Opal Tower - Sydney Australia 28

[CivilEngAus]

https://www.news.com.au/national/ns...s/news-story/dad28af4aad5ff11cf968102635824f9

https://www.smh.com.au/national/nsw...ns-building-may-collapse-20181224-p50o4b.html

This could be an interesting and developing story in Sydney Australia. A 34 storey near new residential apartment tower in Sydney has been evacuated this afternoon over fears it is in structural distress with cracking noises heard during the day and one or more cracks developing; emergency services are treating it as a major incident.

Given we already have some of the toughest building codes in the world (although little to no registration requirements for engineers) it will be interesting to see how this plays out and what the crack(s) looks like to cause such a major emergency response.

 

[Ingenuity]

...and who pays/compensates the apartment owners who have now lost usable floor space with the thickened wall/column elements. Going to get ugly!

 

[rapt]

Ingenuity,

Hope you had not bought one!

They will lose a lot more in collapsed resale value than they will from 100-150mm off the side of a room. Add that to Sydney prices falling anyway and most will have debts higher than value forcing higher interest rates and impossible to re-mortgage, especially those who started on interest only loans. And they cannot just hand it back to the bank like they can in many USA states.

Doubt that consultants insurance company will volunteer to cover those losses.

 

[QSIIN]

Even with the changes to the 2018 code, there's no such thing as slenderness or second order effects for walls in Australia, remember? It just doesn't happen.

So consultants, don't forget to make all your columns D/b>4 (whoops, I mean L/tw, thank God for chapter 11) and you're good to go!! Confinement?, ductility?....earthquakes? What's that...?

 

[rapt]

QSIIN,

It is not just AS3600. Most design codes have a simplified wall design method and most are as basic as ours. The AS rules were based on the BS8110 rules when first written.

Even the changes we have made in the 2018 code to try to make them more robust have been heavily criticized in both the industry and some in academia as being far too conservative.

If you think they should be changed in some way, make constructive proposals for change!

 

[degenn]

All walls.

AS3600-2018 gives cases, in 11.7.4, where wall can be designed as columns not requiring restraint of vertical steel according to 10.7.4. The new last paragraph says vertical reinforced is to be restrained to 14.5.4 if over 50MPa. I believe the correct interpretation of this only if restraint reinforcement is required by the paragraphs above. If the column complies with 11.7.4(a) then it is most likely below the dotted line in Figure 10.7.3.1(A) and doesn't need special confinement reinforcement anyway.

If a wall has a stress level less than about 0.2f'c, then it probably does not require vertical bars to be restrained (i.e. no ties though the wall thickness) , for strengths 50MPa to 100MPa.

 

[QSIIN]

RAPT,

Just because another code does something one way, doesn't mean we should assume it's correct and copy. How's the new Canadian beam shear capacity working out by the way?

Anyway, I've already expressed my opinions and recommendations on simplified wall designs in this very thread. Technology and computing power has come along far enough that some of these simplified methods can be done away with. Seriously, are interaction diagrams so difficult to produce?

As degenn states, you don't have to have ties in a column if you meet applicable criteria, and you can go less than 1% too. At least you're still checking for slenderness, cracking and strain compatibility. Let's at least get rid of the ambiguities as to when the simplified method is and isn't applicable.

And hopefully then we can stop seeing slender 200x800 "walls" being built.

 

[rapt]

DGENN

We are looking at the 14.5.4 referral. It obviously applies to all walls being designed for earthquake. And with the new earthquake code, any building using > 50MPa concrete is going to be required to be designed for earthquake anyway as no-one would waste money using > 50MPa concrete on the buildings that escape the minimum earthquake design requirements as the load levels would not be high enough to require it.



QSIIN,

You obviously do not know how hard it is to get something removed from the code or modified when it has been in a design code for 50 years and is liked by designers because of its simplicity.

You are not the first person to point out how simple it is to generate an interaction diagram these days. But for some reason, many on the code committee do not think designers should be forced to do real calculations that require a computer when a simple hand calculation method has been accepted for 50 years and not many have fallen down! And that includes some academics.

It is the 150 * 800's with central mesh and 80MPa concrete that I am trying to stop initially!

 

[hokie66]

150 x 800, mesh, 80 MPa? Where have you seen that, rapt? Whatever the code provisions, that is nuts.

 

[rapt]

Hokie

Been told about them, especially by someone with a lot of experience in precast. He was fighting against it.

Even Opal was 180 * whatever and I understand 80Mpa centrally reinforced (not mesh at the ends at least) with N28's at 100 central at the ends. 2009 code does not say you cannot do it, and many designers these days will do it if their interpretation is that they are not specifically told they cannot!

RE the confinement provisions and DGENN'S case that walls will be under the limits of 10.7.3 for confinement, the problem is that the earthquake design load with mu and Sp factors for Moderately Ductile and Limited Ductile walls mean that the elastic design load that you are basing the need not to supply confinement reinforcement on are not the real loads under earthquake action where post elastic deformations are extreme. Using those rules with a mu = 1 would be ok, but not with mu = 3 and Sp = .67, where the design elastic load is thus reduced by a factor of 4.5 and ductility of the structure is supposed to make sure the structure can survive under the plastic sway after the elastic design load condition is passed.

We have had to make the same distinction for whether walls are in tension or compression. The decision cannot be made on the reduced design load. It has to be on the full earthquake load with mu = 1.

 

[QSIIN]

RAPT,

I totally agree, and I too have seen extremely thin 80MPa similar to what you are describing.
And you're right, I don't know how difficult it is, though I have tried to reach out to the code committee before regarding various matters and have been turned away. I imagine it must be very frustrating to deal with.

Regarding panels greater than 50MPa and wall ties to chapter 14, engineers will still use the logic of "if wind governs, then you can ignore earthquake". And therefore the mentioned requirement won't be applicable. The idea that earthquake and wind are two totally different mechanisms is still not well understood.

 

[rapt]

I do not know who you approached.

If you have any like that which really need looking at, send them to me on my work email and we can sort out a how to approach them. We do not know every stupid interpretation everyone is making! If you have specific clauses you think need improvement send them to me. I cannot guarantee anything but at least they will be looked at.

Agree with your second comment entirely. Unfortunately they are probably correct as far as ultimate strength design calculations go for section strength design. That is why we had earthhquake moved into the main code instead of as an Appendix. Now we just have to convince people that the reduced load to calculate strength for earthquake is not actually the full earthquake load and very special detailing requirements are needed to make it all stand up.

 

[Tomfh]

RAPT,

The code really ought to spell out the earthquake strength load vs the *full* earthquake load more clearly.

And if it’s not going to spell it out, the full load ought to be the starting point.

 

[rapt]

Tomfh,

We are limited on what can be said in the code itself. It is not allowed to be a text book. If we tried to add what you want in the code itself, even if it got through the main committee (we tried this time and failed), it would be rejected by the Standards editors and BCA. The commentary to the 2018 code will go into a lot more detail.

But any engineer designing to the earthquake provisions of the code should know the background or should be able to research the background logic before he starts designing. You do not learn to design by reading a design code.

It is not as if AS3600 is alone on this, the methods used are basically the same as NZS and ACI so there is a lot of information out there.

It is not the job of the design code to teach an engineer show to design.

 

[hokie66]

And therein lies a big problem. A lot of engineers expect codes to be in cookbook form.

 

[Tomfh]

The code needn’t be a cookbook, but a clear statement that the code earthquake loads are a small fraction of the actual earthquake loads would be a useful thing to include.

I see many connections and walls designed based on assumptions that the code load is the load a wall to connection must resist. I’ve done it myself in the past. And I understand this is why some of the critical NZ failures occurred. The connections simply broke under the actual earthquake loads.

 

[rapt]

Tomfh

That was part of what we tried to add and it was watered down to what is in clause 2.1.2.

But when you divide your earthquake load by 4.5 (mu = 3 and Sp = .67), surely you can figure out the rest has to go somewhere. If you are unsure why, you research it.

Commentary to AS1170.4 goes into some detail on it.

Also, in several places in the 2018 code, it has been specified that the checks be done with a mu = 1 no mater what the design mu value, mainly in the Walls chapter.

We missed the one in confinement. It will be added in the next amendment.

 

[human909]

It is not the job of the design code to teach an engineer show to design.

Except that is how are universities have been approaching 'teaching' structural design for quite some time....

But when you divide your earthquake load by 4.5 (mu = 3 and Sp = .67), surely you can figure out the rest has to go somewhere. If you are unsure why, you research it.

But when all you are doing isn't following the code letter by letter you aren't thinking about the why only about crunching the numbers. I must admit I am not immune from this but I try not to fall into such traps.

Just two days ago I had a call from a structural consultancy asking me why my supplied foundation loads had much greater earthquake shear in one direction rather than another. They seem confused in my reply I mention that the period of the structure differed considerable in either direction. They were about to send out a formal project RFI about the discrepancy, but I haven't seen that yet.

The idea that earthquake and wind are two totally different mechanisms is still not well understood.

Very, very true.

 

[degenn]

We missed the one in confinement

Can you elaborate?

 

[rapt]

Human909,

Says something for our universities then doesn't it. Our design codes are already far too complex trying to bridge the gap between text books, design limits and allowed simplified methods.

A design code should simply be a set of limits. Theory is in text books. But design codes then started to define accepted simplifications to make design calculation easier. Some very basic examples are rectangular stress blocks for concrete, Branson's formula for Ieff and kcs factor for long term deflections and span/depth ratios for deflections. Using a slide rule or abacus, these things were too difficult to do on a daily basis so design codes gave acceptable simplifications. This has extended and extended until the line between a design code and a text book became blurred. But a design code will normally not define design theory, simply accepted methods and limits.

All of these simplifications are not accurate and can cause significant error in calculations! Designers have to understand these limitations and allow for them in their design.

Design is not simply plugging numbers into code formulae. RAPT or any other design program does that for people using it, but it is NOT a designer, it is a calculator.

Dgenn
I elaborated a few days ago about 8-10 posts up.

 

[MDEAus]

The final report is out, I haven't had a chance to read it yet.

Final Report