NCC 2016 - Robustness 1

[Trenno]

Hi all,

I recently attended a Concrete Institute of Australia's seminar: Seismic Design and Detailing for Reinforced Concrete Buildings and there was slide detailing the new clauses that we, as structural designers, need to recognise and address. I'd like to get people's opinion on these new requirements, because they could well have a significant impact upon daily structural design practices.

From what I understand it's a Eurocode requirement and may only need to be implemented for buildings with Importance Level of 4. Apparently there are some deemed-to-satisfy rules for regular buildings that designers can adopt and implement (ductility, redistribution, confinement etc).

BP1.1 From Pg 70 from NCC2016 Volume one

(a) A building or structure, during construction and use, with appropriate degrees of reliability, must—
(iii) be designed to sustain local damage, with the structural system as a whole remaining stable and not being damaged to an extent disproportionate to the original local damage


BV2 Structural robustness From Pg 75 from NCC2016 Volume one
Compliance with BP1.1(a)(iii) is verified for structural robustness by—

(a) assessment of the structure such that upon the notional removal in isolation of—
(i) any supporting column; or
(ii) any beam supporting one or more columns; or
(iii) any segment of a load bearing wall of length equal to the height of the wall,
the building remains stable and the resulting collapse does not extend further than the immediately adjacent storeys; and

(b) demonstrating that if a supporting structural component is relied upon to carry more than 25% of the total structure a systematic risk assessment of the building is undertaken and critical high risk components are identified and designed to cope with the identified hazard or protective measures chosen to minimise the risk

 

[IDS]

I think this is an important new development as well. The ABCB documents are all available for free download:
NCC 2016

There is also a handbook on structural robustness
Structural Robustness

This all ties in with a paper I presented at the Concrete conference in Melbourne last year (attached).

This all raises several questions:
How do we ensure that practicing engineers are aware of these requirements (other than waiting for a building to suffer disproportionate collapse due to the failure of a single column)?
How does this affect requirements covered under "safety in design" legislation?
How should these requirements be covered in the Australian Standards relating to building design?
How should these requirements be applied to non-building structures?

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

Trenno,

There is currently a clause in AS3600 (2.1.3) on Robustness, which refers you to AS1170.0 Section 6 which requires the check on all buildings and provides a deemed to comply solution. This clause is similar in wording to the NCC deemed to comply clause (but not updated to the latest).

The calculation method for this is new in NCC and the next AS codes will be modified to give better guidance on it but it an option now (badly worded in my opinion) in AS1170.0 clause 6.1. It has been in BS8110 since the 1960's and is in Eurocode as you mention.

 

[IDS]

It seems to me that the new requirements in the NCC are much more stringent than those in AS 1170.0.

The general statement:
"Structures shall be detailed such that all parts of the structure shall be tied together both in
the horizontal and the vertical planes so that the structure can withstand an event without
being damaged to an extent disproportionate to that event."

might be interpreted to require consideration of removal of a column (or might not), but then it says:

"Clause 6.2 is deemed to satisfy this Clause."

and Clause 6.2 explicitly does not require removal of a column.

On the other hand, the current commentary to AS 3600 does require this:

"The Standard requires that concrete structures be designed to
be robust in accordance with Section 6 of AS/NZS 1170.0 (Ref. 2) ...

A structure is to be designed such that should a local
accident occur, the damage is contained within an area local to the accident or, should one
member be removed, the remainder of the structure would hang together and not precipitate
a progressive collapse. This requires that the structural members and the connections
between them have adequate ductility (Ref. 7)."

So even before the new NCC requirement, legislation required all risks to be removed, "so far as is reasonably practicable", and the statement in the AS 3600 Supplement can be taken to indicate that designing structures to remain standing after removal of a column is "reasonably practicable", so everyone should be doing it already.

I don't work on building design, but I suspect that this is not general practice at the moment. Certainly designing some bridges to remain standing after removal of a column might prove difficult.


Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[IDS]

Tim Minchin on risk assessment

... amongst other things.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

IDS,

But the new NCC also gives the option of a deemed to comply solution that does not require removal of each supporting member. Same as the deemed to comply in 1170.0.

The main new change in NCC is the option for a calculated solution which does require this. And as you said, 6.1 in 1170.0 could be read as requiring this or it could be read to say anything you want to read into it, the way it is currently written.

So deemed to comply is less conservative than the calculated solution in both documents.

That is why we have to sort out the logic in both NCC and AS!

 

[IDS]

But the new NCC also gives the option of a deemed to comply solution that does not require removal of each supporting member. Same as the deemed to comply in 1170.0

Have you got a clause reference for that? I couldn't find any deemed to satisfy clauses for the support removal requirement.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

IDS

BV1!

 

[IDS]

IDS

BV1!

As you know, I don't work in areas where the NCC applies, so I'm certainly not an expert on the interpretation of the words, but my reading of BV1:

Compliance with BP1.1 and BP1.2 is verified for the design of structural components and connections when—

is that when you are checking the design of the components and connections after removal of a support, they should comply with the listed statistical requirements. I can't see how it could be taken as a deemed to satisfy provision for BP1.1(a)(iii), because it specifically refers to "structural components and connections", and BV2 refers to structural robustness.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

IDS,

I do not agree with the current wording, but the NCC Handbook specifically says you can use either the Performance Solution (what you are talking about) or the Deemed to Comply Solution or a combination of the 2!

And the Deemed to Comply does not allow for loss of a supporting element, so as far as I am concerned it does not comply, but the NCC currently says that it does!

 

[IDS]

I do not agree with the current wording, but the NCC Handbook specifically says you can use either the Performance Solution (what you are talking about) or the Deemed to Comply Solution or a combination of the 2!

Thanks, I've now had a look at that (Handbook - Structural Robustness), and my conclusion is that although there is much in the handbook that is well presented, overall the situation is now a bit of a mess.

As you say, the handbook is quite specific that the "Deemed to Comply" solution is acceptable:

Compliance solutions are the means of satisfying the Performance Requirements. The NCC provides different options for compliance being: a Performance Solution, a Deemed-to-Satisfy Solution or a combination of these...

There are two possible compliance solutions contained within the NCC. These are described in the following sections.

The problems being that:
- The "deemed to satisfy" provisions refer to AS design codes that either have no specific requirements for robustness, or have specific requirements that in general will be much less effective in ensuring a robust structure than the removal of a support method.
- AS 3600 does have a requirement that structures should be checked for loss of a support (even though the handbook says it doesn't), but it is buried in the commentary, where many engineers will be unaware of it, and those that do know of it probably ignore it, because there are no specific requirements in the body of the code.
- Neither the NCC documents (so far as I have seen) nor the AS documents make any reference to the safety in design legislation, which requires that all risks shall be removed or reduced "so far as is reasonably practicable".

When we have our next Newcastle earthquake, or worse, and the lawyers are looking for individual engineers to blame for buildings collapsing, it's hard to see how following the minimum horizontal force requirements in AS 1170.0 could be justified as removing risks "so far as is reasonably practicable", even if the NCC Handbook says that that is all you need to do.



Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

IDS

I think the correct reading of 6.1 in 1170.0 second paragraph which requires

"the structure can withstand an event without being damaged to an extent disproportionate to that event. "

is to require that the structure be able to survive the removal of any individual supporting element. As I said, very badly worded and can be interpreted any way you like.

But then the next paragraph allows the designer to use the deemed to comply clause 6.2 which has no requirement to check for this.

Same problem as the latest NCC!

I have passed my comments on this to both the AS3600 and AS1170 committees over the last few weeks and the problems are being discussed.

In the meantime, everyone should be checking for individual supporting element removal using a combination of ties and tension members as that is going to be the result when it is all sorted out.

 

[Trenno]

Say you have an irregular grid flat plate - do you have test the removal of every single column and design each situation accordingly?

If this clause were to be retrospective, surely a large number of buildings would be found to be non-compliant.

This new clause could have huge repercussions, some guidance and clarification on the matter is definitely needed!

 

[rapt]

Trenno,

You will never make it work with corner columns in a flat plate, unless you have significant edge beams (which we used to always provide but seemed to go out of favour when they let Project Managers take over the control of the design process)! Add significant cantilevers and there is no chance. And then we have Transfer Beams!

I would think that currently most buildings are non-compliant, but the Robustness clause is currently in the AS3600 and AS1170.0. Presumably no one reads it at the moment.