Documentation
About Archicad's documenting tools, views, model filtering, layouts, publishing, etc.

Consensus on accuracy of composite wall components?

Anonymous
Not applicable
I've made a bunch of wall composites for our new v14 template but now I'm thinking about the accuracy of the wall components.
My composites use real-world dimensions, eg 64mm metal studs and 13mm plasterboard linings. Now in the world of AC where we can't round our dimensions (please someone tell me this has changed!) this leads to many, many, what we call 'non-rational' dimensions. These generate comments like, "4026mm? ...the builder'll just laugh at that! He can't build to 26mm!" To which I usually reply "Bite me. I know what the builder can build to, so does he. He'll get as close as he can. I'm NOT making every non-rational dimension use custom text! It's a waste of time!" etc etc, you get the idea. ...and then I end up using custom text all over the place and worrying about updates (I use a different pen colour just to make them obvious).

Anyhow, I'm looking at these composites and now wondering if I should bite the bullet and change them all to 'rational dimensions' - say all to increments of 5mm. Forget the 'real world' component dimensions.

So the question is - In the absence of rounding dimensions, how does everybody else size their wall components?
12 REPLIES 12
Barry Kelly
Moderator
Personally I'd show it as it really is.
Let the builder decide where he wants to round dimensions.

If you manually change the dimensions as you have pointed out you will have to manually maintain them when changes occur.

If you rationalise your composites then you will start getting errors creep into your plan unless you are very careful.
i.e. the first wall will be 3mm to wide. The next wall will be that 3mm plus another 3mm out. The next 9mm, etc.
Before you know it you are 25-50mm out of whack.

As the old saying goes - "model it as it is built."
Then you can't go wrong.

Barry.
One of the forum moderators.
Versions 6.5 to 27
i7-10700 @ 2.9Ghz, 32GB ram, GeForce RTX 2060 (6GB), Windows 10
Lenovo Thinkpad - i7-1270P 2.20 GHz, 32GB RAM, Nvidia T550, Windows 11
Anonymous
Not applicable
Thanks Barry. Good point. I feel less crazy.
50mm can be a world of pain with the BCA or your not-so-friendly developer!
Link
Graphisoft Partner
Graphisoft Partner
We actually make our composite cores nice round numbers. A 64mm metal stud, we'll create at 65mm. A 76mm metal stud, we create at 75mm. We leave the skins at real world dimensions, for our interior department, but otherwise only dimension to the cores.

And all of our buildings are still standing (I hope)!

Cheers,
Link.
Anonymous
Not applicable
Perhaps this is one advantage of imperial measurements. We generally make the composites accurate to 1/8" increments (though I've seen some people put in 1/16" of lead lining). We can then set our dimensions to round to 1/8" or 1/4" (or even whole inches in schematic) as appropriate but generally the contractors here are used to seeing 1/8s on the drawings.

The other thing that helps with rational dimensions is we typically dimension to only one side of the wall (framing or lining depending on the type of project and or standard of practice at the particular firm). This way the thickness is only of consequence in the enlarged fixture plans and such when measuring from face to face of the lining and in those cases we want it noted to 1/8" accuracy.
Link
Graphisoft Partner
Graphisoft Partner
There's an idea: convert from metric back to imperial!

Cheers,
Link.
Erika Epstein
Booster
Link wrote:
There's an idea: convert from metric back to imperial!

Cheers,
Link.
Link, You always know how to bring out a smile
Erika
Architect, Consultant
MacBook Pro Retina, 15-inch Yosemite 2.8 GHz Intel Core i7 16 GB 1600 MHz DDR3
Mac OSX 10.11.1
AC5-18
Onuma System

"Implementing Successful Building Information Modeling"
Anonymous
Not applicable
No argument from me. I think base 12 has advantages over base 10!

I guess an equivalent to Matthew's way would be dimensioning to the center line of walls and leaving the actual wall component dimensions to the wall tag, or to detail drawings. I do wonder what the builder would make of it but it sounds like a way of avoiding many tiny dimensions on a wall set-out drawing + improving clarity.

Link - what happens when you do your RCPs? you still dimension from stud?
Anonymous
Not applicable
Angry wrote:
No argument from me. I think base 12 has advantages over base 10!

I guess an equivalent to Matthew's way would be dimensioning to the center line of walls and leaving the actual wall component dimensions to the wall tag, or to detail drawings. I do wonder what the builder would make of it but it sounds like a way of avoiding many tiny dimensions on a wall set-out drawing + improving clarity.

Link - what happens when you do your RCPs? you still dimension from stud?
I've heard of some doing centerline dimensions but don't recommend it. Builders don't lay out to centerlines. We dimension to one or the other face of stud; usually the dominant one (ie corridors first, rooms second, accessory spaces third).
Anonymous
Not applicable
Link wrote:
There's an idea: convert from metric back to imperial!

Cheers,
Link.
For my own personal work I've converted to nits, tics, and stiks.

The nit is the basic unit, a tic is one seventy-second of a nit, and a stik is twelve nits. The nit can be divided into hits, tits, quits, sits and twits for halves, thirds, quarters, sixths and twelfths of a nit. This makes them 36, 24, 18, 12 and 6 tics each which is quite handy. The non-standard measurement known as a nix, which is 9 tics, is also useful at times.

For smaller scales the whit is a small but undefined fraction of a tic. Thus the phrase "not a whit of difference between them." This is also the equivalent of the PH, CH or RPH commonly used by north american carpenters.

For area measure a square nit is a nat, a square stik is a stak and a square tic is a tac. For larger areas twelve stiks make a flit and a square flit is a flat, which at over 100m2 makes for a comfortable dwelling.

Of course for volumes there are nuts, tucs and stuks.

The areal and volumetric equivalents of the fractional nits are not commonly used and can, in at least one instance, lead to some embarrassment if not carefully pronounced.