Glass Curtain Wall Drift/Racking

[EngStuff]

I seen a lot of threads on here about glass curtain wall and deflection. I also have Design Guide 3. I understand that it is recommended to use AAMA TIR-A11 of L/175 up to 13.5’ and L/240 + ¼” above 13.5 feet. From my understanding, that is the max center deflection of the curtain wall?
I do not see anything about drift perpendicular to curtain wall or racking(In plane drift).

My current situation is I have a store front glass curtain wall which is 28 feet high, and no girt system. So it spans from foundation to top of building. I have a steel building that will likely be seismic category B, wind will probably control. Would I use H/60 to H/100 for 10 year wind for drift perpendicular to curtain wall? What about racking (in plane drift) H/500?

Thanks!

 

[jjl317]

Some comments -

My understanding is that the AAMA CW deflection limits are for the curtain wall framing members, perpendicular to the face of system, in part based on the ability of the glass and typical connections to maintain air/water tightness. Additionally, there would likely be other "local" (based on the size of the glass units) deflection limits, depending on the connection system between the framing and the glass.

In terms of "drift perpendicular to the curtain wall", do you mean the horizontal translation of the upper anchorage relative to the lower anchorage point? If so, the ability of the system to accommodate these movements will in large part be dependent on the anchor type.

And in terms of racking in the plane of the glass, the situation can be unclear. In my experience, it is difficult if not impossible to accurately address the response of a curtain wall system under racking via analysis. In fact, in more custom unitized curtain wall applications, the performance under racking loads are typically tested during a mock-up phase, to better understand the performance.

Finally, a 28' clear span is very large for standard systems. Wind loads girts might be beneficial.

 

[NorthCivil]

28' clear span is very large. check thermal movements and capacity for vertical deflection

See

https://aees.org.au/wp-content/uploads/2013/11/Sivanerupan-et-al.pdf

page 7 for calculations to bouwkamp. check your racking capacity of the lites within the frame, based on the building movements

might have to liaise with the building structural engineer for particular building movements at your area of curtainwall

 

[EngStuff]

JJL317

Thanks for the response, Yes I meant the horizontal translation between the upper and lower anchorage point. Is there a typical value that we can use? also with racking?

NorthCivil

Thank you, the article was informative. To be clear, I'm not designing the curtain wall system. I'm the building structural engineer, and trying to design my lateral system to meet the requirements. I emailed and called the manufactures A/E services with no response. I don't want to design an extremely conservatives structure if not necessary, but I have a whole face and sides that are to be glass curtain walls. Their documents only give center deflection of L/240 + 1/4" above 13.5 feet. that's why I am looking for the horizontal translation between upper and lower anchorage points and racking?


Owner/architects are 100 percent against girts. but the curtain wall manufactures spec sheet shows it can handle up to 30'-0

 

[NorthCivil]

EngStuff

appreciate your concern. When i design curtainwall, it is a struggle to get the building structural engineer to provide drifts. usually they give me code maximum.

if you design your building to meet code maximum drifts (2.5%? depending on your location?)

likely the curtainwall engineer will be able to accomodate those movements, as they are standard.

 

[jimmytwotimes]

Nice to see a building engineer thinking about the curtain wall guys!

Most curtain wall systems are tested to H/100 for serviceability and H/66 for strength (standard non-project specific test). Typically the test is done with (2) rows of panels, each 13'-6" tall, and racked at service level to 1 5/8". The overload test is done based on 1.5 * 1 5/8" = 2.43" (H/66)

The serviceability test just means the wall gets racked at H/100 and then needs to pass an air/water test. For the strength test (the "overload" test) the pass criteria is basically just nothing falls off the wall.

Then verification comes from comparison of wind drift w/ service level test, and seismic drift w/ strength level test. If I was asked what wind drift I would want the building limited to it would be H/400. That seems pretty standard for most structural engineers that actually think about the facade.

10 year vs. 50 year wind drift values seems to be a point of debate if it even gets questioned. But in my opinion, is something you need to be clear with to the architect/design team. The curtain wall guy is highly unlikely to ask. We just want to see a clearly defined line in the spec that tells us what we are expected to meet. Because 9/10 times we RFI this and the response is "design for code minimum"...