Subject: RE: Reverse Camber (was: RE: Long cracks in
From: Gil Brock <gil(--nospam--at)raptsoftware.com>
Date: Wed, 10 Jan 2001 09:41:26 +1000
This effect was noticed in many single span (with integral end columns) and
multi-span parking garages in Australia about 20 years ago. Once we
determined the cause, we then started designing for and detailing for
temperature differential effects in all prestressed roof members.
As you said, it is necessary for the bottom reinforcement to be fully
developed into the supports at both end and interior supports and it is
quite often necessary to add extra reinforcement in the bottom to contain
the cracks to reasonable limits as the bottom reinforcement required for
flexural strength is often very nominal in these members, especially with
bonded prestress as is used in Australia and Asia. For wide flat beams
(bands) we have found that a minimum area of reinforcement of #5 @ 8"
(16@200) is necessary in the bottom.
It is also necessary to check for this effect at midspan as the tensile
stress will add to any tensile stresses due to loading and will result in
cracked sections at mid-span while the designer may think he has uncracked
sections in these areas due to normal loading. The temperature stresses in
a previously uncracked member due to temperature differential alone could
be as high as 500 to 700psi thus cracking the concrete by itself. It is not
logical or economical to try to resist these stresses with more prestress
so a partial prestress design is needed allowing the concrete to crack and
providing sufficient reinforcement to contain the cracks to reasonable
levels. Fortunately, once the concrete cracks the temperature differential
stresses reduce (they are stiffness dependent) and less reinforcement is
required than would be for the full temperature differential effect on an
Interestingly, in some of your colder climates, the reverse is true and
continuous reinforcement would be required at the top at mid span to
counter a reverse temperature differential (colder on top) for concrete
roofs of enclosed buildings. In fact, it may be necessary to allow for both
effects in areas with relatively warm summers and very cold winters.
Even though similar temperature differential effects (but reduced due to
the effects of the higher levels of normal cracking) occur in reinforced
concrete members, the cracking does not occur or is not noticed in
reinforced buildings as
- the compression stress due to self weight of the member tends to
counter the temperature effect in the areas of the members where we do not
normally expect tensile stresses (at the bottom at a support or top at
mid-span) thus resulting in net compression in these areas and
- at areas where both effects cause tension (bottom at mid-span or
top at supports) we have bonded reinforcement which has been added for
flexure to contain the cracking. In this case, extra reinforcement may
still be needed for crack control so if you have a reinforced concrete roof
or spandrel member which is cracking more severely then expected in the
bottom at mid span in hot weather or at the top at supports in very cold
weather, the cracking could be caused by temperature differential effects
adding to normal loading effects.
At 09:14 9/01/01 -0600, you wrote:
This was an interesting one that I read about several years ago. I believe
that it was in Concrete International.
A long span PT girder was cast for a parking garage top deck. The project
was in an area of low seismicity, in the sun belt, and the girder was cast
integrally with the columns, but was not designed as a lateral frame. As is
generally the case with PT, the bottom steel was minimal, and just embedded
into the column, and not fully developed.
Cracks developed at the bottom of the girder at the column. They were
epoxied periodically for several years, and they continued to open up. Some
studies were performed where they measured the elevation of the girder at
several points along the girder. As they were doing the measurements, it
was noticed that the elevation varied in a cycle during the course of the
day. They then measured the temperature of the girder across the profile,
and the deflections throughout the day.
As the sun heated the top deck the concrete expanded relative to the bottom
of the girder. The result was an increase of camber, reverse rotation at
the columns, and cracks as the girder cycled. As the thermal gradient
equalized the girder flattened out.
The solution was to cut joints where the cracks could appear that they were
planned, and epoxy inject the random cracks. The cut (control) joints were
filled with an elastomeric sealer. The problem was solved.
My own practice is to develop the bottom bar at PT girder to column
interface, and actually try to calculate the amount of anticipated cracking
due to column rotation for creep, shrinkage, and thermal cycling. This
involves intense calculations, the killing of a chicken, a full moon, and a
dart board. (humor intended, no animals were harmed in the preparation of
Harold O. Sprague
> -----Original Message-----
> From: Roger Turk [SMTP:73527.1356(--nospam--at)compuserve.com]
> Sent: Monday, January 08, 2001 5:47 PM
> To: seaint(--nospam--at)seaint.org
> Subject: Reverse Camber (was: RE: Long cracks in
> Harold Sprague wrote:
> . > There have been cases of the sun causing reverse camber in precast
> . > members.
> Could you expand on this?
> A. Roger Turk, P.E.(Structural)
> Tucson, Arizona
Regards Gil Brock
Prestressed Concrete Design Consultants Pty. Ltd.
5 Cameron Street Beenleigh Qld 4207 Australia
Ph +61 7 3807 8022 Fax +61 7 3807 8422