Is automating structural drafting possible/worth the effort

[labeattie]

Lately at work, I have found drafting to be a bit tedious, which is no big deal. But as I have enjoying learning to automate things (like learning excel macros), I was wondering if anyone has any experience in using autocad scripts and macros for structural drafting. I was curious if this might be worth learning in free time, or if I wouldn't get much utility from it. A potential issue is that my company uses autocad whereas most other structural firms I know use microstation, so if the learning process for these automation procedures is very difficult and very different between the two programs, the knowledge could be obviated upon changing jobs.

Thanks guys,
labeattie

 

[labeattie]

(and gals)

 

[EngineeringEric]

I don't know how/what I would automate anything in CAD.... I do however have some dynamic blocks and files full of similar details that i borrow from for starting points. In example, i have a file called Concrete Slab, which has things like Sawn Joints, multiple stair landings, and so forth; and other files for steel framing, bar joists, etc.

 

[cvg]

scripts and macros are what made Autocad so successful. they are easy and can certainly increase production.

if your peers are using microstation, then are you doing bridge design? Outside of state and county public works, most do not use microstation. Civil 3d is much more likely to be used

 

[manstrom]

To an extent, yes. I wouldn't spend 100 hours learning learning to code in order to write a few simple macros. But I would spend a day if it would save me 30 seconds worth of drafting 200 times and week.

Script files for cleaning up architectural plans, drawings monotonous things on plans, setting up templates, etc are fairly handy. At my previous job, we used layer scripts extensively.

It doesn't need to be complicated. Learn enough to write one script. Tweak an existing one if you have to. Start with the most simple, but time consuming job you have to do.

 

[Brad805]

The question of automation depends largely on how consistent your type of jobs are. If you are constantly working for different architects that come up with new hair brained ideas, then I would say the time to create an automated system would not work out well in the end. If you are always doing routine tasks the time to improve your skills with CAD will prove to be a savings in the end. Take a look at the pre-eng steel industry. Their buildings do not vary significantly, so they use the API tools in modeling software to create systems that reduce the time in every step from estimating, engineering and production.

 

[glass99]

I would be interested to hear about consulting engineers with specific instances of AutoCAD automation being successful. Agreed with Brad805: If I ever do the same thing twice I feel like something is wrong.

 

[hawkaz]

My drafting team loves using scripts in AutoCAD. We had a real pro at these- but unfortunately he left to go work at Argonne labs.
I don't believe they work in Revit.

 

[mtu1972]

This was from a former job and was in another CAD system. I supervised the CAD IT director (Ray) and given my limited CAD skills it made reviewing his work difficult. One year during his review, we got on the subject of utilizing the "computer's power" so as to eliminate mundane tasks. We were in the corporate engineering office of a large paper manufacturer.

Ray asked what was I thinking about and I said that I'd like to use a dialog box to define things like bay spacing's, column line designations, etc. I said that I would like to layer the grids, dim's, etc. so that they could be then used for all of the plans - Floor Plan, Foundations, Framing, etc. In that system, we drew in what would be "paper-space" in AutoCAD. Each sheet was on it's own.

Ray thought he could do that. What else would I want? For a single story warehouse structure, we would input the loads (DL, LL, SL) and the allowable soil bearing pressure and then design/draw the footings (square footings only in this first trial) for those loads and bay spacing's. Within a couple of weeks (in his spare time) he came back with the first effort. It worked pretty good. We tweaked a few items and then had a floor plan and foundation plan for our fictitious building. Wind Loads and braced bays were never in this preliminary effort.

When we presented it to the entire department, there was a lot of skepticism, such as: "Not all of our buildings have constant bay sizes for their entire length or width". Yes that was true, but with a "stretch box" those idiosyncrasies could be easily remedied.

Ray then asked what else was a typically mundane task. Drawing and detailing stairs came to mind. As he was not a structural drafter, we reviewed the last couple of projects to see how the stairs were drawn. Standard details were used to define all of the connection's, etc. The plans and elevations were shown for each stair system.

We reviewed the basics - max. step height (7.5"); one less tread than riser; landing lengths, Code limitations, etc. So with this dialog box, one would put in the elevations of the floors, the number of runs between levels, any limits on individual stair runs, etc. Again, in about 2 weeks, we had a working model that drew the plan and elevation views of the stair in question. It could be repeated for each set of stairs.

I'd like to tell you that it was a resounding success, but I was downsized out later that year when the company was purchased by a foreign competitor. The engineering staff was gutted (70+ to 6). They were forced to switch to AutoCAD and they contracted out virtually all of their design/drafting. Ray still was there, but mainly to convert any existing drawings to AutoCAD or PDF's and to provide them to any of the consulting firms that were going to do the work.

I've related this idea at my subsequent employer's but no one was ever interested. We billed by the hour and there was no incentive to reduce any hours.


gjc

 

[slickdeals]

Is Ray looking for a job?

 

[TehMightyEngineer]

Our company developed a bunch of automated lisp routines in AutoCAD to create 3D shapes we commonly use. We have programs to create all the structural shapes, pipes, valves, hangers, bolts, nuts, flanges, motors, and so on. If you're trying to do 3D drawings and do not have some way to automate this then you're doing it wrong. To have to draw every 3D shape by hand isn't worth it.

All our drawings are done with full models similar to how BIM projects are done, we've been doing it this way for 20 years and are still baffled that people don't do the same. The speed, quality, and ease of use when using full 3D models with automated routines is well worth the time to develop them (or the money to purchase them).

Maine EIT, Civil/Structural.

 

[labeattie]

Thanks everyone for the responses. I may learn a little about it in the future, primarily out of interest, but my application may be limited to drawing setup, as our jobs are not usually very boilerplate, and I never have worked in 3d cad. It's very interesting to hear about how you all have used or not used these capabilities differently. Thanks!

 

[RareBugTX]

While macros and scripts are very useful to run simple commands, AutoCads real strenght lies in AutoLISP, DCL (dialog box language), ARX, C++, and even XDATA where you can assign values to specific AutoCAD entities like lines arcs circles and blocks. There are a bunch of books on the topic that you can google or amazon that you can get for a few bucks. Sounds to me that you like programming as I do. You should give it a shot it is so much fun. [pre][/pre]

 

[labeattie]

Ok great! Yes that sounds like something i would be very interested in checking out. Do you have any particular reference recommendations?

 

[jvvse]

My firm currently has a whole suite of LISP routines that make drafting very fast. I'm looking into improving it as we are branching out into newer areas, but without this automation you're really hurting yourself for time.

 

[3DDave]

There's a demo of an in-use application I saw for a company that makes structural towers - lots of beams, angle, and gusset plates. They automated to the point that the model drove all the beam and angle cut lists, and placed all the gusset plates and generated output for plasma cutters.

One result of attempting to automate, even when it does not succeed, is the examination of the design process that is required as preparation. Automation can't help in areas that are unclear; attempting to automate makes those areas obvious. It has to be a genuine effort to work right, but it's worthwhile.

Simple things like examining notes and specification references to see that the automation will reflect what is implied or required can uncover areas that are either more costly than need be or are assumed correct for wide application when they are only applicable to special cases, and vice versa.