is a business sketching package. It is quite intuitive and all you do is copy and paste to get them into mathcad. It wont do drawings as accurately as a bona fide draughting package, but if its just a sketch representation I think its pretty good.