[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: GSBN: Rubble Trench Foundations

Dear Martin Ð many thanks for that very good description of how you make RT
footings Ð it seems that the approach is much the same in many different

Here are a couple of replies I received to direct queries about RT footings
from a couple of Californian based engineer colleagues  (with a few personal
notes taken out)

I see that there may be little alternative  to pouring at least a reinforced
concrete footing on top of the RT trench, with possibly some sheer pinning
as well Ð it is similar in a sense to a method we have been adopting here of
digging footing trenches and running in a self levelling flowable concrete
that a reinforced concrete beam is poured on top of Ð but gets rid of the
cement in the bulk of the footing and gets the building process out of the
ground fastÐ it seems that I have found an engineer down-under here now who
is willing to give it a go  - all your help has been invaluable.

Thanks and best wishes,


Hi Graeme, 
> Stone rubble foundations. By that, I presume you are talking about stone or
> gravel filled trenches. Possibly a reasonable idea in a dry, non-seismic area.
> But, it must be done with care in a wet and/or seismic region.
> Obviously, in a wet climate the base of the wall needs to be above the grade
> level and water needs to be carried away in a drain placed in the trench. This
> means that at a minimum a concrete grade beam needs to be poured on top of the
> stone rubble to allow a clearance of at least 6 inches (152 mm) between grade
> and the base of the wall.  Seismic regions require more attention to the
> anticipated behaviour of the walls and the overall structure during ground
> motion. I agree with Richard Walker regarding the shear wall being founded on
> a concrete foundation. But, this could be a reinforced concrete grade beam on
> top of a stone rubble filled trench.
> However, the idea of the structure floating on a rubble filled trench in a
> seismic zone is that the stone rubble will allow the building to slip and
> slide at ground accelerations that are much less than the peak acceleration
> produced by the earthquake. This means that the forces in the building walls
> are much less than they would be if the building were restrained from slipping
> by building on top of a conventional foundation. To make this work properly,
> however, the building has to move as a whole, which means, the building needs
> to be founded on a slab of concrete, or a continuous grade beam that is strong
> and stiff enough to make this possible. The basic idea behind the seismic
> advantage of the rubble filled trench (actually, a gravel filled trench would
> work better) is the coefficient of friction. The coefficient of rolling
> friction of a flat concrete surface on a plane of gravel is relatively low. To
> make this work properly, the coefficient of friction needs to be the same at
> all locations under the walls. If the coefficient is different at one corner
> of the building than at another, the area with the higher coefficient gets
> hung up and there occurs differential movement of the structure, causing
> stresses in the walls that could do serious damage. The best bet would be a
> flat slab foundation with rounded edges that would not dig into the foundation
> rubble if tilted. The building would move as a whole and there would be no
> differential wall movement. Anyway, that's the beginning of the discussion. It
> would be a good PhD topic, one ripe for testing on the shake table.

Best regards,

> Fred Webster

I don't think I could add much to what Fred just wrote.  RT foundations are
fantastic where you have the right kind of soils.  I can't remember offhand
which types are and are not good -- and allowed, in the USA -- for RT
foundations;  there was a good article about that some years ago in Building
Standards magazine.

They are less fantastic in high seismic risk areas, unless you think them
fully through as Fred did there, and even less so for earth wall structures
(it seems to me).  I just designed a house comprised of used shipping
containers sited directly astride a creeping fault rupture -- the site
shifts about a cm. per year -- and is founded on a very sturdy (0.5 m thick)
reinforced slab over a greased, sacrificial 5 cm. slab with a very smooth
top;  a crude base isolation system.  To work, it depends on the great
strength and rigidity of the thick slab, else all would come apart, much as
Fred described.  I can only imagine it would be more difficult with earthen
(brittle) walls. 

I suppose we should bear in mind the lessons from the historic (surviving)
earthen structures in seismic regions, which made it through big shakes by
having very thick walls, or more specifically very low h/t ratios (correct
me if I'm wrong anywhere in here, you guys both know more than me).  Graeme,
you're trying to economize on overall material use in a wet, shaky country;
knowing nothing else, I'd rather have the 0.5 m wide walls and concrete
foundation (or cap on rubble trench foundation) than have to build 1 m thick

Feel free, so far as I'm concerned, to condense and forward these comments
to the GSBN list  I think they would apply just as much to SB walls.
They're less massive, yes, but also more sensitive to cracking and the
ensuing water intrusion problems.

best , 

Bruce King, PE 
Director, Ecological Building Network  ( www.ecobuildnetwork.org )
Publisher, Green Building Press  ( www.greenbuildingpress.com ) 

--- StripMime Report -- processed MIME parts ---
  text/plain (text body -- kept)