New Brunswick Structural Engineer Trouble 4

[kissymoose]

https://www.cbc.ca/news/canada/new-...-engineer-buildings-john-richardson-1.7244456

Summary:
A structural engineer was called in to assess foundation cracks in a 63 unit apartment under construction back in December and concluded there were deficiencies in the design.
He calls the EOR and is informed they have produced many of similar design, then calls the New Brunswick association and hands it over to them.
Since then the engineer's licensed has been suspended/rescinded from multiple associations and multiple lawsuits have been filed.
The city of Dieppe said it has identified 20 buildings associated with the engineer, but who knows what the scope of work is to identify any other deficiencies.

"Match admits to under design of the transfer slab and certain elements, including columns and supports, supporting and connected to the slab..."
"Richardson described the main issue he observed as a concrete transfer slab over the parking garage that's too thin, with too little supporting steel."

 

[Tomfh]

Well, 14.06/13.81 = 1.018, I suppose there's a truncation in the guts of their code, so when it chops it off to 1 decimal it's a 1.0 and passes.

The software states "1.02", meaning it rounded 1.018 up to 1.02, as opposed to rounding it down to 1.0.

Some people and some software consider 1.02 or 1.03 a passing value. This isn't a case of the computer failing to recognise that 1.018 exceeds 1. Rather, the PASS/FAIL threshold is set slightly above 1.

 

[lexpatrie]

The display logic does not need to match the internal logic of "pass/fail". That's what you're misconstruing. This is a printout, not the program code, so you can't tell what it's doing behind the curtain.

 

[Just Some Nerd]

Some people and some software consider 1.02 or 1.03 a passing value.

Imo some people and some software need to re-assess what they're doing....

----------------------------------------------------------------------

Why yes, I do in fact have no idea what I'm talking about

 

[Tomfh]

The display logic does not need to match the internal logic of "pass/fail". That's what you're misconstruing.

The computer is capable of distinguishing between 1.0, 1.018, and 1.02, as is clear in the display output. The software is simply calibrated with the PASS/FAIL bar set slightly above 1, as some people are known to do.

Imo some people and some software need to re-assess what they're doing....

I don't agree with it either, but it's what some people do.

 

[EngDM]

Well, 14.06/13.81 = 1.018, I suppose there's a truncation in the guts of their code, so when it chops it off to 1 decimal it's a 1.0 and passes.

It's not 1.0, it is 102%. The 1.0 is a Canadian code reduction for load duration factor and it just displays it so you know it is used. For instance, with high-ish dead loads on wood members that have snow and live loads, the LDF can go as low as 0.7 so your capacity is 0.7*factored capacity.

 

[lexpatrie]

You both are getting confused here, I'm saying it's a "pass" because the logic on the pass/fail is separate from the capacity calculation, for whatever reason, INSIDE the software. I'm referring to the "PASSED (102%)" - the pass fail that displays "pass" isn't using the calculation to the right of it (102%, It's using some other internal logic.... it's a bug in other words due to sloppy coding).

i.e.

=if(rounddown(stress_ratio,1)<=1.0,"PASS (","FAIL (")&(stress_ratio)&")"

EngDM - it's a coding error, report it to Forte.

 

[Tomfh]

I'm not confused; I simply disagree with your claim that the only way software could PASS a value of 1.02 is because of a “bug” in the software. Many engineers and various pieces of software intentionally pass values that slightly exceed 1.0. It’s no accident.

It’s relatively common practice, and whilst questionable, is an entirely different scenario to a structure that’s overstressed 50%+, 100%+, 200%+, etc. For the whistleblower engineer to have hit the panic button the level of overstress must have been in that sort of range, as opposed to 2-3% over.

 

[lexpatrie]

I doubt the fellow in New Brunswick is upset about 102% I'll grant that.

I think it's a bug because I've never had Forte do what the output shows, pass with over 100% on the check. Can't really share Forte data files anymore with the web version.

 

[ChorasDen]

Forte has a pass limit where it returns anything with a DtC ratio <=2% over capacity as passing, for both strength and serviceability checks. This is an intentional feature in the software and is not a bug. see the note and link below for more info:

[URL unfurl="true"]https://www.forteweb.com/Help/Content/C_Tabs/11_report_tab.htm?Highlight=passed[/url]

 

[jerseyshore]

Yeah I was going to say I see this in Forte all the time. especially, when checking existing framing. There are plenty of times you'll see 100-102% as PASS in green. But I make sure any of my calcs or designs are passed at 99% or [well] below.

 

[skeletron]

One issue that I have mixed feelings is Richardson not making a formal complaint. I mean, I get it because that would indefinitely involve him in the complaint process that could extend for years. But, if Richardson had reviewed the building, observed on site, talked with Match, why didn't he go through the next step of submitting a complaint for that one building and listing concern that this could be a repeated issue.

On the other hand, it's a bit of a cop-out for the Association to say:

the association's CEO and registrar, Lia Daborn, said the lack of a formal complaint by Richardson "hampered our efforts."

Like, give me a break?! Efforts were hampered by not having a more robust system of checks and balances in place. So, in that respect, I understand that Richardson turned down the opportunity to review other buildings as part of the investigation.

 

[lexpatrie]

ChorasDen, thanks for clarifying. I get it now. It's not a "bug" it's a "feature."

I'll be extraordinarily candid for a moment.

I don't know, but as to there being a lack of a formal complaint, most of the U.S jurisdictions I know would refuse to investigate. It's actually nice that they've exceeded the "scope" of whatever complaint they actually got (informally) to look for similar structures by the same engineer. Now that's Health, Welfare, and Life-Safety being taken seriously.

(side note, last I heard a few states still lack the authority to investigate "unlicensed architects" as their explicit authority is over licensed architects....

I really think in the U.S. "blind" complaints from non-engineers get taken more seriously to a far greater degree than "I'm an engineer and this is done wrong." The sheer existence of this opinion, inside me, is a problem for the U.S. boards. The States all seem particularly uninterested in these kinds of issues, ("cat fights between engineers", if you would, considering things differences of opinion versus actual errors....) and I have some actual basis for this opinion, it's not arrived at randomly. (e.g. "he withdrew from the jurisdiction, so it's all good, we aren't investigating, thanks for calling." Meaning don't bother sending anything in formally, because we won't investigate anyway. Thematically I think there's a lot in common with what's going on here, i.e. if you don't know what calculations you're supposed to be doing, you won't do them on repeat projects, and without a "senior" looking at your work or a peer review, or formal training, the calculations you do that are wrong you will continue to do wrong, whether there are flaws in your homebrew spreadsheet, or you are setting up calculations incorrectly in off-the-shelf software.

Although in the "case" I was tangentially involved in, it appears the individual had catastrophically messed up the design of something they'd never done before (which goes to competence, (Buildings are not coffer dam repairs), i.e. Harbour Cay, a first-time design by some NASA folks who bought the ACI 318 and started a design project, and which collapsed during construction due to a combination of design errors and an aggressive pour schedule, lack of reshores, and shoring deficiencies, weight of wet concrete), so there aren't multiple buildings YET, but there's nothing preventing them from doing another building with the exact same problems, it just won't repeat itself in the jurisdiction, so it's all FINE.

Not even a formal reprimand or a warning, they withdrew their license so "we're all good." But they do like to send off their $1,000 fines for not renewing on time or without the appropriate number of continuing education units, which are somewhat subject to their discretion as to which ones are acceptable.)

This case it seems rather they did multiple buildings incorrectly enough that they limped through construction and haven't yet seen their design loads (sounds like the situation here).

I think it's almost better not to lead them to a specific flaw, here in the U.S. if you think something is deficient, you can call it in, the burden of proof isn't on a member of the public with a concern. Anyway, with an engineering report, if you hand the board the one flaw you found, and some commentary on other things that should be checked but you didn't perform (you don't OWE them a full structural analysis, the public doesn't), they may not necessarily bother looking for any other flaws, and if they don't look, they won't find flaws you didn't find, i.e. independent review.

SO, for Harbour Cay, bad depth of reinforcement, slabs too thin and deflection checks not in the calculation package, no punching shear calculations, column reinforcement beyond the area percentage limit, .... These calculations at one point, existed, and now seem to have vanished into the garages and personal effects of the investigators who are now retired and in poor health, (or are in some warehouse next to the Arc of the Covenant). I know I'm going to hear from others that you can calculate whatever you please, if the contractor deviates from the plans you'll have a difficult time knowing, but that's a false argument. That's on the contractor. If the design actually works on paper it's a lot more likely to work during construction if the contractor is sufficiently skilled and careful. Many, MANY of the OSHA collapse reports involve the contractor asking the engineer about issues and getting a formal or informal brush off. (FIU springs to mind, particularly).

I can only hope that these buildings and their calculations eventually become public record so the engineering community can learn from the errors that were made, if there were any. And further, that any future engineering articles on the flaws are written by people who actually have the calculations and aren't basing them off the say-so of the parenthetical comments from an OSHA or NIST report.

 

[hokie66]

A lot of discussion here that the calculations were wrong. But what says there were any calculations to speak of? There was a design, as these buildings did get constructed, but we don't even know if the designs were done with the benefit of analysis.

 

[lexpatrie]

Yeh, granted, Florida has a few disciplinary actions where there were no calculations done, but I think the issue here (for others) is the fellow alerting the authorities may (or may not) have done their own calculations

 

[hokie66]

He is an engineer with duty to the public. That duty does not extend to doing his own calculations. Many times, it does not take calculations to exhibit failure. I have done similar notifications several times over the years, when a failure was obvious visually.

 

[Tomfh]

Calculations are somewhat redundant when a building is in a state of failure.

 

[lexpatrie]

I think you mean irrelevant, Tomfh.