Post Tensioned Free Standing Firewall

[jike]

Has anyone on this forum ever done a post tensioned, free standing firewall in precast concrete or concrete block? Just looking for some alternatives to conventionally reinforced, free standing masonry.

 

[Structcon]

The "firewall" part is curious ... 1hr, 2hr, 3hr, 4hr? Both get you the ratings, but you need a certain cover and you need 6" or more. The "p/t" part is structural, and not fire-related ...

 

[msquared48]

Done sound walls, none PS. No fire walls.

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[FFENGINEER]

I have seen them done out of Tilt-up, or site cast precast.

 

[jike]

I am looking at doing a 4 hour firewall. Yes, the P/T is the structural part and not fire related except the cover.

 

[ishvaaag]

Sometimes precast panels are prestressed because of the lessened size of the reinforcement, and help to deal with tensile stresses; a high fire rating will help because of the thicker sections making less relevant the effect of geometrical imperfections (like in curved shells in bridges upon application of the PT forces the attempted straightening might try to eject -spall- some section of the structure, or deform it). Also, remember that in general tests on column-like structures have been made to slendernesses 1/30, going above that uses not to be a problem bur somewhat unwarranted. And, make checks for a relatively ample range of variation of PT loss, to ascertain the usefulness and performance of your design.

 

[cntw1953]

I came accross several websites that claim 4 hours fire endurance of precast concrete pannels and solid CMU, not sure what were the reinforcing in their's products. I think cover thickness and heat effect on post-tensioning reinforcing are the most crucial factors in design. Have you checked PTI or PCI? May have some useful info there.

 

[Structcon]

4 hour firewall - structural cantilever - non conventional ... ok then. For the fire rating, you can check the IBC chapter 7 - it tells you minimum thickness based on material (cmu or conc.) and minimum cover to steel (more critical is you do go ahead and use p/t). For cmu, I would need an 8" solid grouted wall for a 4 hour rating. In terms of using a non-conventional structural solution, I wouldn't waste my time unless I had a huge area of wall to make it economical for the contractor to build - prices come in much lower (especially now) if it is conventional. If you do want to use p/t, then it would normally be concrete and not cmu, since the tendons have way more tensile stress than you would want to bear onto cmu - often cmu post-stressing is done with threaded rods - there are some tech guides from NCMA on this. In terms of using p/t and concrete, you would encase your base anchors down in the footings, so I wouldn't think precast would be very practical to thread the cables thru, so I would do cast-in-place. Finally, I wouldn't recommend doing this wall as p/t unless you've done p/t slabs and beams before.

 

[jike]

I have talked to a precaster that has provided freestanding 4 hour firewalls in the past, but was just wondering what others experience has been.

There certainly is no problem achieving the 4 hour rating. I was more interested in the method that they used to anchor the panel to the foundation.

 

[Structcon]

So the "firewall" part is irrelevant now, ok, it's just a structural issue then... my guess is the precaster has anchored these free-standing walls with an anchorage detail very similar to the hold-down detail for a tilt-up wall. See tilt-up wall guidebooks. I've never seen a precaster provide a conduit longitudinally in the panel for placement of a tendon, unless you're referring to 8" or 12" prestressed planks. I think you need to do some more checking with you local contractors/precasters to find out what they can do, since we (SE's) can do anything!

 

[concretemasonry]

For masonry, the fire rating is based on an actual test of a panel (unreinforced block with face shell mortar bedding) and it depends on the equivalent thickness of the units and the type of aggregate. In the interest of testing time/speed and economy, the same wall is used for the fire resistance portion (temperature rise and passage of gas/smoke) and the stability/thermal shock when hit with a hose stream. Many walls of other materials use two separate tests, instead of combining the two portions in one more severe condition.

Lightweight aggregate will always have a superior fire rating because of the lower heat transfer rate, consequently, the equivalent thickness for lightweight units do not have to be as high and often normal units can be used.

As an example, for a 2 hour rating a normal hollow 8" block will provide a 2 hour rating, while a normal weight unit must have a 1 5/8" face shell thickness instead of a 1 1/4" thickness for lightweight units. When you go to a 4 hour rating, some standard 12" thick units made using lightweight aggregate will qualify. For 8" walls, even with lightweight aggregate the face shell thickness will have to be greater (possibly as much as 2 1/8").

The use of reinforcement is strictly a structural question and not a fire resistance requirement for masonry.

The U.S. engineers are not a familiar with the structural aspects of P/T masonry as foreign engineers are, so there may be some good references out there.

If this is a industrial/refinery situation or a power substation the blast effects could be the critical item and not fire rating.

Dick

 

[JLNJ]

That would need to be one heck of a hold down detail if the pc/pt wall were free standing (i.e. cantilevered). I'd love to see a detail of how this would be accomplished.

 

[JAE]

jike,

Using P/T vs. mild rebar - both creating what you are looking for - a vertical cantilevered wall that can be free standing if structure on either side falls down in a fire.

I don't see why the P/T option is really any different than developing vertical bars into a footing like a cantilever retaining wall. One thing to watch for is the shrinkage of the concrete vertically reducing the pretensioning but that is typical P/T loss calculations.

I think the problem is more constructability - how do you hang the tendons prior to the casting of the concrete.

Also, with prestress, you want to have tendons on both faces. If you use tendons down the middle, and have a small eccentricity in the tendon, you might run the risk of a wall buckling suddenly.

I have done a similar vertical "prestressed" detail before but not for a fire wall. This was a small bell tower in which the architect wanted to have six 3 5/8" x 7 5/8" mini-columns radiating around the bell cage at the top. The mini-colums held up the roof.

We used straight stacked brick and installed two vertical threaded rods through the brick holes - welded to a bottom plate and after the last brick was placed on top, added a plate and tightened the bolts down to prestress the brick and create a structurally competent column.

 

[jike]

We will be constructing the firewall out of pre-tensioned precast panels. Footing 4 feet down and wall tied in at floor similar to tilt-up with coil inserts.

 

[structuresguy]

I have looked into doing single, free standing, masonry fire walls many times. I have never been able to find a good solution for anything higher than about 15-20 feet using single wythe of block, either PT or mild reinforced. My problem with PT walls in this applicatin is the tendency to buckle due to lack of restraint. I have done slender PT masonry walls that were braced to roof before, but won't use one in cantilever. In the end, we always use two 8" wythes, back to back, one attached to each side. If one side collapses, there is still one wythe of block wall left standing.

I have had to put openings through this type of wall also, and there I just used a CIP frame over the opening, designed with all the cover requirements etc. to meet the fire rating. The design assumption here is to make the arch strong enough that it won't collapse in the even that either side does. And then put a fire door inside the opening, or on either side of the opening.

 

[jike]

structuresguy:

I have done exactly the same thing for double walls, using a CIP frame at openings, detailed similar to what NFPA recommends.

My single CMU walls have gone higher than 20 feet. We are not in a significant earthquake zone and we design them for 5 psf plus the wind load on the parapet. Many times the wall is capable of more than that.

 

[vandede427]

I do them out of 12" CMU all the time, cantilevered from the base up to heights of 20'. Above that I'd use a cast-in-place conc wall. I wouldn't mess with PT on a firewall.