On Wed, 21 Apr 2010 01:29:05 -0700 (PDT), sm_jamieson
wrote:

On 20 Apr, 17:25, Bruce wrote:
On Tue, 20 Apr 2010 07:11:49 -0700 (PDT), sm_jamieson



wrote:
On 20 Apr, 11:54, Bruce wrote:
On Tue, 20 Apr 2010 10:41:01 +0100, Tim Watts wrote:
If you do - this trick is fairly easy:

Drill into the side of the old slab at foot intervals and resin-mortar in
some heavy studs (M10-M12), say 4" into the old slab and 4" protruding.
That should provide a good mechanical lock.

That's not anywhere near long enough. *Thermal movements, shrinkage
and creep will be enough to break the bond, and the dowel bars (that's
what they are called) will be worthless. *I would recommend 200mm into
the old slab and 200mm protruding. *Otherwise, don't bother. *Also, I
would avoid studs and use plain mild steel bar - not the ribbed kind
either.

I've seen this done on a much larger scale when they widened the M25 bridge
decks near Jnc 7-8 - bloody great big bars where set into the sides of the
original deck and then the new slabs were poured incorporating these bars.
Couldn't see the details of how they tied the rebars to the mortared in
bars as the shuttering was up at that point - I assume they were well tied.

There is no need to tie the dowel bars to the rebar. *They are
performing two quite separate functions, each of which involves
interaction between steel and concrete. *Steel to steel interaction is
neither necessary nor particularly desirable, as it can induce
cracking of the concrete.

If a waterproof joint between the old and new slabs is important (the
OP has not made this clear) fix a small softwood fillet about 10 mm
wide by 15 mm high against the top edge of the existing slab. *After
the new slab has cured (at least a week) carefully remove this fillet
and clean out the resulting slot. *

Then almost fill the slot with a good quality polyurethane mastic.
Leave around 3mm clear at the top to allow upward expansion. *This
will give a watertight joint while allowing for the inevitable small
amount of movement. *

However, if a waterproof joint is not needed, this can be omitted.

This is a 5 inch thick kitchen floor slab - a 2 x 3m slab being poured
up against a 4x5m slab, on top of 75mm celotex. No mesh in the slab.
Will be screeded over both slabs, 75mm screed.

In which case, there is probably no need to seal the joint. *But even
with dowel bars, you will inevitably get a fine crack at the joint.
This will eventually be reflected through the screed and any tiles
that span the joint, although the good, thick 75mm screed will help to
minimise and/or delay this. *

I would be tempted to ensure that there is a joint in the screed and
also in the tiles exactly above the line of the joint between the
slabs. *That would help to confine any opening up to that one joint in
the tiles, which could be re-grouted when needed, and avoid any risk
of the tiles splitting. *

On the other hand, you could just ignore it and screed right across.
But there will be short term shrinkage of the new slab, and long term
creep of both slabs, and while the dowel bars will help reduce it,
they will not eliminate it. *Laws of physics (and chemistry) and all
that.

I guess most builders would just screed over and be done with it.


Sometimes a degree of ignorance can be a good thing. Or, to put it
another way, knowing too much can get in the way of an uncomplicated
life.


Floor slab joints must be very common with extensions etc.
Best thing is to make the dowel bars of a reasonable size I suppose.


The dowel bars and thick screed will probably do the trick.

Another way of minimising the problem is to include some chicken wire
or brick reinforcement mesh in the screed where it goes over the joint
between the slabs. One brick's width should be enough. Instead of
one concentrated crack above the joint, it will create many tiny
micro-cracks that are too small to see with the naked eye and won't
affect the tiles above.