http://www.bbc.co.uk/news/magazine-27509559
number 7 is interesting. I would have been completely paranoid about level and line from each side, would it be so hard to check before constructing?
My greatest miscalculation was when I advised my mother-in-law not to buy Google at the IPO. I still hear about that when I see her.
Good examples of why folks that aren't accustomed to metric should stick with what they know.
I layed out an Olympic Trials Soccer field in Edmond, Ok. in 1983 for construction and dirtwork. One week before the games the Olympic committee sent the fellas with the roll-up tapes to "verify" the size. It was 20m too short.
All the finger pointing was at guess who? Ha.
Olympic soccer fields have a 10m "end zone" area. The architect that designed the field had for some reason misread the overall length. A little scrambling with some extra fill and everything was copasetic for the games.
"Stick to what you know!" :snarky:
Pretty obvious how 7 happened. Have had this before on sites where they changed the datum and we were give a conversion factor eg 0.27m. Of course there are potential mix ups whether you added or subtracted converting one way or another, so you can potentially end up with double the error. These errors can occur on site or in the design office so everyone has a chance to mess it up.
If it is really necessary to use a new datum then the answer is generally common sense to use a completely false datum ie add 100m so you can tell the difference.
Of course yes, the engineers/surveyors who did not bother to sight and check across the water should hang their head in shame. Especially on a project of this expense, you would have thought it would focus the mind. I bet they were obsessing about mm as well on the decks as they were constructing them.
The thing that surprises surveyors from countries with professional licensed surveyors is that we in Europe (well speaking for UK at least) send the young engineers out to do the setting-out while the experienced staff sit in meetings. The formal training in surveying these youngsters have had is patchy.
Having said that whilst working abroad I have seen experienced surveyors make simple levelling errors (not moving level and closing back on original BM) and cause $10000's remedial work. So what was their excuse.
As for the original story about the platforms, I bet there were plenty of people aware of the platform edge issue but they weren't listened to as the management were focusing on something other BS. When I'm in a situation like that I tend to vote with my feet - so I'm rarely around to say I told you so.
I can relate. We did a metric roadway project in the 90's. The Party Chief measured the sanitary and storm structure inverts with an Imperial rod, so the measure downs were subtracted from the elevation of the top of the manholes. It was very embarrassing when the Client called and asked how we got 50 foot deep invert elevations. A new QC/QA plan was quickly put in place. At least, it was caught before construction.
awesome. thanks for sharing that.
If the survey crew was told that "THIS" end of the bridge needs to be lowered 0.27m (subtract 0.27m from the plan elevations) to make it fit, they may have just subtracted 0.27m from the benchmark elevation instead. "Yea, that will be much easier."
OOPS
James
I think they struggled to get 10 good ones, but still worth reading. Thanks
similar to #10 is this: http://en.wikipedia.org/wiki/Hyatt_Regency_walkway_collapse
> I think they struggled to get 10 good ones, but still worth reading. Thanks
>
> similar to #10 is this: http://en.wikipedia.org/wiki/Hyatt_Regency_walkway_collapsebr >
sort of... but the problem at the hyatt was just simple statics. The steelwork and connections just weren't strong enough. It was mentioned in one of the chapters in this book I read many years ago. Really makes you think about things in a a different way. http://www.amazon.co.uk/Design-Paradigms-Histories-Judgment-Engineering/dp/0521466490
The millenium bridge was altogether more complicated dynamics. Marching soldiers have known about it for a while though - they have to "break stride" (go out of phase) when going over bridges. Civilans quite reasonably don't know to do that I suppose. I don't think the millenium bridge was ever going to fall down though, it was just unpleasant for the pedestrians so they fixed it (pretty quickly actually).
In the 1970s, at a leading computer manufacturer, the engineering types who designed the horizontal layout of integrated circuits (ICs) designed in thousandths of an inch, and their equipment was calibrated in the same units. The physics types who designed the vertical characteristics of the ICs worked in Angstroms. A program was introduced to change to metric, but keep using the existing horizontal equipment. This was done by treating an inch to be exactly 25 mm long; it was called "pseudometric". By the time I started working there, they were using real metric and "pseudometric" was considered obscene language. No one was willing to say exactly what went wrong.
true about the youngsters.
setting out is seen as a job you do before going to sub agent then agent.
The first one I could think of I figured sure to not be on the list. Turned out being #1. But I thought that happened back in the mid-early 80s.
Almost had one of those mis-calcs of my own once. Me and my Bud, Curtis, were building a workshop for a retired surveyor turned inventor, Tony. One of us forgot our tape one day so Tony lent us one of his. We were trying to figure out why we could get our foundation square using 3 tapes like both of us have done in the past many time each. I noticed one of the tapes wasn't marked the same. "Hey Tony, what's up with the markings on this tape?". Neither of us had ever heard of a tape marked in tenths!!! Funny, he couldn't understand how we could make a square foundation using 3 tapes. I guess he knew he'd given us a tape in tenths. When he got us a "proper carpenter's" tape, the foundation "magically" became square and he understood. Afterwards we measured diagonals to check our work before we started on the floor system. We were both proud to be 1/16" out on a 16x18' building. Tony was impressed as well.
The Regency walkway collapse that Peter mentioned has been featured on the "Engineering Disasters" show. I've seen it a couple times.
Same with the crane collapse at Milwaukee Stadium. The heavy lift along with the high winds were to blame.
In both cases, I would call those bad judgments. Not necessarily mis-calcs. I don't recall the exact details of walkway collapse so don't hold me to that one.
> In the 1970s, at a leading computer manufacturer, the engineering types who designed the horizontal layout of integrated circuits (ICs) designed in thousandths of an inch, and their equipment was calibrated in the same units. The physics types who designed the vertical characteristics of the ICs worked in Angstroms. A program was introduced to change to metric, but keep using the existing horizontal equipment. This was done by treating an inch to be exactly 25 mm long; it was called "pseudometric". By the time I started working there, they were using real metric and "pseudometric" was considered obscene language. No one was willing to say exactly what went wrong.
Not to beat a dead-horse but was that inch from international feet or survey feet? The difference when going to Å could be significant. Based on my cals, the difference is 508.00101599097252 Å. I would imagine the EEs and the physicists are going to complain about that.
Regardless, 25mm for 1" is neither an international inch nor a survey inch.
It would seem they dropped the 0.4 mm from either unit which equates to 4,000,000 Å.
By the way, I wrote a little program to get those ridiculous number of decimal places. 🙂
First one that I thought of was the Tacoma Narrows Bridge; surprised that didn't make it on there but I guess that was a little more than just a miscalculation.
Yeah, Lee, I thought of that as well.
I'm not sure that was a mis-calc rather than a lack of anticipation of windages and/or model mock up of those effects.
No one ever mentioned survey feet or international inches. I believe that when mils (thousanths of an inch) were in their heyday, the masks needed to manufacture ICs were made with a machine that was sort of a marriage of a phototypesetter and a plotter. It would have been designed by mechanical engineers, not surveyors or architects, so presumably would have been designed in international inches.
In the 70s I don't think you would have found many ICs bigger than 2 mm on a side. An error of 2 parts per million in the definition of an inch would have resulted in a maximum error of 6 nm which would have been negligible in those days.
It just wasn't really anticipated that a bridge would have a resonant frequency.
Your first bridge is a lump of rock and over thousands of years, gradually your bridge designs become more and more slender - inevitably the failure modes change.
For example we got caught out in the UK with impact collisions to the piers (columns) of bridges. A lot of hasty assessment and remediation has been done over the last 20 or so years.
The hyatt problem was due to a design change during construction not being properly considered. The original design wasn't that great either. The whole thing was relatively simple, but not apparent at the time as everyone was thinking about different things - like putting up a hotel in the shortest time and getting a pat on the back (or $$$$).
Lake Peigneur
>The steelwork and connections just weren't strong enough.
There were two mistakes. The designer specified something strong enough but impractical to build. The construction people substituted something that looked the same to them but was not. The drawing put the tension on long rods. The as-built put the tension on too-thin cross members with short rods interconnecting them.
Lake Peigneur was the first one to cross my mind... there's a film on YouTube indicating that it was an 'engineering' error... OK...
