if my memory isnt messing w/ me, the Ghery exhibition at the MoCA here in LA (i doubt its still there) had a few sheets of the CDs.

They get each structural member out of CATIA straight to the steel fabrication shop.

curtclay - there is a publication out on the making of the Stata that was put forth by MIT Press.
_Building Stata_ by Nancy E. Joyce. Photos by Richard M Sobol.
I'll post it on the books section when I get a chance.
It's on Amazon.
It's not quite what you're after but it's something. They also used 3d models to build Stata, not just paper. I recall seeing the CD's at the NY Guggenheim show.

As Brim already mentioned, Gehry's office essentially does their own shop drawings, which gives them a lot of control over the design, but the liability must increase exponentially. The bigger question for me is the guys actually doing the construction. Does he tend to use the same contractor? I can't imagine you'd hard bid one of his jobs.

I worked with a guy who had a friend working for one of the more well known LA studios(don't remember which on) who said they had parties when the completed a set of CD's, not a beer in the breakroom but real parties where you wake up on your boss's lawn & shit.

fear is only for those who are ignorant.

i led a panel discussion with a fellow from gerhy technologies a year ago. you get bigger fees for stepping over the line and if you know what you are doing, you do it right, there is no risk. move past the fear of the 80s man. this building is shit but we should all be glad gerhy is paving the path of 3d working drawings CAD --- to CAM. hey, you all are doing nurbs modeling any way, arent u? CATIA, is a big step up from the poxy viz/max/rhino/maya enviroments, but even if you may not like his stuff, we should recognize he is a ground breaker.

The Guggenhiem NY show did have a few sets of CDs -- freakin crazy it was like a phone book for just window casings. Every single window was different and completely detailed in 3d and 2d.

I read somewhere that multiple contractors have quit Ghery jobs because they couldn't read the CDs.

I liked the decon type Ghery stuff before he slid into wavy titaniam ad infinitum, but as TED mentioned you have to give him ups for pushing contractors as far as you can. . .

Gehry's design process has some/all the following elements/processes:
1. Volumetric analysis of brief
1a. Gehry sketch
2. Assembly and arrangement of volumes on site model
3. development and actual model built of concept
4. 3d Point scan of model/s into computer into either Rhino or Catia
5. Sections cut through Rhino model and then rationalised
6. Sections exported into Catia, design development model commenced, documentation commence.

Rhino is used because of its ease of transportabilty between Catia and other programs. Dont forget that CATIA is/can be controlled to a large extent by a related excel spreadsheet, and Rhino can export stable CSV format files (effectively x,y,z locations for points) that can be linked to a CATIA file. Of course alot of programs can do this. So the windows can effectively be controlled by a spreadsheet.

Also, Gehry has the clout to effectively force construction company's and subcontractors to use CATIA and submit to their own quality and programmatic demands. You cant work with Gehry unless you adopt Catia basically. There is a strong relationship with Permasteelisa. Gehry also has a strong internal team set up (for a number of years now - I dont know if they are the same guys froming the basis of Gehry Technologies) to exclusively learn, develop and set-up IT and software practices. These guys basically know Catia (and geometry) inside out.

I spent a very brief time at Gehry in 2002 - not as an employee. I dont know much about the processes of documentation, but I know that at Disney for example, they had a site office with at least 5 employees and at least 2 Catia machines still documenting as the building went up.

Confluences: The Design and Realization of Frank Gehry's Walt Disney Concert Hall
great book just came out, showing the range of the process, including the CATIA work, the math behind the wholw thing, and the construction includes an interactive cd-rom

Also;http://www.gehrytechnologies.com/

I've never felt less informed in my life. I understand the process outlined above, but I can't even begin to imagine the coordinating necessary to make it all work.

Good god, I feel old.

When Gerhy had his MOMA show there were many CD sets around to look through. I spent about 30min with one set trying to figure where to start (datums, 3d coordinate reference points…) but I did not see how referencing, sizing and locating the organic, sculptural forms was done. It struck me that if I were a contractor I would need the computer model on site to be figuring things out.

Our office built the home of one of the contractors (masonry) for the MIT building. Next time I run into him I’ll have to ask him how he was able to both bid the job and construct it.

This aspect of Gehry's work (though the execution of it is to be attributed to others of course) is the most interesting thing about the office. As far as new methods of going from design to completed building, and new ways to conceive of Construction Docs, they are right up (and perhaps past) the forefront of architectural practice. I am hoping their research will trickle down to the rest of us in due time. Though no one is going to put up the cash to teach me Catia at work anytime soon.

Many props to the crew at Gehry Tech. They are kicking ass over there.

Metamechanic,

I understand the theory you are talking about and it is truly very seductive. But… the reality of construction is very different – it is all about tolerances and field dimensions and variances of an inch or so happen all the time.

In Gehry’s case, I would not be able to sleep at night as the project architect thinking about how the hell all those points that my CATIA modeled shop drawings that are specifying my steel are going to really meet once the framework starts going up… yikes!

Someone mentioned having a civil crew on site to plot actual construction points. That makes sense to me but imagine what that adds to both constructions costs (construction delays) and a/e fees!

The cad to fabrication route does work in smaller cases (industrial designers are doing it all the time) but there are so many variables at the scale of a building that meeting the perfection (and 1/32” tolerances) of a computer model are just not reasonable.

... I know people at companies that are afraid to make the jump to Architectural Desktop yet alone to something like Catia. The software would have to be incredibly efficient or these buildings wouldn't be going up as quick as they are. The majority of architecture firms aren't creating buildings made of forms that require complex geometrical calculations. So many will be hardpressed to believe they need to adopt the method of creating documents that Catia provides because they "don't do that kind of architecture". But it obviously has the ability to turn a coordination nightmare into a breeze.

I'll check for the Confluences book. Thanks for the tip.

~C

I partially agree with you that contractors are going to have to move up quite a few levels to produce this kind of work but even then this methodology must be so expensive.

Gehry can do it because he is a brand. You want a Gehry, you know you have to pay. But for the rest of us…

I have been an architect for around 8 years now, I worked as a framer and I build furniture. I can tell you that (for me at least) the most successful thing when I build furniture is not to cut each piece per my cad drawing but to first construct the governing framework and then field scribe (no measuring involved) all the of secondary parts to the frame. I get the tightest joints and fewest errors.

This same kind of thing happens in making buildings – someone else who does construction contract admin back me up on this one… Beams are ordered longer for field scribes and as-built dimensions are sent to fabricators to work from – not your architectural plan dimensions. It is common practice and pretty straightforward with nost buildings.

What I am saying is that I don’t understand how you can 4-axis CNC a set of beams toleranced to 1/32” and not have room in there for field deviations. Ask your structural engineer how much those beams are going to deflect under just the dead load once your building is going up – you would be surprised! Is all that deflection calculated for in the cad model and built to? Getting that stuff to meet perfectly…??? Again, someone mentioned having a survey crew in the field shooting points and doing SK drawings on the fly – this makes sense to me but seems like it would drive the costs up astronomically.

As to my statement you both mis-read and miss-quoted me. What I wrote was “cad to fabrication route **does** work in smaller cases.” The world is changing and I have no problem with that but I am not a technocrat either. I would personally love it if someday I can create a building in 3d wearing my VR goggles and feedback gloves to literally move walls and beams around then send it out for fabrication knowing it will all go together like my childhood Lincoln Logs. Think about the reduction in liability (errors and omissions) alone!

one of my instructors brought in a CD set for the disney concert hall when i was in undergrad. it was easily four or five inches thick. i mean, i was just in awe. this was also only 1 in a 4 part set mind you.

I just posted 2 snapshots of Disney hall in "under construction gallery."
I've seen this building go up. It is nothing more than conventional steel framing (a lots of it). all the undulating forms attached to this rather rectangular frame. Kind of how they build New York street scene in Hollywood film lots. There is no innovation whatsoever behind those so called sculptural forms.

except that the steel contractor had to use lasers guided by the computer model, co-oridinated in three dimensions to figure out where to put all that conventional steel constuction...

I think some of you are getting lost in tech-speak, when in reality most of Gehry's buildings have been construced by super advanced space alien steel masons that swoop down at night and sculpt each piece on by one The humans that arrive the next day to do their work are drugged into submission through invisible chemical clouds the space aliens have left behind.
the humans do no work. they just stand, confused, but trying to look busy.

that is all.

I agree to some extent with OA. I toured Disney in April 2002 and saw alot of the structure. An important thing to remember that hardly any of the iconic surfaces are structural or watertight.

Usually what happens is that the interior forms are constructed conventionally, and then sheeted in ply (or whatever) and then sprayed with concrete to achieve watertightness. The exterior cladding is then fixed to a custom steel frame attached to this conrete layer. This is why his buildings, particularly Bilbao, have rather conventional interior spaces. There is alot of drainage and services going on under the skin.

The breakthrough with Gehry is the design development phase, the guarantees of tolerance and fit acheived through this, and the flowthrough of this quality through to fabricators and constructors.

As the the tolerance issue, Gehry's design for Chicago's Millenium Park suffered quite a bit from this. The panel supports were shop fabricated based on the digital model and when they arrived on site 489 of the 2064 units had to be remade or "field engineered" in order to work. So it's not a perfect process.

ENR Article

metamechanic,

I don't think the conversation is at ALL about eliminating the SD and DD phase of a project! I think the discussion is centered around simply recognizing that when you look up at the ceiling inside of a Gehry building, if you have been out in the field, produced CD's or been a part of any phase of construction for that matter, you HAVE to recognize that it was not easy to accomplish.

This then moves the conversation into also recognizing that the only way to accomplish forms, planes and the ability to create spaces such as this can ONLY be accomplished by embracing 3D modeling techniques and working with contractors who embrace them as well.

daver
05/27/04 7:07
When Gerhy had his MOMA show there were many CD sets around to look through. I spent about 30min with one set trying to figure where to start (datums, 3d coordinate reference points…) but I did not see how referencing, sizing and locating the organic, sculptural forms was done. It struck me that if I were a contractor I would need the computer model on site to be figuring things out

//really, scalable 3-d (CD) plans might be best. i assume that does not yet exist.

e909,

The drawing sets at the MOMA exhibition (and there were quite a few projects with CD sets to view) were pretty fantastic (and pretty thick). They also had a live cam set up covering MIT (the foundation was underway) with a time lapse from start to the current day that must have looked pretty amazing by the completion.

Has anyone out there worked for Gehry and could you comment on the process from SD through CD and construction? I really would like to hear how it works (and works out) from first hand experience.

metamechanic,

no apologies necessary. I would simply reframe the problem you present. Rather than looking at the issue as "a lack of most architect's visual skills" I would say the issue is more "finding new ways to construct buildings". In order to understand this methodology, there is a certain visual skill set that must be embraced by those who do not currently have it.

With fees getting tighter and tighter, projects are moving quicker and quicker so more often than not, a project goes from quick rendering done by some guy who doesn't know the building program to CD's. I can't count on my hand the number of times I've heard people say, "hey, lets look at the rendering and see what it shows." that's the extent of 3D visualization in many cases. This Catia / 4D thing is something completely new. I'd like to embrace it, just trying to figure out how...

The technology is not there yet, simple as that.
You can use these programs to fabricate directly but there are tolerances that can not be met.
this is not a reason not to do something however, it will get better.
I know CE's that have to go out to sites to fix these problems, it is work for them at least.
This process will only get better.
Contractors will get more experience with the process as well.
until then the CE's get to make money.
There are changing being made, it is underway it will just take a long time to become perfect.

metmechanic,

I see. What came to mind when you said 3D and 4D documents was an office full of people with oversized dark goggles on and gloves with wires hooked up to them. ... like Tom Cruise in Minority Report.

I like the idea and think it's great, but that shifts the conversation into... what client is paying architects to spend the time to do 3D CD sets? We barely have time to complete them now!

I interned at a firm in Boston that for some reason that I never found out, had a set of CD drawings (not sure if it was a complete set or not, as the firm was not involved in the project) from the MIT Stata Center stored in the back room with all the other rolls of old drawings. I'm not a fan of Gehry, so I never tried to make copies of the drawings or anything like that, however I was quite surprised to find them, to say the least. The best bet to finding a set for something would be to contact the local firm that worked with Gehry on a project. I'm sure they'd be a lot more accomidating than Gehry's office as far as coming in and letting you have a look at what they have in the flat file for the project. If you give them a call, just be sure to stroke their ego for being able to work on such an "amazing project", etc. etc....