Just out of curiosity (I'm trying to learn about concrete), can anyone
tell me how weak a cold joint is. Assuming there is rebar crossing the
joint, that is. All the books and articles seem to pretend that the
joint is completely not connected at all, and even a "key" between the
two will only kind of help a little and the two pieces will still be
free to rotate against each other, shift, crack, etc.
I see buildings going up all the time here, with what must be hundreds
of pours (these are complicated university buildings, lots of foyers,
atriums, suspended walkways, etc.). Are those all flimsy cold joints?
Or do they do something special to make them stronger.
BTW -- I just learned about slip-forming... looking at the Millau
bridge project (http://bridgepros.com/projects/Millau_Viaduct /) --
Yikes, that is one rediculous huge concrete project.
Thanks in advance!
All huge slabs are poured in sections. Where possible, the seams are
placed where walls will cover them rather than in the middle of a room.
Seams are good - they allow the concrete to expand & contract without
It would be good if you noted what sources you were refering to
two will only kind of help a little and the two pieces will still be
free to rotate against each other, shift, crack, etc.<<<<
Depends are where the joint is located & in what type of member
a logical place for a cold joint is between a footing & a wall or
column. It's done all the time & I would not refer to the joint as
A pinned or keyed (when needed to be) cold joint properly detailed is
You mean how strong? Tension, compression, or shear?
Concrete is much weaker in tension vs compression, and cold joints even
Tile and stone mortar is a species of deliberate "cold joint". Look at
pull-off strength for some cementitious products on a manufacturer's Web
site, like http://www.mapei.com . Should specify something like 50 psi.
Compare that to 3500 or 5000 psi for concrete in compression.
If a 4-inch thick wall 8 feet high would pull apart at 50 psi, that amounts
to about 10 tons total force, best case.
Of course. All a key or hardware will do is convert the failure from that
of poorly bonded surfaces to homogeneous tension concentrated over a
smaller area. Such a conversion cannot be much stronger, and may in fact
Another factor is the "coldness" of the joint. Freshly set concrete is
still active on the surface and will bond well to a fresh pour, if done
within hours or a day or so. Aged concrete has an inert surface that
doesn't bond well to new.
Thanks for all the replies... but now I'm more confused. Some of these
replies seem to make no sense. See below
I think the expansion and contraction is for expansion joints, right?
This is where you see a deliberate cut down the middle of a garage
floor, for example. Or rather, where you see a thin wood board between
the garage floor and the house foundation. I am about where you don't
want separation -- you want the two pieces to stick together as if they
were one piece.
And I see pictures online of huge slabs being poured as one huge pour.
Seems to be pretty common. Here is one example:
Richard J Kinch wrote:
Huh? Seems like adding steel, which is just fine in tension, will add a
huge amount of strength, since it will take the tension, and the
concrete will take the compression.
And if a key makes it weaker, why would anyone ever add a key? Again,
it seems like a cold joint resisting a shear force might be flimsy
because it is just poorly-bonded concrete in sheer, but a keyed cold
joint puts part of the bond in compression, and puts the well-bonded
concrete into sheer instead.
This is useful (if it is correct). So that means if you pour the second
section just a day later, it is much better than if you wait a week or
The problem is not the steel, but the steel-concrete bond. Why is that
any better per unit area, or any more area, than the concrete-concrete
bond? The steel just pulls out of the concrete.
All a key does is trade an area of bond tensile strength for a smaller
area of bulk tensile strength. Whether the trade is an improvement
depends on the geometry and the relative values of the two types of
strength per unit area. The problem is that neither is that good.
Imagine that you joined the edges of two slabs with a butt joint versus
a dovetail key. In the former you have the bond strength per unit area
over the full area of bond. In the latter you have the tensile strength
per unit area, but over a much smaller area that is the base of the
dovetail. If the tensile-vs-bond strength ratio has to be large enough
to overcome the geometrical disadvantage.
Consider a dovetail joint in a wooden drawer. It is an improvement only
if the tensile strength of the wood is much larger per unit area than
the pull-off strength of adhesives, because the tensile area of the
dovetails is much less than the bonded area of a butt joint.
Absolutely. Concrete takes weeks to cure to nearly its ultimate
strength. The chemistry of the bulk material and surface activity
changes during that time.
humidity. If the slab is too large without controlled expansion
points, the slab will crack where you don't want a crack. That's why a
large slab is poured in sections, with the section lines located along
future wall locations - if the joints expand or contract you won't see
them. To make multiple sections stick together as one huge piece is
asking for trouble down the road. Maybe you should define how large a
pour you're talking about.
Look closely at the image and you will see the concrete is being poured
and worked in sections. There are obvious horizontal lines between the
columns and in the upper right next to the shadow you see a verticle
line. Another verticle line is to the left of the workers feet
extending up into the overpour of the next section. True, it's being
poured at the same time in the photo, but work could stop for any
reason and resume the next day with no problem. The verticle at the
workers feet seems to be a board - the typical way to make a tongue &
groove joint for the next day's pour.
Sorry, let me go back to the original post Forget slabs, forget
expansion joints, forget control joints to control cracking. The
question is about *cold joints*, where we want to permanently join two
pieces of concrete. For example, the joint between a footer (say a 1' x
1' x 4' block, to be specific), and a short stem wall sitting on the
footer (say 4" thick, 4' long, and 2' high). Imagine the two pieces
could be poured separately, on different days (say, 1 month apart), or
together at the same time on the same day with no pause between the
In the latter case, I think it is clear that there is no joint at the
intersection of the wall and the footer, unless someone wants to
*deliberately* put one there by sawing, trowling, etc. In the former
case, there is, what I think is called a cold joint. A new piece of
concrete is joined to a cold, already cured piece.
Most of you seem to be missing the question, I think. The question is
simply, how much worse is it to do it 1 month apart, rather than at the
same time? Will the rebar crossing between the footer and the wall
completely more or less make up for the fact that it was poured
separately? Or if a key is installed, or a combination of key and
rebar, would that make up for it? Or is the tensile and shear strength
of concrete so low anyway that the further weakening due to the "cold
joint" negligible, because even the continuous pour would be so weak as
to be negligible.
Well, maybe my terminology is not up to speed -- please let me know.
But when I say "poured together", I mean that the concrete is poured,
you know, together. On the same day, from the same truck, at the same
time. It really has nothing to do with whether someone cuts a control
joint, or puts in a board or key or whatever as an expansion joint. I
think they only way to get a cold joint (which is what I am asking
about) is to do two separate pours, on (say) different days of the
week, where one side has already cured substantially.
These have been answered. Tensile capacity of concrete is ignored in the
design. Tension and shear are taken up by reinforcing rods.
If cold joints were such a problem, concrete construction wouldn't happen.
I think much of the confusion is coming from terminology.
A cold joint in concrete implies to me that the pour was held up
(form failure/truck breakdown/pump failure, etc) that was not
planned. These can be cosmetic and/or structural problems.
A construction joint is a deliberate stopping point that is part
of the building design. Dowels, keyways, etc are designed
primarily to work in sheer. Reinforcing steel or mesh in slabs
doesn't keep the concrete from cracking, it keeps the cracks from
spreading. In structural concrete, the steel reinforcing is used
to overcome concrete's poor tension properties.
A contraction joint is a deliberate tooled or sawn joint to
accommodate concrete shrinkage. Concrete will crack, the
contraction joint tries to tell it where to do it.
Keep the whole world singing . . . .
DanG (remove the sevens)
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.