|
| Author |
Message |
KinAlberta
Member
|
# Posted: 23 May 2016 22:01 - Edited by: KinAlberta
Reply
I'm a layman just for fun wondering about different ways to solve problems (hopefully not create them. 
So when I think of boxes or I imagine such things as shorter shipping containers, the inherent strength of the sides and entire unibody design prevents any sagging. Lift one corner and the entire 'box' lifts evenly along its length.
Now with a small building on piers using regular sized beams you can't lift a corner and expect the wall to stay straight. (Similarly, you can't cantilever much of the building.) Lift a corner and you can expect sag somewhere along the length of the wall as the wall and underlying beams flex. Huge deep beams under the cabin would be the preferred method to prevent dangerous sag. (Ignoring issues with the centre of the building/box for now.)
Now, since the depth of a beam is critically important towards its strength, why not essentially create deep beams and build the cabin between them? (Standard full height framing with plywood sheathing on both interior and exterior sides of the wall might do the same.) So, on a perimeter wall without doors or low windows could a very deep 'site built' laminated beam be used that extends up above the floor. The floor would basically hang off the beam rather than sit on top of it.
In common cabin designs often the doors are on the front and back, the side walls often have windows placed fairly high off the floor. So could those front-to-back walls essentially be turned into big beams? Say stacking 3 2x12" for a 36" depth.
Eg.
|| wall
||
||
[]
[] beam
[]----------flooring ---
[]
[]
Looks like this might be called a "through girder bridge" in the bridge building world.
|
|
Don_P
Member
|
# Posted: 23 May 2016 22:33 - Edited by: Don_P
Reply
I've wondered when this would come up
Yes, the easy way is to design it like a covered bridge, a truss. The openings are the tough part. A modified queenpost would probably give the most options. This makes it rigid in one direction but the two sides still can distort in relation to one another... truss all 4 sides.
Stacking 3@2x12 does not create a 36" deep beam unless you can eliminate horizontal shear, not easy but possible with an engineer (which any of this would need really). Typically stacked beams like that are viewed the same as 3 2x12's side by side as in a typical built up girder. Now you are building a girder bridge. I've readily ordered 24" deep LVL's. Steel is an option... bolt it to well designed steel legs and those piers are rigidly braced, a moment resisting frame.
|
|
KinAlberta
Member
|
# Posted: 23 May 2016 22:44 - Edited by: KinAlberta
Reply
Yeah I was thinking glue and pins between the 2x12s all along their length and steel plates on the sides might tie them together. They'd be laminated horizontally and vertically. 3 wide 3 tall or whatever.
Also since headers are often 12" could a perimeter header be built and tied to the floor well enough (cables or strapping) that it too would work as a beam suspending walls and floor under it.
A framed wall with diagonal bracing and clad on both sides with say 3/4" plywood might increase the rididity of the wall as well. The diagonals would have to bypass windows though.
|
|
Don_P
Member
|
# Posted: 23 May 2016 22:54 - Edited by: Don_P
Reply
Google "keyed beams" and you'll probably turn up images of historic ways of doing this.
Finally found an old pic on p-bucket. I was playing around modeling a queenpost truss. The bottom chord is 3 pieces ~ 2' long each. There is an X of tension wire in the center panel.
|
|
KinAlberta
Member
|
# Posted: 24 May 2016 00:15 - Edited by: KinAlberta
Reply
Very cool!
The bottom is 3 pcs! Pretty amazing.
So if you build a lower wall like that, shouldn't one be able to build more trouble free small cabins on piers. The load gets spread out across multiple piers, so much further than without, that pressure points are reduced. It would behave more like slab on ground then wouldn't it? (In my city there are literally thousands upon thousands of detached garages just build on concrete pads sitting on gravel bases. Good drainage and everything moves in unison if at all.)
Yeah, I recall seeing some show on some ancient culture (possibly Ancient Rome) where their earthquake resistant building included diagonal braces in the walls, etc. And it was many, many centuries before those building practices were rediscovered/reapplied.
I've also wondered if wrapping beams with wire like a rope around a mast wouldn't work wonders. Or cables going from the top of a beam at each end to the bottom at the centre and vise versa couldn't offer enough tension to add much rigidity to a beam.
|
|
bldginsp
Member
|
# Posted: 24 May 2016 07:36 - Edited by: bldginsp
Reply
Quoting: KinAlberta So if you build a lower wall like that, shouldn't one be able to build more trouble free small cabins on piers.
The point of making a beam or truss is to span a large distance, so the point of making walls that are beams would be to reduce the number of bearing points. You could build a cabin on four corner piers, with beams spanning between. But you would still have the basic problem with piers, which is that they move independently of each other because they are not connected.
Quoting: KinAlberta The load gets spread out across multiple piers, so much further than without, that pressure points are reduced.
A standard stick framed wall with plywood is so strong in itself that it is strong enough. Actually probably stronger- so what becomes critical are the connections at the top and bottom, and the quality of what you are connecting to. A rigid wall, or beam, is useless if it isn't connected well to its adjacent components. And it's always best if it is connected to a rigid, immovable foundation, which piers aren't.
Good basic construction thinking though. The plywood box you describe is the basic idea behind shear walls and lateral bracing in general.
Another commonly used idea along these lines is the wall frame, which is a shear wall with an opening, such as a window, in the middle of it. Ordinarily shear walls can't have any openings, high or low. But if you put horizontal blocking on either side of the window, at header and sill, for three bays beyond the window each side, and then strap horizontally across all this blocking, it causes the whole length of wall, including the window, to act as one shear wall, or one rigid panel, rather than two rigid panels with a window in between.
|
|
Don_P
Member
|
# Posted: 24 May 2016 07:49
Reply
Quoting: KinAlberta cables going from the top of a beam at each end to the bottom at the centre and vise versa couldn't offer enough tension to add much rigidity to a beam
"Bowstringing", it is used in repairs sometimes. If you are in an old CCC era picnic shelter look up at the trusses, they were often designed a little on the light side. I've seen retrofit bowstrings in several.
Generally, the prescriptive codes lean towards uniformly distributed loads rather than concentrating loads. This lowers unit stresses. For a simple span bridge to last it needs substantial footings and abutment walls rather than piers. The walls stabilize it laterally in the downstream direction, the banks that the bridge is pinched between stabilize it in the cross stream direction. You need to pick up that lateral resistance somehow. It won't be 8" sonotubes or willowy water pipe piers. It is quite possible and would be a fun project maybe with an engineer intern or as a class project.
Wrapping a wire around wood, there is a post frame company doing that to posts using a synthetic fiber and some wonder goop. An easier way to get continuous fiber wrapping a wood beam is to use the log in the round. The fibers forming the timber are then unbroken as they travel around knots and travel in a path that didn't consider that we would later try to make that wonderful natural round cable/column into a rectangle, shearing through many of its individual chords. Round wood design values are significantly higher than sawn timber... into the injunears playground again.
I've deleted this several times in the past because it can and has lead to unfortunate results. Traditional Japanese timberframing is unbraced in our sense of the word. They resisted seismic events by swaying elastically rather than behaving rigidly... do not try to go there, centuries of trial and error, and we've seen photos of many collapsed in major shakes. Just pointing out there are several working paths through time.
|
|
KinAlberta
Member
|
# Posted: 24 May 2016 08:52 - Edited by: KinAlberta
Reply
Thanks for posting all the interesting information!
Regarding piers (and building on blocks)
Some benefits I see are that the increased building stiffness reduces uncertainty of piers not being on equally firm ground and it increases easy of simple cost free corrections. It also reduces potential damage from pier movement. It might help for building on stuff like permafrost or other grounds where full foundations don't work very well. It also may be a way to avoid the deep or extensive foundations that are always recommended but aren't always feasible due to location or cost.
Full perimeter foundations though always sound like the way to go (to do it 'right'), but then I wonder why not then go with a full basement and a potential doubling of the usable floor space. It's just a case of moving a bit further along the cost continuum.
Improved cantilevering though seems to be the most interesting result.
|
|
|
Don_P
Member
|
# Posted: 24 May 2016 22:31
Reply
You made one of those leaps there somewhere. You are thinking vertically, that is not the only force acting on a foundation and is not the reason piers require engineering.
A rigid frame does a better job of transferring all the forces from above to the foundation, no energy is absorbed by a swaying frame... the same reason we want a stiff racket, bat, club or stick, it delivers all of the force applied. Hit a golf ball with a limp noodle and it isn't going anywhere. Or the adage "load goes to stiffness".
Google continuous beam reactions, which is probably closer to what we are taking about. The load delivered to each support along a continuous beam varies. Start kicking out supports and then look at the loads... assuming you have built that much rigidity into the frame and foundations $. We've spent more for a less sound structure.
|
|
KinAlberta
Member
|
# Posted: 4 Aug 2018 00:38 - Edited by: KinAlberta
Reply
Will do.
I have no plans on reducing supports. Could easily add more blocks along the length of the old garage walls. The boathouse. It sits on an overgrown sand beach. Not sure what more blocks and added beams would do. Probably help.
However since water can soften ground to different degrees along the length of a wall, or variable snow loads before freeze up or after thaw could put substantially more weight over different portions of a wall, my thinking is that a stiffer wall would spread the downward forces across more supports and minimize any uneven settling /sag and risk of separation of sheathing and metal panel siding. (On the old garage package sitting on blocks, I suppose I could jack the building up say 12†and run a new beam under the walls.)
I’m still learning about shear walls. This article was very interesting to me:
Double-Layered Single-Sided Shear Walls? | JLC Online | Engineering, Panels, Building Codes, American Society of Civil Engineers
https://www.jlconline.com/how-to/framing/double-layered-single-sided-shear-walls_o
|
|
rockies
Member
|
# Posted: 4 Aug 2018 18:30 - Edited by: rockies
Reply
In thinking of the wall purely as a structural device are you forgetting about how to insulate the lower part? You'll have to build another wall around the beam in order to insulate, run pipes or wiring. It seems that having the beam come up higher than the floor surface creates an equal number of problems to the one it might solve.
|
|
KinAlberta
Member
|
# Posted: 5 Aug 2018 04:31 - Edited by: KinAlberta
Reply
Good points. Lifting would mean that electrical height would no longer be to code. Windows would be higher too. (In my case the old garage is uninsulated and has no windows. It’s essentially just a storage shed with a dirt floor. Maybe $5k Cdn in materials to replace since the one in this sales brochure is $6682 but has soffits, windows, insulated door, but like ours, no siding )
http://www.windsor-edmonton.com/Images/products/garagepackagespdf499450946.pdf
|
|
KinAlberta
Member
|
# Posted: 22 Mar 2019 21:29 - Edited by: KinAlberta
Reply
Design and Construction of a Monocoque House
Technical Series
90-227
http://publications.gc.ca/collections/Collection/NH18-22-90-227E.pdf
See page 20 of 28:
“In 1990, Natural Resources Canada was eager to field test a new rigid monocoque building design developed for possible use in permafrost locations. Designed to be constructed on-site with dimensional lumber and plywood, the test structure was built at Yukon College in 1991 by pre-apprenticeship carpentry trainees. The project evolved to become a test bed for off-grid hybrid alternate energy technology using a wind generator and photovoltaic solar collectors to power
a battery/inverter system. It even had a composting toilet.
The original building design worked but high material costs and space-limiting construction details meant that it was not tried again. The four bearing points of the building foundation are now hidden behind skirting. “
http://www.energy.gov.yk.ca/pdf/pushing_the_envelope.pdf
|
|
rockies
Member
|
# Posted: 23 Mar 2019 18:03 - Edited by: rockies
Reply
Ah, permafrost. Permanent frost, although with climate change all that permanently frozen ground is starting to thaw and the buildings are tilting and sinking.
Up north in the Arctic circle the government was looking for a way of building on unstable soils and they came up with the Multipoint system.
http://multipoint-foundations.com/
This system was adapted from the old geodesic dome design of interlocking struts (rather than curving into a circle the system was flattened out to become a platform). It's very strong and remains rigid and level on unstable soils (just adjust the posts whenever the ground sinks under a spot).
|
|
|
