ASCE 7-10 Wind Loads On Roof Mounted Equipment

Evolving Design Wind Forces On Retail RTUs
Wind forces on Roof Mounted Equipment for typical retail and restaurant structures are receiving, at long last, a lot of attention by the engineering community, with somewhat surprising results. Not entirely surprising though, because there has been a growing body of evidence, primarily from post hurricane field surveys, that the standard roof top unit (RTU) anchorage has been insufficient.

1970s
In the old days, we'd apply a wind pressure against the face of an RTU, likely the same used on the windward wall of the building, design overturning anchorage if the weight of the equipment seemed insufficient to hold it down, and call it good.

1990s
The late 1990s saw the first significant change in wind design in the past twenty years, noted in, what are now called, our legacy codes.

2002 ASCE & 2003 IBC, current code in a few states
The 2002 edition of ASCE 7, which became the basis of design by reference for IBC 2003, was the first code reference to specifically address roof mounted equipment as requiring special attention for wind design.

The lateral force applied to rooftop equipment is given by: F = (qz)x(G)x(Cf)x(Af) (lb) where qz is the velocity pressure evaluated at height z of the centroid of area Af using the appropriate exposure category. G is the gust-effect factor, Cf is the force coefficient, and Af is the projected area normal to the wind except where Cf is specified for the actual surface area.

A typical G would be .85, a typical Cf would be 1.3 or 1.4, yielding a product around 1.0 or 1.1 typically. We'll use this product, (G)x(Cf), as the mode for comparison with subsequent codes. All else being equal, design forces can be seen to change in direct proportion to the changes in this product (G)x(Cf).

The ASCE 7 committee was vague about providing guidance for dealing with the possibility of increased loads exerted on the RTUs because at that time they felt there was no basis to make a recommendation. Uplift forces due to wind across RTUs received no consideration in the methodology at all.

2005 ASCE, current code in more than 40 States
The 2005 edition, ASCE 7-05, is the basis of IBC 2006 and IBC 2009 which are the current codes in the vast majority of the US at the time of this writing, late 2012. ASCE 7-05 used the same equation but provided more specific guidance in consideration of the increased loads on RTUs, and specifically for roofs less than 60 feet high, which fits the majority or our restaurants and retail stores. The familiar equation is F = (qz)x(G)x(Cf)x(Af) (lb) but with added requirement that the product of the factors G and Cf for roof mounted equipment shall be adjusted from 1.0 to 1.9 based upon certain geometric factors. But in most cases for typical packaged RTUs, the correct number to be used is (G)x(Cf) = 1.9, which is almost double what it was under ASCE 7-02. The net effect is that design wind loads upon RTUs developed under ASCE 7-05 will be almost double the same loads developed under ASCE 7-02.

Regarding wind induced uplift effects upon RTUs, ASCE 7-05 does not address it in the body of the code, but buried in the commentary pages is this observation, "The designer should design for uplift."

2010 ASCE, The Upcoming Building Code
The 2010 edition, ASCE 7-10, the basis of IBC 2012 and not widely adopted as of this publication, but obviously soon to become law, uses the same equations as its predecessor but adds a specific uplift force requirement. The lateral wind pressure on an RTU is determined from the equation, F = (qz)x(G)x(Cf)x(Af) (lb) This is the same equation with (G)x(Cf) = 1.9, generally, as before.

But then ASCE 7-10 adds that the uplift wind pressure on an RTU shall be considered to act simultaneously with the lateral pressure and shall be determined from the following equation, F = (qz)x(G)x(Cf)x(Ar) (lb) with (G)x(Cf) = 1.5, generally based again upon certain geometric conditions. This uplift force essentially cancels out the "holding down" effect of the RTU weight.

The net result is that, not only will the lateral design wind loads upon RTUs developed under ASCE 7-10 will be almost double the same loads developed under ASCE 7-02, but much of the holding down effect we might consider from the sheer weight of the equipment, has been eliminated from our design. RTU anchorages must increase in capacity yet again.

The Current Florida Building Code, Harbinger of Things To Come?
Florida has taken it a step further with its March 2012 adoption of the FBC Section 1609.8. The design methodology is the same as under ASCE 7-10, requiring (G)x(Cf) of 1.5 for RTU uplift design, but adjusting (G)x(Cf) for RTU lateral design to 3.1. Yes, the word would be "triple".

We're obviously moving in a direction to eliminate the problem of detaching RTUs. And this is a good thing. Buildings that otherwise have performed rather well under hurricane conditions have nevertheless turned in huge insurance claims due to the water damage ensuing upon the consequences of gaping holes in roofs after RTUs have been detached.

Major Texas Retailer Receives Engineering Report From TR2

TR2, The Roof Reinforcer, completed and submitted its final evaluation report for this eastern Texas major retailer today. Tim McCarthy P.E. was consulted to investigate a deflected 20T RTU mounted on the roof. "We performed two site visits altogether," said Mr. McCarthy, principal with The Roof Reinforcer. The first visit yielded measurements of deflections at the top of the RTU and the surrounding roof membrane. The second visit was necessary to measure structural element elevations inaccessible without proper equipment."

The project required two weeks to investigate the structure, analyze the acquired data and develop conclusions.

"The deflection in the roof mounted equipment was obvious," noted Paul, Southern Region Project Manager for TR2. "And there were elevation divergences in the roofing measurements and the structural supporting elements too. But what was unusual was that the RTU measurements and the roof top and structural measurements were telling two different stories. The RTU measurements indicated an unacceptably large deflection to one side, unacceptable from the viewpoint of HVAC performance. The structural measurements corresponded in direction but were much less in magnitude, a magnitude that was well within design standards." The expected solution, an inadequate structural roof system for the support of the HVAC, was not indicated by these measurements. Further investigation was indicated.

Tim McCarthy P.E.: "We performed calculations of the center of gravity of the 20 ton RTU in relation to the supporting structure and found it to be significantly off center, about 50%. We further discovered that the mounting system was significantly smaller than the HVAC base. The combination of these two geometries renders this particular RTU significantly vulnerable to laterally imposed loading conditions." Further investigation revealed that this store was within the current code prescribed hurricane region and likely subject to significant wind gusting.

"This appears to be a classic case study for the significant acceleration in recent codes of wind loads on roof mounted equipment," noted Tim McCarthy. "It's only been since 2000 that industry research and the building codes have directly addressed wind gust loads on roof mounted equipment. And the results have been surprising. Current wind codes mandate design loads frequently double or even triple what many current buildings have been designed for."

The report recommends a contemporary wind analysis be performed for this deflected RTU, that a proper load distributing curb be designed, and anchorages installed in accordance with the wind design.