Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Fire in AS3600-2009

Status
Not open for further replies.

degenn

Structural
Jul 26, 2009
39
There appears to be some major errors in in the column fire resistance section 5.6.

Data for axis depth of longitudinal bars in columns has been derived from Eurocode EN 1992-1-2:2004. Several errors in the conversion of this document have been noted. They include simple typographic errors as well as misinterpretation of the original document.

The effect on calculations is significant and not conservative. For example, axis depth for bars in a 400 column for 120 minute FRP is 53mm in EN1992-1-2:2004 and 38mm for the equivalent design using AS3600. 38mm gives a FRP of only 50 minutes according to EN1992-1-2.

1.1 Calculation of Load Level for Table 5.6.3
The load level is given as N*f / Nu in 5.6.3 of AS3600. The equivalent in 5.3.2 of EN1992-1-2 is Ned.fi / NRd. NRd is the “Design Value” which is the “Characteristic Value” reduced by a “Partial Factor”. The load level is therefore more correctly represented as N*f/?Nu.
The effect of this is roughly increasing load level from 0.5 to 0.7.

1.2 Typographic Errors in Table 5.6.3
Table 5.6.3 is copied from Table 5.2a. The data for FRP values 120-240 are copied incorrectly to the row above. This creates the obvious abnormality that distances for FRP120 are less than FRP90.

1.3 Calculation of Load Level for Table 5.6.4
The value of ? is calculated using characteristic values for steel and concrete. In EN1992-1-2:2004, the value of n is based on design values. The value of ? is therefore underestimated.

1.4 Allowable Eccentricity for Slender Columns
Values tabulated in Table 5.6.4 are based on an eccentricity (e/b) of <=0.025. The values in the table, however, are derived from Tables C2, C5 & C8 in EN1992-1-2 (using Le/r of 30). These tables are for e/b <= 0.25.
Minimum e/b in AS3600 is 0.05, so the requirement, as written, is unusable.

1.5 Reinforcement Ratio
Table 5.6.4 uses the reinforcement ratio p, with values of 0.01, 0.05 and 0.1. The values used in EN1992-1-2 use a reinforcement ratio ? defined as Asfyd / Acfcd. The conversion ratio used is 0.1 to a get steel percentage. Allowing for the appropriate partial factors, this conversion is correct for f’c = 65MPa. As AS3600 is applicable for concrete ranging from 20MPa to 100MPa, the implied error in reinforcement ratio can be large.
A reinforcement ratio of 10% is very unlikely in a column, but the equivalent ?=0.1 is easily achieved with 3% steel in a 20MPa concrete column.


I don't suggest using this section of the new code.
 
Replies continue below

Recommended for you

Thanks Degenn

They had 8 years between editions and it still looks like it was rushed out at the end.
 
For slabs, Table 5.5.2(B) specifies a minimum slab thickness of 200mm for a 90 minute FRP. This includes PT, as there is a cover increase specified for PT.

It also requires 20% continuous top reinforcement.

This is derived from Eurocode2.

This is at odds with current Australian practice where much thinner slabs are routine.

Any comments?
 
Table 5.5.2(B) does only apply to flat slabs, does that affect your comment?
(I haven't designed a flat slab for many years).
 
The table refers to "flat slabs including flat plates". The definition of a flat slab is a "continuous two-way solid or ribbed slab with or without drop panels, having at least 2 spans in each direction, supported internally by columns without beams....".

I consider this applies drop panel slabs and to banded slabs, since a band is a slab thickening (i.e. if no shear ties).
 
Your point re banded slabs is the contentious one.

I would say that they aren't classified as flat slabs; it's a one-way slab spanning between band beams.

Perhaps others can share their view.
 
I agree with apsix. A band beam and one way slab system is not a flat slab. Besides, many band beams do require shear reinforcement.
 
dgenn,

Have you discussed any of this with the committee or are you floating it here first.

Getting the relevant sub-committees comments would help people a lot in deciding how to approach this!

RE the band beam is a flat slab question, in terms of general design, I would agree with your comment, but as far as fire rating is concerned, I would think it is regarded as a beam and slab system.
 
I've forwarded this to Bob Potter who was on the sub-committee but I think now retired. He agreed the table errors (point 1.2) are a typo and will be fixed in first amendment. Other issues have been referred to the sub-committee for consideration.

Eurocode 2 classifies slabs as
- simple supported solid
- continuous solid
- flat slabs
- ribbed slabs

Continuous solid slabs can be 120mm for 120 minutes while it is 200mm for 120 minutes for a flat slab. Only the flat slab requirement of 200mm has been passed to AS3600.

AS3600 includes a note in 5.3.1 refering to methods in Eurocode2. Presumeably this allows us to take rules from Eurocode 2 and still comply with AS3600. That is, consider a banded slab as continuous solid with a minimum of 120mm.

 
degenn,

Now you will just have to wait for 6 months for an answer, like I am still waiting for my question to them!!

I would have thought a flat slab was a solid slab, either simply supported or continuous! I will see if I can get some definitions of the EC2 stuff.
 
I'd say that a flat slab is one not supported by beams or walls, and with at least 2 continuous spans in each direction.
 
The term ‘flat slab’ has no universal definition. Eurocode 2 defines flat slabs as slabs supported on columns. BS 81102 explicitly includes waffle or coffered slabs. Thus slabs supported on beams and walls are not flat slabs.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it
 
At least AS3600 and EC2 are compatible in their definition.
BS8110 has a broader definition.
 
RE Banded beam and slab construction,
I had a talk to Ian Gilbert on this yesterday. In his opinion (not an AS3600 response) as long as the slab is designed to span purely in the direction transverse to the bands, not utilizing any strength parallel to the bands, and as long as there is a true band beam present, it can be considered as a one way slab supported by beams for fire design to AS3600.

RE Fire rules in AS3600
5.4.1(b)(ii) the wording is a complete mess and is meaningless
5.4.1(b)(iii) the wording is a complete mess and is even more meaningless than (ii)
5.5.2(b)(i) anyone know how to satisfy this? It is meaningless also
5.5.2(b)(ii) is extremely erroneous

I have reported all of these to the committee along with some other problems I have with ductility, rectangular stress blocks and biaxial bending.

RE AS3600-2009
There is a meeting in the first week in June to consider problems with the document. If anyone has any other problems they have found with the new code, can they please pass them on to me (or post them here if you do not know where to find me) and I will pass them on to the committee.
 
Hi Rapt,

The problems seem to come from taking bits from Eurocode 2 but not explaining them.

The attachment shows a diagram from Eurocode 2 showing how redistribution can be done of positive moment, based on reduced fire load to the negative (cold) steel. This is ,I think, what the redistribution comment refers to.

The methods in Eurocode 2 are very good in the rigorous approach they take, but they are very time consuming and can't be easily simplified as the code committee has tried to do.

Perhaps there is a need for some new software! I have afew ideas.
 
 http://files.engineering.com/getfile.aspx?folder=73999248-22c5-43ef-8bfe-b19d38fef73f&file=fire_redistribution.GIF
Status
Not open for further replies.

Part and Inventory Search

Sponsor