jonochshorn.com
Selected writings
Follow @jonochshorn
contact

Appeal regarding building code violations in Cornell's Fine Arts Library

Prepared by Jonathan Ochshorn


animation gif showing hypothetical smoke exhaust at roof-top art gallery in Mui Ho Fine Arts Library at Cornell
Smoke exhaust vents for the library atrium form the northern boundary of the roof-top assembly space, as discussed in Exhibit 2 (Violation #5). Animation and underlying photo by Jonathan Ochshorn.

The Capital Region Board of Review Appeal Hearing, open to the public, took place virtually on Wednesday, March 17, 2021. Of the nine "violations" alleged in my complaint, the hearing Board ruled against me on six of them, i.e., they ruled in favor of the City of Ithaca's determinations. For violations 6, 7, and 8—all of which relied on an out-dated 2016 Code variance—the Board voted to modify that 2016 variance decision by reducing Rand Hall's total allowable occupancy from 518 to 387 persons. In a sense, these "modifications" supported my complaint, but, overall, the outcome of the hearing was disappointing.

On April 1, 2019, I filed a Code complaint concerning the Mui Ho Fine Arts Library at Cornell University with the City of Ithaca Building Division under Title 19, which was resolved on May 3, 2019 with a statement from the City of Ithaca Director of Code Enforcement, Michael Niechwiadowicz, stating that the "proposed work is in compliance with applicable codes." Not one of my detailed Code-based objections was discussed in the determination or refuted. On June 10, 2019, I filed a Code appeal (Complaint #4660) with the NYS Division of Building Standards and Codes Oversight Unit, which was "closed with prejudice" on September 26, 2019. Again, not one of my detailed Code-based objections was discussed in the determination or refuted. What follows is my appeal of these determinations to the Syracuse review Board of the NYS Division of Building Standards and Codes, which I filed on September 17, 2020, with an Appendix added on November 23, 2020. The Appendix contains the response of the the architect of record (STV Architects) to my detailed analysis of the library's Code violations as well as my comments on the architect's response.

The library was designed by Austrian architect Wolfgang Tschapeller (with STV of New York the architect of record). Links to all of my writings on the Rand Hall Fine Arts Library can be found on this website. The 2015 New York State Building Code can be found here. I have included links to the original (pdf) appeal documents. A web-based version appears in the table below. This table displays my analysis of nine Code violations in the left column (from Exhibit 2 and 3 of the appeal), while the architect's response—and my comments in rebuttal—appear in the right column (from Appendix 1 of the appeal), along with a timeline.



Original appeal application, exhibits, and appendix ("official"pdf versions):

Appeal application

Exhibit 1—Determinations

Exhibit 2—Reasons to reverse determinations

Exhibit 3—Supporting emails from ICC and DBSC

Exhibit 4—Photos of architectural Code
Analysis drawings

Exhibit 5—Atrium smoke control report

Exhibit 6—Appeal Petition 2013-0250

Exhibit 7—Variance Petition 2013-0456

Exhibit 8—Variance Petition 2015-0432

Exhibit 9—Variance Petition 2016-0269

Appendix 1—Response to My Code Appeal (including timeline)

Appendix 2—Egress diagram

Appendix 3—Comments on revised Atrium Smoke Control Report


Web version (below) of appeal material from Exhibit 2, Exhibit 3, and Appendices 1–3:


Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Introductory comments

Before being connected to Milstein Hall in 2011, Rand Hall was a 3-story Type IIB non-fireproofed steel frame building and E. Sibley Hall was a 3-story Type VB building (with non-fireproofed wood exterior bearing walls at the third floor). Since the two buildings were connected to 2-story Milstein Hall without fire walls (Figure 1), Rand-Milstein-Sibley Hall became a single building with Type VB construction.

Sibley, Milstein, and Rand Halls photo and site plan
Figure 1. (a) View, from left to right, of interconnected Sibley, Milstein, and Rand Hall; and (b) second-floor plan showing net second-floor areas of Milstein, E. Sibley, and Rand Halls.

Prior code variances: In 2012, shortly after the construction of Milstein Hall, Cornell's Fine Arts Library was moved to the third floor of Rand Hall in violation of the New York State Building Code, which prohibits library (A-3) occupancies above the second floor of Type VB construction. After this violation was successfully challenged at a hearing of the Capital Region-Syracuse Board of Review on July 18, 2013 (Petition No. 2013-0250), Cornell applied for and received a code variance later in 2013, which allowed the library to remain on the third floor of the combined Rand-Milstein-Sibley Hall. Two subsequent code variances were granted in 2015 and 2016 based on the submission of preliminary, and evolving, plans for a new Fine Arts Library in Rand Hall, still without a fire wall to separate Rand Hall from Milstein Hall. The code waivers granted by the three variances can be summarized as follows:

  • The 2013 variance (Petition No. 2013-0456) permitted Rand Hall, with a library on the third floor (and potentially on the second floor), to exceed floor area and height limits for A-3 occupancies in Type VB construction.

  • The 2015 variance (Petition No. 2015-0432) allowed 2-hour fire barriers, instead of a 3-hour fire wall, to effectively create a separate "Rand Hall" building with non-fireproofed steel (Type IIB) construction. The library spaces proposed at the time consisted of two stories, each with a mezzanine, so the building—including the first-floor F-1 shop occupancy—was still considered to be three stories high.

  • The 2016 variance (Petition No. 2016-0269) allowed a proposed four-story A-3 occupancy in Type IIB construction, a smaller elevator not sized for an ambulance stretcher, and an alternate to standby power for atrium smoke control. The library space outside the bookstack floors (or "vertical opening" in code language) was reformulated as an "atrium" connecting stories two through four (with a single mezzanine within the fourth story), so the building—including the first-floor F-1 shop occupancy—was now considered to be four stories high. Various structural elements in the building were allowed to have no fireproofing, i.e., to remain as non-fireproofed Type IIB construction instead of being upgraded to Type IIA fireproofed construction as would have been required for an A-3 library occupancy in a four-story building.

Because the current Rand Hall library is substantially different from prior proposals for which these code variances were granted, those variances are no longer in effect. Specifically, the library as built contains, for the first time, a roof-top art gallery that more than doubles the library's occupant load compared to all prior proposals. Furthermore, the current structure is a five-story building (although packaged as a four-story building with a mezzanine—see explanation of Violation #3 below), whereas prior proposals in 2013, 2015, and 2016 were for a three-story or four-story building. Unlike the as-built library, the four-story proposal permitted under the 2016 code variance had a mezzanine that conformed with the Chapter 2 mezzanine definition and with Section 505.2.1 area limits that require a mezzanine to be between the floor and ceiling of any story and within a room or space that is at least twice as big as the mezzanine floor.

New York State Code variances only apply to the specific building proposal for which the variances were granted—and cannot be interpreted to give implied approval to a different proposal. All three Rand Hall variances (Petition Nos. 2013-0456, 2015-0432, and 2016-0269) contain the following written disclaimer: "Furthermore, it should be noted that the decision of the Board is limited to the specific building and application before it, as contained within the petition, and should not be interpreted to give implied approval of any general plans or specifications presented in support of this application." Since the library as built is substantially different from all prior schemes, the code variances that supported those prior schemes cannot be applied to the current building.

Atriums and mezzanines. Even though prior code variances—requested for prior versions of the Rand Hall library and shop—are not germane to the current proposal because the current as-built scheme is substantially different from all prior versions, it may be useful to compare the noncompliant use of what are called "atriums" and "mezzanines" in the current scheme with the use of atriums and mezzanines in prior versions for which variances were requested.

The proposal corresponding to the 2013 variance had neither an atrium nor a mezzanine, since it involved only a change of occupancy in the existing three-story building. However, the two subsequent proposals corresponding to the 2015 and 2016 variances both were substantial alterations of, and additions to, Rand Hall, each of which created four library levels above a first-floor shop occupancy.

In the 2015 scheme, based on the 2010 New York State Building Code, the four library levels were configured as two stories, each with a mezzanine. Because both mezzanine levels were within a double-height story, and because both mezzanine levels were within code area limits for sprinklered buildings, Rand Hall was properly defined as being three-stories in height, with 1 story of F-1 shop occupancy and two stories (each with a mezzanine) of A-3 library occupancy. The four library levels were all spatially connected with a "vertical opening," but since these four levels consisted of only two stories, this vertical opening was in conformance with Section 707.2 exception 7 ("does not connect more than two stories") of the 2010 New York State Building Code, and therefore did not need to have a shaft enclosure or be defined as an atrium.

In the 2016 scheme, based on the 2015 New York State Building Code, the four library levels were configured as three stories, with the top story containing a mezzanine within it. Because the mezzanine level was within the double-height portion of the top story, and also within code area limits for sprinklered buildings, Rand Hall was properly defined as being four-stories in height, with 1 story of F-1 shop occupancy and three stories (the top one with a mezzanine) consisting of A-3 library occupancy. Because there was a vertical opening connecting what were now three library stories, the vertical opening was properly defined as an atrium.

The current proposal, like the prior scheme, has four library levels connected by a vertical opening that is defined as an atrium, as well as an F-1 shop occupying the first story. The scheme differs from the 2016 variance proposal in that it no longer has a double-height fourth story in which a mezzanine can be placed. Rather, the atrium's vertical opening now extends through the roof level, and a small fifth story is placed on the roof, with an exit access stairway connecting this new fifth story through the atrium to the fourth story below.

Relevant building code: The Fine Arts Library proposal counts not just as an "alteration" to Rand Hall, but primarily as an "addition." This is because, per the definition of addition in Chapter 2 of both the 2015 Building Code and Existing Building Code of New York State, it is "an extension or increase in floor area, number of stories, or height of a building or structure." Specifically, the building's height has increased from approximately 43 feet above grade to approximately 55 feet above grade, and the number of stories has increased from 3 stories to either four stories (according to the architect's drawings) or five stories (counting the roof-top bathrooms, elevator, and corridor as a fifth story rather than as a mezzanine within the fourth story). There is virtually nothing remaining of the original Rand Hall structure above the first story; it is true that exterior riveted steel columns are still embedded within the old brick exterior walls, but they have been reinforced with new steel columns that extend from the foundations up to the new roof structure and they are braced by an entirely new lateral-force-resisting system. Most of the existing third-floor and roof-level structural elements have been demolished, and all of the bookstack structural floors and supporting elements are entirely new, as is the main roof structure over the library stacks. Even existing footings in Rand Hall have been reinforced to accommodate increased structural loading. In effect, a new building has been inserted into the brick shell of the old building, leaving only the first story more or less intact.

As an addition, according to Section 1101.1 of the 2015 Existing Building Code, the building proposal "shall comply with the International Codes as adopted for new construction." In fact, the construction documents for this project reference the Building Code for new construction, and not the Existing Building Code, so the question of which code governs the design of this project does not appear to be contested: as an addition, the proposal must conform to the 2015 Building Code (for new construction) of New York State.

Code violations

The Mui Ho Fine Arts Library in Rand Hall at Cornell University is in violation of the following provisions of the 2015 New York State Building Code:

Notes on Appendix 1 comments

The Appendix 1 comments in this column (starting with Violation #1 below) are not intended to replace the detailed enumeration of Code violations contained in Exhibit 2 of my Application for Appeal (left column), which was issued September 17, 2020. Rather, they are intended to rebut the arguments contained in the architect's response, issued November 3, 2020. To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font). The architect's response was signed by Senior Vice President and Chief Architect David Miles Ziskind of STV Architects—henceforth identified as "the architect" or "the architects."

Violation #1: Unenclosed egress stair in the atrium.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code sections: 1006.3; 1019.3; 1022; 1023.2; 1023.1; and 1023.3.

Explanation: "Exit Stair B" is an unenclosed exit stairway that extends from roof-top bathrooms and a roof-top art gallery down to the second-story atrium floor (see Exhibit 4). Such a means of egress component needs to be enclosed. However, the architects have alternatively used two lines of argument to challenge this requirement—one based on requirements for an unenclosed interior exit stairway, and the other based on requirements for an unenclosed exit access stairway. Both arguments are flawed, as was confirmed by code opinions from both the International Code Council and the New York State Division of Building Standards and Codes (see Exhibit 3).

The first argument suggests that Section 1023.2, exception 2, permits an interior exit stairway in an atrium to be constructed without enclosures consisting of fire barriers and/or horizontal assemblies. However, this exception applies only to the "construction" of an interior exit stairway and not to Section 1023.1 (General) or Section 1023.3 (Termination), which require that interior  exit stairways shall "lead directly to the exterior of the building or shall be extended to the exterior of the building with an exit passageway" and "shall terminate at an exit discharge or a public way." James Harding of the New York State Division of Building Standards and Codes, in an email to me dated Feb. 27, 2019 (Exhibit 3), explains why it is incorrect to call this type of stair an "unenclosed interior exit stairway." He writes: "Thus, unless the stairway ends at an exit discharge or a public way, it must end at an enclosed exit element as described in the exception. This guarantees that the level of protection for occupants is not diminished along the exit path, which is required for all exits by Section 1022. There are two exceptions provided in section 1028 to the requirement that the exit discharge be to the exterior of the building, a lobby and a vestibule. Since both of these exceptions are for such areas at the level of exit discharge, they may not be applied to the situation described. A final consideration regarding this question is whether the atrium might also be classified as an enclosed exit element (stairway, ramp or passageway). No.  Section 1022 also provides that an exit, 'shall not be used for any purpose that interferes with its function as a means of egress.' An atrium is provided for ambiance and other purposes as needed and would not be considered exclusively for use for exiting." In fact, this particular atrium is designated by the architects as a library reading room (occupancy group A-3), with an area of 4,928 square feet and, per Table 1004.1.2, an occupant load factor of 50. In other words, it functions as an assembly space with 99 occupants.

The second argument is based on the fact that Section 1019.3 (line item #5) states that an exit access stairway in an atrium need not be "enclosed with a shaft enclosure constructed in accordance with Section 713." However, Section 1006.3 states that "the path of egress travel to an exit shall not pass through more than one adjacent story." Taking both sections together, it is clear that while shaft enclosures for exit access stairways are not required in atriums, such unenclosed exit access stairways cannot pass through more than one adjacent story, even in an atrium. James Harding of the New York State Division of Building Standards and Codes, in an email to me dated Feb. 27, 2019 (Exhibit 3), explains why it is incorrect to designate this type of stair an "exit access stairway." He writes: "The requirements for exiting from all areas of a building are provided in chapter 10.  Any portion of a building that serves as part of the required means of egress system must comply with all applicable provisions of chapter 10. Referring to Section 1006 for exit access, travel through more than one story of the building is prohibited."

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. The east stair is not an unenclosed exit stair within the atrium and an "egress stair" is not a code defined term. It is an unenclosed exit access stair within the atrium.

In my appeal (Exhibit 2, p.5), I have used the term "egress stair" to encompass two types of stairs that may be used as components of means of egress—i.e., interior exit stairs and exit access stairways. I wrote:

Such a means of egress component needs to be enclosed. However, the architects have alternatively used two lines of argument to challenge this requirement—one based on requirements for an unenclosed interior exit stairway, and the other based on requirements for an unenclosed exit access stairway. Both arguments are flawed, as was confirmed by code opinions from both the International Code Council and the New York State Division of Building Standards and Codes (see Exhibit 3).

If the architects are now claiming that this stair is "an unenclosed exit access stair within the atrium," my arguments pertaining to the stair as an "interior exit stair" are no longer pertinent, and only my arguments pertaining to the stair as an "exit access stairway" need be examined. I have already explained, in Exhibit 2, why this particular exit access stairway is noncompliant.

b. Section 1006.3 does not address the number of stories through which an unenclosed exit access stair may pass. It requires at a minimum, an (one) exit on every other story along the path of egress travel so that the path of egress travel does not pass through more than one adjacent story. Rand Hall has an exit on every story.

First, the architect's contention that "Section 1006.3 does not address the number of stories through which an unenclosed exit access stair may pass" is simply false: Section 1006.3 clearly states that "the path of egress travel to an exit shall not pass through more than one adjacent story." An "exit access stairway" is, by definition, "within the exit access portion of the means of egress system." The exit access stairway is therefore regulated under Section 1006.3 and, as such, "shall not pass through more than one adjacent story." Since the unenclosed exit access stairway in the Mui Ho Fine Arts Library passes through more than one adjacent story, it does not comply with this criterion and is therefore noncompliant.

Second, the fact that Rand Hall has "an exit on every story" is not relevant in this context. As I reiterate in my comments on the architect's "Exhibit 3 Response," the path of egress travel in Rand Hall with "an exit on every story" is part of a completely different means of egress system than the one encompassing the exit access stairway being discussed. Table 1006.3.1 in the 2015 NYS Building Code requires two exits, or access to exits, per story, and Table 1006.2.1 limits the common path of egress travel to 75 feet. In other words, after traveling no more than 75 feet from any point, an occupant must have "separate and distinct access to two exits or exit access doorways" (Chapter 2 definition of Common Path of Egress Travel). In Rand Hall, one of those exits is an interior exit stair labeled "Exit Stair A" and the other one, at the floor of the atrium, is labelled "Exit Stair B." It should be self-evident that in a building required to have two discrete means of egress systems—after the common path of egress travel distance is exceeded—both must be compliant, not just one. And the enclosed interior exit stair ("Exit Stair A") cannot be used for both means of egress systems, as the architect is suggesting in his response. The architect's reasoning here is both sloppy and dangerous.

c. Section 1019.3 regulates when exit access stair enclosures are required. Enclosure is required when access stairs do not meet any one (not all) of the conditions listed. Rand Hall complies with 1019.3 Condition #5 for unenclosed exit access stairs in an atrium that complies with Section 404.

Yes, the exit access stair complies with Section 1019.3, as I indicated in my appeal. The problem is that the stair does not comply with Section 1006.3. Means of egress components must comply with all relevant Code sections, not just those that support a particular argument. My reasoning here is supported by code interpretations from both the ICC and the DBSC.

d. Section 404.9.3 stipulates exit access travel distance within an atrium is limited to 200 feet. The exit access travel distance in the Rand Hall atrium is less than 200 feet.

Because the "path of egress travel to an exit" passes through more than one adjacent story, in violation of Section 1006.3, the requirements in Section 404.9.3 are not relevant in determining whether the exit access stair is, or is not, compliant. To be compliant, all requirements of the Code must be satisfied.

e. The Rand Hall east exit stair (Stair B) provides a continuous exit from the second floor to the exterior to meet Section 1023.3.

Because the "path of egress travel to an exit" passes through more than one adjacent story, in violation of Section 1006.3, the requirements in Section 1023.3 are not relevant in determining whether the exit access stair is, or is not, compliant. To be compliant, all requirements of the Code must be satisfied.

As I indicated in my appeal (see Exhibit 2), the idea that an exit access stairway in an atrium can pass through more than one adjacent story was rejected by James Harding of the New York State Division of Building Standards and Codes, and the logic underlying Mr. Harding's Code interpretation was later supported by Kevin Clark, Assistant Director for Technical Support and Manager Building Standards and Codes, DBSC in an email to me dated Feb. 27, 2019 and included in Exhibit 3. Mr. Clark wrote:

Upon further review of the questions and answers provided, I believe the responses initially rendered by Jim Harding to be correct. Question No. 1 is clearly addressed in Section 1006.3 of the 2015 IBC where it is indicated that 'The path of egress travel to an exit shall not pass through more than one adjacent story.'  Per this section, if using the atrium as part of the path of egress travel as the question stated, it can only pass through one adjacent story or once the traveler as reached an exit.  Per the definition in the 2015 IBC, an exit is 'That portion of a means of egress system between the exit access and the exit discharge or public way. Exit components include exterior exit doors at the level of exit discharge, interior exit stairways and ramps, exit passageways, exterior exit stairways and ramps and horizontal exits.' Whether or not this portion of the specific building discussed in subsequent emails is the path to an exit, part of the exit, etc. and therefore potentially allowed to pass through more than one story, was not a part of the answered provided and will need to be evaluated by the local building department for compliance.

But this final question posed by Kevin Clark—whether or not this portion of the specific building is the path to an exit, part of the exit, etc.—has now been answered by the architect: the unenclosed stair in question has been defined as an "exit access stairway," which places it squarely under the purview of Section 1006.3.

Appendix 2: Egress diagram
Both means of egress systems in the Mui Ho Fine Arts Library contain exit access stairways and interior exit stairways. The means of egress containing Exit Stair A (shown in blue) is compliant. The means of egress containing Exit Stair B (shown in green) is noncompliant because its path of egress travel passes through more than one adjacent story, in violation of Section 1006.3 of the 2015 NYS Building Code.

Egress diagram for Mui Ho Fine Arts Library
Appendix Figure 2.1 Schematic diagram showing means of egress systems in the Mui Ho Fine Arts Library in Rand Hall, Cornell University. Drawing by Jonathan Ochshorn based on life safety plans submitted by STV Architects.

Violation #2: Inadequate number of plumbing fixtures in the roof-top bathrooms.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code sections: Table 1004.1.2, Table 2902.1, and Section 2902.2.2 of the New York State Building Code; Section 419.2 of the New York State Plumbing Code.

Explanation: The stated occupancy for the roof-top art gallery of 131 is found by using an incorrect gallery area of 911 square feet and an incorrect occupant load factor of 7 square feet per occupant (Figure 2). The actual occupancy is 263, based on an actual area of 67'-2½" × 20'-6" = 1,315 net square feet and an occupant load factor of 5. In this calculation, I have computed the gallery area by first subtracting 6 inches from the circulation zone edges and then subtracting 20 square feet for a utility enclosure on the north side of the gallery. This actual area includes the hypothetical egress pathways at the perimeter of the gallery space that are colored yellow on the architect's plans, sheet LSP-103 (Exhibit 4), apparently, and deceptively, to imply that they might represent something literally separated from the orange-colored main gallery space (Figure 2). In fact, the entire actual gallery area—i.e., both the orange and yellow zones drawn by the architects—must be accounted for when computing the occupancy of the space.

Rand Hall roof plan with occupancy calculations
Figure 2. Portion of architect's roof plan showing the roof-top art gallery with fictitious circulation zones (colored yellow) that have been subtracted from the actual occupiable gallery area, along with a tabulation of occupancy group (A-3), occupancy load (7 occupants per square foot), gallery area (911 square feet), and number of occupants (131). Plan drawn by Jonathan Ochshorn based on architect's roof plan, sheet LSP-103. Colors, dimensions, and room data matrix appear in the original.

The architects inappropriately subtracted this yellow colored circulation pathway area from the actual gallery area and used the incorrect load factor of 7 to compute the fictitious values shown on sheet LSP-103, and reproduced in Figure 2, i.e., 911 square feet for the art gallery area and 911/7 = 131 occupants. The architects, in their "Proposed Toilet facilities Calculations" shown on LSP-100 (Exhibit 4), use an even lower number: 93 occupants.

Note that the correct occupant load factor of 5 is for "standing space" (corresponding to the actual use of the space, and consistent with the occupant load factor used, for example, in the nearby Milstein Hall art gallery) whereas the occupant load factor of 7 is for "chairs only—not fixed" a factor which does not represent the actual use of the roof-top space (see Table 1004.1.2 reproduced in Figure 3).

Table 1004.1.2 from 2015 NYS Building Code
Figure 3. Excerpt from Table 1004.1.2 of the New York State Building Code showing occupant load factors for assembly functions without fixed seats. The correct load factor of 5 is for "standing space."

With an actual occupancy of 263 on the roof and 36 occupants on the fourth floor for a total occupant load of 299, the required number of plumbing fixtures, per Table 2902.1 of the New York State Building Code, is 2 WCs and 1 lav (male) and 3 WCs and 1 lav (female). The actual number of WCs provided (1 for males, 2 for females, as shown in LSP-100 reproduced in Exhibit 4) is therefore inadequate. Specifically, Note "a" for Table 2902.1 states that "fixtures shown are based on one fixture being the minimum required for the number of persons indicated or any fraction of the number of persons indicated. The number of occupants shall be determined by this code." Section 2909.2.2 (Fixture calculations) states that "to determine the occupant load of each sex, the total occupant load shall be divided in half. … Fractional numbers resulting from applying the fixture ratios of Table 2902.1 shall be rounded up to the next whole number." With an actual occupant load of 299, the number of male or female occupants is 150. From Table 2902.1, the required number of male WCs (with up to 67% allowed to be replaced with urinals per Section 419.2 of the Plumbing Code) is 150/125 which rounds up to 2; and the required number of female WCs is 150/65 which rounds up to 3.

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. There are sufficient fixtures for the occupant load at the roof terrace.

This statement by the architect is not supported by any argument. I have demonstrated in my appeal that it is false.

b. Section 1004.1.2 (Areas Without Fixed Seating) allows the building official to reduce the maximum number of occupants permitted is [sic] an occupied space. The code official determined that the maximum number of occupants for the exhibition roof terrace is 131. For this number of people one toilet and one lavatory men and two toilets and one lavatory for women are required.

The "exception" to the requirements in Section 1004.1.2 states that the building official may approve a number of occupants that is "less than those determined by calculation." This exception does not apply to the roof-top art gallery in question because the number of occupants assigned to the space (i.e., 131) is not "less than those determined by calculation." In fact, this number was determined by the architect's explicit calculation—and was not reduced by the building official from a higher calculated number. Drawing LSP-103, provided as a pdf attachment in Mr. DiTullio's email to me dated November 17, 2020, and reproduced in my appeal as Exhibit 4, shows exactly how the calculation was made by the architects: first, by incorrectly computing the floor area of the art gallery as 911 square feet; and second, by incorrectly assigning to that space an occupant load factor of 7 square feet per occupant based on the inappropriate category of "Assembly without fixed seats, concentrated (chairs only-not fixed)." Dividing the area of 911 square feet by the occupant load factor of 7 results in a calculated occupant load of 131. The calculation of occupants was flawed in two respects (i.e., floor area and occupant load factor), but it was still a calculation, and did not invoke the "exception" provision of Section 1004.1.2.

To contend at this point—after the two errors in the architect's calculation have been exposed—that it was the building official who somehow approved a reduced occupancy load of 131, is both cynical and dangerous. It is cynical because it acknowledges that the architect's calculation of floor area and choice of occupancy load factor were deliberately falsified in order to reach a smaller occupant load than what is required by the Code. It is dangerous because the actual occupancy of such a space will not be constrained by either the fake calculations provided by the architect or by any reduction approved by the building official. This code violation is not necessarily an acute life-safety issue in regard to the number of plumbing fixtures available. The true danger is that this same reduced occupant load is being used in the Atrium Smoke Control Report to grossly underestimate the potential danger of egressing from the roof-top art gallery during a fire event (see my discussion of Violation #5 below).

Michael Niechwiadowicz, the City of Ithaca building code official who approved this fraudulent calculation, used the same cynical strategy in approving a dangerously low occupancy for a major assembly space in the basement of adjacent Milstein Hall, claiming at first that the 4,978 square foot "crit room" space was only 3,600 square feet and replacing the assembly occupancy load factor of 5 square feet per occupant with a "business" occupancy load factor of 100 square feet per occupant. These fraudulent and dangerous calculations were then used to compute an occupancy load of less than 49 occupants, so that only one legal exit was provided, when the real occupant load was closer to 995! Photographs of the space taken soon after occupancy (see Appendix Figure 1.1) show literally hundreds of alumni, students, and faculty crowded together there, making a mockery of the building department's assumptions. The same level of crowding is to be expected in the high-profile "art gallery" on the roof of the Fine Arts Library building and, for the same reasons, playing games with its occupancy numbers is both dangerous and an abuse of the building official's discretion. After I successfully appealed Mr. Niechwiadowicz's determination regarding Milstein Hall to the Capital Region-Syracuse Review Board, Cornell was forced to demolish a reinforced concrete and glass enclosure in order to provide a second exit for the space. I mention the example of Milstein Hall because, taken together with the occupant load calculation for the Mui Ho Fine Arts library, it points to a consistent pattern of bad behavior by Cornell, their architects, and the Ithaca Building Division.

Milstein Hall Crit Room with one exit and hundreds of occupants
Appendix Figure 1.1. Based on the same fraudulent strategy used to reduce the calculated occupant load for the roof-top art gallery in the Mui Ho Fine Arts Library and approved by the same code enforcement official directing the City of Ithaca Building Division, Milstein Hall's "crit room" design allowed hundreds of occupants to gather in an assembly space with only one legal exit. Image from "AAP buzzes as hundreds of alumni, students, and faculty gather during Celebrate Milstein Hall," March 15, 2012, AAP News & Events, (image source)

Violation #3: Fifth floor incorrectly labeled as mezzanine within the atrium.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code sections: Chapter 2 definitions of mezzanine and atrium; Sections 404.6, 505.2, 712, 1510.2, 1510.2.3, and 1510.8.2.

Explanation: The enclosed area on the roof, open to the atrium, is improperly classified as a mezzanine within the fourth story. Instead, it should be classified as a fifth story adjacent to an atrium. Because these roof-top spaces are adjacent to a vertical opening, their status is governed, in part, by Section 712 (Vertical Openings), which lists 16 options for compliance. Of these 16 options, only three are potentially relevant: 712.1.7 (Atriums), 712.1.11 (Mezzanine), and 712.1.12 (Exit access stairways and ramps). Section 712.1.12, as explained in the discussion of Violation #1, does not apply: James Harding of the New York State Division of Building Standards and Codes writes: "The requirements for exiting from all areas of a building are provided in chapter 10.  Any portion of a building that serves as part of the required means of egress system must comply with all applicable provisions of chapter 10. Referring to Section 1006 for exit access, travel through more than one story of the building is prohibited." Of the two remaining options, only Section 712.1.7, describing an atrium complying with Section 404, is applicable to the roof-top spaces. In other words, the roof-top spaces containing bathrooms, elevator, and corridor must be classified as a fifth story adjacent to an atrium.

The enclosed area on the roof cannot be classified as a mezzanine. To be a mezzanine, it would have to be "an intermediate level or levels between the floor and ceiling of any story," as the definition in Chapter 2 of the code requires it to be. Instead, it is on top of the ceiling of the fourth floor. The explanation of the architects, forwarded to me in an email from the college's Director of Facilities, Frank Parish, dated Jan. 25, 2019, states: "The enclosed area at the roof level is a mezzanine to the fourth floor. The fourth floor is open to and part of the atrium." The architects write in their "Mezzanine diagram" on LSP-103 (Exhibit 4, reproduced below as Figure 4) that the "aggregate area of the mezzanine (580 GSF) is less than one-half the floor area of the room (7,412 GSF) in accordance with BC 505.2.1, exception 2." This room of 7,412 GSF that is shown on the architect's diagram is labeled as the "atrium" rather than as the fourth story, although the boundary shown for the 7,412 GSF corresponds neither to the fourth story nor the atrium. This statement reveals two principal errors. First, a mezzanine cannot be in an atrium, since a mezzanine is "an intermediate level or levels between the floor and ceiling of any story," and an atrium is not a "story"; rather, an atrium is an "opening connecting two or more stories." Second, the roof-top enclosed spaces cannot be a mezzanine within a fourth-story room or space since there are no double-height fourth-story spaces in which a mezzanine could be placed.

annotated mezzanine diagram
Figure 4. Mezzanine Diagram from sheet LSP-103 with the fourth-floor plan superimposed in order to clarify the relationship between the enclosed roof-top spaces (black rectangle) the fourth-floor bookstacks (bounded by a red line), and the third-floor mechanical room for Milstein Hall (bounded by a blue line). Text has been enlarged, red and blue boundaries added, and the 4th floor has been superimposed by the author. The black boundary line is in the original mezzanine diagram.

A schematic section through the roof-top enclosed space is shown in Figure 5 (left). This section clearly shows that there are no double height spaces in the fourth story within which a mezzanine could be placed. To be a mezzanine, the space would need to be configured within a double height space, as shown, for example, in the hypothetical section of Figure 5 (right). In other words, the architects are inappropriately using the atrium opening to demonstrate conformance with Section 505.2.1, rather than using an area within a fourth-floor room or space as the section requires. Both Figures 4 and 5 show that the story incorrectly labeled "mezzanine" (bounded by a black line in Figure 4) is only marginally above any fourth-floor room or space (bounded by a red line), and is not between the floor and ceiling of any fourth-floor room or space, as required by the code. The fourth story below the enclosed roof-top spaces is little more than a corridor, while the greater part of the enclosed roof-top spaces is directly above—not the fourth floor—but rather the ceiling of a third-story mechanical room. The exit access stairway ("Exit Stair B") connecting the fifth-story roof-top spaces to the fourth story is in the atrium, not in any mezzanine, as can be clearly seen in Figure 5.

The original intent behind the definition of a mezzanine was to make special allowances for relatively small platforms or lofts within larger rooms with which they shared a "common atmosphere." Mezzanines, in other words, are only possible within double-height spaces which can contain smaller raised floor levels within their confines. The fifth-floor bathrooms, elevators, and corridors are not within any fourth-floor room or space and therefore do not satisfy this basic criterion.

annotated mezzanine diagram
Figure 5. Schematic section through the as-built fifth-story roof-top enclosed spaces and atrium, which cannot be defined as a mezzanine (left); contrasted with a section showing a hypothetical example of a Code-compliant mezzanine within a fourth-floor room or space (right). Drawn by Jonathan Ochshorn.

By incorrectly defining the mezzanine as being within the atrium, the architects have unwittingly confirmed that it is the atrium, and not some fictitious mezzanine, that actually extends above the fourth-floor level in order to provide access to a roof-top story containing toilet rooms and an elevator, both of which have been made necessary by the addition of a roof-top art gallery with 263 occupants.

Furthermore, what the architects incorrectly call a rooftop "mezzanine" cannot be defined as a rooftop bulkhead/penthouse, and must be counted as an additional story, per Section 1510 (Rooftop structures) of the 2015 NYS Building Code.

First, the Code defines penthouses according to permitted uses, which consist of "the shelter of mechanical or electrical equipment, tanks, or vertical shaft openings in the roof assembly." (1510.2.3) For the record, Section 1510.2.2 the 2018 IBC, recently adopted by NYS, extends the use limitations for penthouses to include elevators and related machinery. However, bathrooms are not permitted within penthouses in either the 2015 or 2018 versions of the IBC.

Second, the Code defines bulkheads, "used for the shelter of mechanical or electrical equipment or vertical shaft openings in the roof assembly," according to the same permitted uses as penthouses (1510.8.2) and expressly states that such rooftop structures "used for any other purpose shall be considered as an additional story of the building."

Since the rooftop structure in the Rand Hall project contains elevators and bathrooms, it cannot be counted as a penthouse/bulkhead, and must be considered as an additional story.

Following are code sections relevant to bulkheads and penthouses:

  • Section 1510.2 Penthouses. Penthouses in compliance with Section 1510.2.1 through 1510.2.5 shall be considered as a portion of the story directly below the roof deck on which such penthouses are located. All other penthouses shall be considered as an additional story of the building.

  • Section 1510.2.3 Use limitations. Penthouses shall not be used for purposes other than the shelter of mechanical or electrical equipment, tanks, or vertical shaft openings in the roof assembly.

  • Section 1510.8.2 Bulkheads. Bulkheads used for the shelter of mechanical or electrical equipment or vertical shaft openings in the roof assembly shall comply with Section 1510.2 as penthouses. Bulkheads used for any other purpose shall be considered as an additional story of the building.

Even if Sections 712.1.7 (Atriums) and 712.1.11 (Mezzanine) were both plausible options for categorizing the vertical opening adjacent to the roof-top spaces, the Code requires that Section 712.1.7 (Atriums) govern, since it has more restrictive requirements than does Section 712.1.11 (Mezzanine): Section 102.1 states that "[w]here, in any specific case, different sections of this code specify different materials, methods of construction or other requirements, the most restrictive shall govern."

There are two immediate code violations that follow logically from the mislabeling of the fifth story as a mezzanine within the fourth story. First, the part of the atrium that extends into the fifth story must be separated from the enclosed roof-top spaces with 1-hour fire barriers, since only three floors can be open to the atrium per Section 404.6, exception 3. Second, as explained in Violation #7, the entire building must be built with Type I construction, rather than with Type IIB or Type IIA construction.

While the fifth story is itself small—containing only bathrooms, an elevator, and a corridor—it provides access to a new unenclosed roof-top art gallery with 263 occupants, greatly increasing the total library occupancy compared to any of the prior versions for which variances were granted. It is also relevant to consider Cornell's intentions for the outdoor roof-top art gallery, advertised on their website for the Mui Ho Fine Arts Library as well as in an exhibit in November 2017 called "Full Scale." Cornell's intention, made explicit in this exhibit and reinforced on the library homepage (reproduced in Figure 6) is that "every few years a new team of students builds a [roof-top] classroom as part of the ongoing research and collaboration in physical making that Cornell is so proud of, as Buckminster Fuller once built one of his geometric domes on the very same roof almost 70 years ago."

annotated mezzanine diagram
Figure 6. (top) Cornell's elevation of the Mui Ho Fine Arts Library shows roof-top pavilions intended to be built every few years; (bottom) Cornell's schematic roof plan is labeled "Experimental pavilion roofscape," explicitly indicating their intention to use the outdoor roof deck to support enclosed, occupiable spaces that would constitute a fifth story.

Pavilions, even if temporary (i.e., "erected for a period of less than 180 days"), still need to follow the requirements outlined in Section 3103, including conforming "to the structural strength, fire safety, means of egress, accessibility, light, ventilation and sanitary requirements of this code as necessary to ensure public health, safety and general welfare." In any case, whether "temporary" or exceeding the 180-day time limit, such covered structures would count as part of a fifth-story, since they are on top of the roof, and could not possibly be designated as mezzanines between the floor and ceiling of the fourth story below, as the definition of a mezzanine in Chapter 2 of the code requires. The current scheme, in fact, contains sprinkler and utility hook-ups at the roof level in anticipation of these future enclosed pavilions. For this reason, a designation of the current fifth-story bathrooms, corridor, and elevator as a mezzanine within the fourth story is not only improper on its own merits, but by implicitly suggesting that the intended future roof-top pavilions will also count as fourth-story mezzanines, Cornell is effectively inviting the Ithaca Building Department to acknowledge and agree, in advance, to that noncompliant vision.

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. Rand Hall complies with Section 502.2, 502.2.1 and 502.2.3 and therefore the enclosed roof area of Rand Hall is a mezzanine within the atrium space below. In addition, the Syracuse Board of Review accepted the enclosed roof level as a mezzanine. See Variance Petition 2016-0269 Nature of Grievance and Relief Sought dated 10/11/2016.

First, the 2015 NYS Building Code sections referred to are not correct: Mezzanines are discussed in Section 505.2, not 502.2.

Second, claiming that the roof-top enclosed space is a mezzanine simply by referring to the sections in the Code that discuss mezzanines, but without actually citing any specific Code provisions, or attempting to refute the detailed arguments in my appeal, does not constitute a useful response.

Third, the Syracuse Board of Review never "accepted the enclosed roof level as a mezzanine," because the various schemes that were brought to their attention for appeal or for variances did not include the roof-top enclosures that are shown in the current plans. In particular, the Variance Petition 2016-0269 cited above, which is included in my appeal as Exhibit 9, does not once even mention the idea of a roof-top mezzanine. The 2016 variance validated the library space outside the bookstack floors as an atrium connecting stories two through four (with a single mezzanine within the fourth story), so the building—including the first-floor F-1 shop occupancy—was now considered to be four stories high.

In other words, the mezzanine in the 2016 proposal was neither in the atrium nor was it a "roof-top" enclosed space, but it was within a fourth-floor room or space and met all Code requirements for a mezzanine (i.e., it was actually within the double-height fourth-floor space and also complied with mezzanine area limits). The current scheme, with its roof-top enclosure above (not within) any fourth-floor room or space has nothing in common with the prior design development scheme considered under the 2016 Variance (see Appendix Figure 1.2 below).

b. Section 505.2 (Mezzanines) states a mezzanine does not contribute to the number of stories in a building.

The statement itself is certainly true, but it is not relevant to the arguments in my appeal because the building in question, as designed and built, has no mezzanine.

c. Section 505.2.1 (Area Limitation) states the aggregate area of a mezzanine or mezzanines within a room shall be not greater than one-third of the floor area of that room or space in which they are located. The space that contains the mezzanine is the atrium; it is not the floor below which is open within the atrium. The mezzanine (enclosed area at the roof level) is less than one-third of the area of the atrium.

To claim that this roof-top enclosed space is a mezzanine within an atrium requires that one violate the definitions of both "atrium" and "mezzanine." As I state in my appeal, "a mezzanine cannot be in an atrium, since a mezzanine is 'an intermediate level or levels between the floor and ceiling of any story,' and an atrium is not a 'story'; rather, an atrium is an 'opening connecting two or more stories.' " To claim that this roof-top space is "in" the atrium—even if it were possible for an atrium to be considered a "story" within which a legal mezzanine must, by definition, be situated—would mean that any portion of any story adjacent to an atrium could just as arbitrarily be classified as a mezzanine in the atrium, as long as it were not more than 1/3 or 1/2 (for a sprinklered building) the area of the atrium. The architect's "explanation" goes against the entire logic that allows both atriums and mezzanines to be exempted from ordinary fire safety stipulations.

Let me be clear: (1) A mezzanine, by definition, must be between the floor and ceiling of a double-height story. If the story isn't at least double height, the mezzanine cannot be inserted within its floor and ceiling and therefore would not comply with the definition. (2) The roof-top enclosed space being characterized as a mezzanine is not situated within a double-height story, but, rather, is a story itself, located on top of (not within) the fourth story. (3) The atrium is not a "story" within which one can situate a mezzanine. Instead, an atrium is an opening that connects two or more stories. Because it is an opening, and not a story, an atrium cannot include a mezzanine, which must be within the floor and ceiling of a story. Of course, a mezzanine can be situated in a story adjacent to an atrium, but such a condition does not apply to the present case.

d. Section 505.2.3 (Openness) states a mezzanine shall be open and unobstructed to the room in which such mezzanine is located except: (1) when the occupant load is not greater than 10 persons, or (2) when mezzanines have two or more exits or access to exits. Although only one exception needs to be met the enclosed area at the roof level meets both exceptions.

The roof-top story in question is not a mezzanine, and so meeting the two criteria listed in Section 505.2.3 is not relevant. Only a mezzanine that meets either of the two criteria in Section 505.2.3 can forgo being open and unobstructed to the room in which it is located. Of course, it is possible for numerous rooms or spaces in a story to either have an occupant load not greater than 10, or to have two or more exits: meeting either criterion does not magically turn such rooms or spaces into mezzanines. If that were the case, then every bathroom with low occupancy and every assembly room with two exits would be a mezzanine—a proposition that is obviously absurd. To invoke this section, the architects would first have to demonstrate that the space in question would meet the definition of a mezzanine if it were open to the room in which it is located. But this roof-top enclosed space is not even located "in a room" at all. It is clearly built directly on top the roof structure defining the ceiling of the fourth floor; in other words, it constitutes a fifth story.

e. Section 1019.3 Condition #5 allows the exit access stair to be open within this mezzanine as previously identified.

The noncompliance of the unenclosed "exit access stair" was discussed in Violation #1. Having a noncompliant exit access stair leading to a fifth story does not somehow turn that fifth story into a mezzanine, nor does it make the stair compliant.

It is possible for a mezzanine to adjoin an atrium, for example, if it is within the floor and ceiling of a double-height room or space adjoining the atrium. Such a hypothetical arrangement is shown in Exhibit 2, page 9 (right image) and also occurs in the Rand Hall 100% design development drawings from November 2016 that show a building configured with four-stories and a mezzanine within the fourth floor. In Appendix Figure 1.2 shown below, the current scheme (left), with an atrium but no mezzanine, is compared with this unbuilt scheme from 2016 (right). The unbuilt scheme from 2016 has a legal mezzanine; the current scheme does not.

comparison of current section and 2016 section of Mui Ho Fine Arts Library showing fifth floor and mezzanine, respectively
Appendix Figure 1.2. Comparison of current scheme (left) and 100% DD scheme (right). The double-height fourth-floor space containing a mezzanine is highlighted in yellow in the 2016 scheme. There are no double-height rooms or spaces in the current scheme that could contain a mezzanine. Instead, what the architects incorrectly call a mezzanine is here called what it really is: the fifth story.

Violation #4: Lack of 1-hour fire-rated construction between the atrium and occupied roof.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code section: 404.6

Explanation: Section 404.6 (Enclosure of atriums) states: "Atrium spaces shall be separated from adjacent spaces by a 1-hour fire barrier constructed in accordance with Section 707 or a horizontal assembly constructed in accordance with Section 711, or both." The architects claim that an occupied atrium roof does not count as an "adjacent space" that requires 1-hour separation from the atrium below. However, even though such a roof occupancy is not a "story," it is still a "space" and the wording in the code section specifies that the atrium shall be separated from "adjacent spaces," and not only from adjacent stories. James Harding of the New York State Division of Building Standards and Codes, in an email to me dated Feb. 27, 2019 (Exhibit 3), explains why it is incorrect to consider an occupied roof space as exempt from the atrium separation requirement. He writes: "Section [404.6] provides that the atrium be separated from adjacent spaces … Though not explicitly stated, space in this context includes any area that may be occupied including the roof as described."

The architects have also claimed that "in the case of Rand Hall, the variance issued permits Type IIb construction with additional sprinkler protection at the roof steel and deck." Even if the code variances were still in effect, the 2016 variance that allowed non-fireproofed horizontal transfer girders was only granted with respect to construction classifications and fire-resistance rating requirements for various elements in order to permit non-fireproofed steel in a Type IIB building with A-3 occupancies on a fourth floor. It said nothing about the need to separate the atrium from all adjacent spaces with fire barriers and/or horizontal assemblies, since there were no roof-top occupancies proposed when the variance was granted in 2016. Therefore, it is incorrect to claim that the 2016 variance allows the roof-top occupancy to exist directly above the atrium with no fire separation.

In addition to providing a 1-hour horizontal assembly between the atrium and the roof gallery, the part of the atrium that extends into the fifth story (see Figure 7 below) must also be separated from the adjacent outdoor roof gallery by a 1-hour fire barrier.

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. We assume the phrase "1-hour fire-rated construction" was intended to mean "1-hour fire barrier" between the atrium and the roof deck.

No, I did not intend for "1-hour fire-rated construction" to mean a "1-hour fire-barrier," since a fire barrier is a vertical wall assembly and therefore cannot separate the ceiling of an atrium from an occupied roof deck directly above it. Rather, the requirement for 1-hour fire-rated construction refers to either a "1-hour fire barrier constructed in accordance with Section 707 or a horizontal assembly constructed in accordance with Section 711" (2105 NYS Building Code, Section 404.6 Enclosure of atriums). In this case, the applicable type of required fire-rated construction is primarily a 1-hour horizontal assembly.

b. The Code does not define the term "space" and typically the code does not require a fire separation between interior enclosed occupied spaces and exterior occupied areas. The only Code required ratings for exterior walls are all related to exits and to exterior fire separation distances.

First, the Code states in Section 201.4: "Where terms are not defined through the methods authorized by this section, such terms shall have the ordinarily accepted meanings such as the context implies." The ordinarily accepted meaning of the word "spaces" that appears twice in Code Section 404.6 ("Atrium spaces shall be separated from adjacent spaces") is clear and unambiguous.

Second, and contrary to the erroneous and misleading statement made by the architect, the Code absolutely and unambiguously does require, in specified cases, a fire separation between interior enclosed occupied space and exterior occupied areas. As I stated in my appeal, Section 404.6 (Enclosure of atriums) requires that: "Atrium spaces shall be separated from adjacent spaces by a 1-hour fire barrier constructed in accordance with Section 707 or a horizontal assembly constructed in accordance with Section 711, or both." I would add that such a requirement to separate interior from exterior occupied space occurs in other contexts and has already been acknowledged by the various architects hired by Cornell to design Rand and Milstein Halls. For example, occupied interior spaces in Rand Hall are separated from the adjacent exterior space under Milstein Hall at the ground floor level with a 2-hour fire barrier, and the exterior space under Milstein Hall is separated from occupied interior spaces in Sibley Hall with a 1-hour fire barrier.

Third, neither the New York State DBSC nor the ICC's technical Code Advisory experts had any trouble reaching the conclusion that a 1-hour horizontal assembly was required in this case: see Exhibit 3 in my appeal for documentation.

c. Rand Hall meets the requirements for an occupied roof per Section 903.2.1.6 (Assembly Occupancies On Roofs) where an occupied roof has an assembly occupancy with an occupant load exceeding 100 for Group A-2 and 300 for other Group A occupancies and all floors between the occupied roof and the level of exit discharge are equipped with an automatic sprinkler system in accordance with Section 903.3.1.1 or 903.3.1.2.

This is not relevant to the requirement that the occupied space be separated from the atrium with 1-hr fire-rated construction.

d. Although not a Code requirement the Rand Hall roof exhibition area was constructed with the following additional safety features:

  1. Sprinklers are designed to wet all roof supporting beams to maintain temperatures less than 1,000 degrees during a fire event.

  2. There is a 2" thick precast concrete deck under the entire roof exhibition area located above the atrium.

  3. Fire alarm and sprinkler services are located within a roof mounted junction box and are designed to be extended to provide coverage of any enclosed, temporary roof structure built as part of the roof exhibition program of the College of Architecture, Art and Planning.

First, none of this is relevant to my appeal. If Cornell wishes to construct a building that violates Code provisions, the appropriate course of action is to acknowledge that they are violating these Code provisions and to provide explanatory documentation as part of a formal Code Variance application, as they have done several times already for this project. It is not appropriate to introduce such allegedly mitigating factors in a Code Appeal hearing, where the only relevant issue is whether provisions of the 2015 New York State Building Code have been violated.

Second, it is disingenuous to claim that these are "additional safety features," as if they were not otherwise required.

Regarding item 1: In fact, sprinklers for the roof-supporting beams were proposed as part of Cornell's 2016 Code Variance in order to substitute Type II-B (non-fireproofed) construction for Type II-A, the construction type required for an A-3 occupancy in a sprinklered 4-story buildings. It is therefore not an "additional safety feature" since it was stipulated as being necessary to compensate for the lack of fire-resistance-rated construction on the structural frame. And the use of sprinklers to justify non-fireproofed construction had nothing to do with the requirement for a 1-hour horizontal assembly separating the atrium from occupied space above, since there was no occupied roof deck in the 2016 scheme. The requirement for a 1-hour fire rating on the entire horizontal roof assembly is not mitigated by having special sprinklers for the non-fireproofed roof beams and providing no fire-resistance rating for the spanning elements between those beams.

Regarding item 2: What is characterized as a 2-in. thick precast concrete deck is nothing more than a grid of roof pavers having no continuity, no fire-resistance rating in this context, little ability to block the passage of smoke, and no relevance to the safety of occupants either on the roof-top art gallery or below in the atrium.

Regarding item 3: Providing alarm and sprinkler service to any "enclosed, temporary roof structure built as part of the roof exhibition program" has no relevance to the requirement that the occupied roof-top art-gallery be separated from the atrium with a fire-rated horizontal assembly. Moreover, the fact that sprinklers are necessary for enclosed rooms above the fourth story (i.e., on a fifth story) is hardly an "additional safety feature" as claimed, but rather a bedrock requirement of the 2015 NYS Building Code, Table 504.4, which does not permit any non-sprinklered A-3 assembly occupancies in Type II-B construction above the 2nd story and does not permit any non-sprinklered college classroom occupancies in Type II-B construction above the 3rd story.

Violation #5: Smoke control system does not protect building occupants.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code section: 909

Explanation: Section 909.1 requires that the atrium smoke control system "provide a tenable environment for the evacuation or relocation of occupants." The Atrium Smoke Control Report (Exhibit 5) prepared by GHD, Inc., dated August 7, 2017, and submitted to the City of Ithaca Building Division, was not updated to include the building's occupied roof gallery with 263 occupants who, in the event of a fire, would need to egress directly through the smoke-filled upper levels of the building. This alone is sufficient cause to overturn the determinations made by the City of Ithaca and the DBSC Oversight Unit. Occupants exiting from the roof gallery during a fire event would primarily use Stair B, which is not only unenclosed as it enters the atrium below, but is interrupted by a long horizontal passage on the fourth story—also completely unenclosed— that leads to the continuation of unenclosed Stair B at the opposite end of the building, thereby exposing hundreds of occupants to the smoke that would tend to migrate precisely to this upper-level area. The Atrium Smoke Control Report draws a troubling conclusion about egress from the low-occupancy fourth story: "Fire location 1 represents the worst-case scenario in regards to smoke accumulation … [T]he smoke layer forms enough to begin descending at approximately 240 seconds. Once the smoke layer develops it descends at an average rate of 0.06 meters per second. At approximately 250 seconds it can be observed that the tenability limit for the top floor has been reached thus meaning that evacuation from the top floor must be accomplished before 250 seconds. By comparing the egress times to the smoke layer descent, it can be determined that complete egress can be accomplished prior to the space becoming untenable to occupants." [Exhibit 5] In other words, the entire viability of the smoke control system is based on the assumption that only 36 occupants are at or above the fourth floor and that these 36 occupants can exit the fourth floor within 250 seconds. However, the Atrium Smoke Control report does not consider the addition of 263 occupants exiting from the roof gallery who also must travel through this fourth floor space in order to reach the continuation of Stair B at the opposite side of the building. These occupants would certainly need substantially more time than the 250 seconds provided by this smoke control system.

The smoke exhaust system not only does not "provide a tenable environment" for emergency egress, but actually places occupants attempting to egress from the roof-top gallery directly in the path of toxic smoke and other exhaust products. This happens in two additional ways. First, many of the roof-top atrium smoke hatches are immediately adjacent to the occupied roof-top art gallery (see Figure 2 and Figure 8), so that smoke being exhausted through these hatches could blow directly onto roof-top occupants. Second, since the atrium extends into the fifth floor (Figure 7), it will act as a primary smoke exhaust vent when the exit door from the roof-top art gallery is opened, even though this door is not labeled, or intended, as an exhaust vent. This smoke control system is entirely passive, with no mechanical exhaust; as a result, art gallery occupants attempting to exit through Exit Stair B will likely encounter toxic smoke emerging from the exit door, which is at the highest point in the atrium and therefore the most likely location for the hot smoke layer to migrate.

section through Fine Arts Library showing smoke plume at exit door to roof decl
Figure 7. Schematic section through atrium showing how the smoke layer would migrate to the highest point of the atrium, directly in the path of egress from the roof-top art gallery through Exit Stair "B"

As described above, it is also uncertain, since it wasn't part of the smoke control model, what the time for evacuation from the roof gallery would be, and how that evacuation would take place with Exit Stair B potentially acting as a smoke exhaust vent at the highest point in the atrium. This is the stair that occupants would be most likely to use, since it is adjacent to the elevator which is the primary mode of entry into the roof-top gallery. In any case, it would certainly take far longer for all 263 occupants to egress from the roof-top art gallery than the 250 seconds that defines the "the tenability limit for the [fourth] floor" through which the roof gallery occupants must travel. Exit Stair B is not only unenclosed from the roof down to the second floor, but is also discontinuous, forcing those evacuating from the roof-top gallery using Exit Stair B to travel horizontally through the fourth floor—by now with the smoke layer making passage on this level untenable—before somehow continuing down Exit Stair B or D (both unenclosed) or across the fourth-floor bridge through the atrium space to enclosed Exit Stair A.

section through Fine Arts Library showing smoke plume at exit door to roof decl
Figure 8. Schematic section through atrium illustrating scenario with fire location at north side of building that was not modeled in the smoke control report prepared for Cornell by GHD

The Atrium Smoke Control Report also does not mention whether the specific building geometry developed for the fire/smoke model takes into account the perforated (i.e., open) floor gratings and six-inch "flue space" under the bookstacks at each floor level. In fact, the requirements of Building Code Section 909.2—that documents must "include sufficient information and detail to adequately describe the elements of the design necessary for the proper implementation of the smoke control systems [and that] these documents shall be accompanied by sufficient information and analysis to demonstrate compliance with these provisions"—are not met, since much crucial information is missing, including the following:

  • Description of fire modeling software, not just evacuation modeling software.

  • Description of building model imported into the fire modeling software (e.g., what floor geometries are assumed; are perforated gratings on the library bookstack floors and floor openings under the bookstacks included in the model, etc.).

  • Description of additional fire location scenarios on the north side of the atrium.

  • Description of the atrium penetrating into the roof-top fifth story and implications for smoke control.

  • Description of the occupied roof-top art gallery and egress/smoke calculations related to the roof-top gallery.

A close reading of the consultant's smoke control report also reveals that, notwithstanding all the missing and incomplete information, a fire at location 3 remains problematic because "there essentially is too much open space with fire sprinklers not located directly in the fire/heat plume to cause sprinkler activation."

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. The Atrium Smoke Control Report has been reviewed again and updated since August 7, 2017 to include the fourth-floor mezzanine and egress of the occupied roof. The smoke control system model successfully demonstrates that all occupants can egress the space before conditions become untenable.

My appeal is a request to review a determination made by the City of Ithaca Building Division (and validated by the New York State DBSC Oversight Unit) to issue a building permit for the construction of the Mui Ho Fine Arts Library at Cornell University. This determination—to issue a building permit—was based in part on an Atrium Smoke Control Report prepared by GHD and dated August 7, 2017. It is not possible to fairly review this determination without examining the actual documents submitted for that permit. Any documents revised after a permit was issued and after my appeal was filed—including a revised and "updated" Atrium Smoke Control Report—are not relevant to this appeal. In other words, it's admirable that the architects and their consultants have finally submitted an Atrium Smoke Control Report that purports to accurately model the building's geometry and occupant load, but any such ex post facto revisions cannot be used to defend a Building Division determination that was made on the basis of inaccurate and, therefore, noncompliant documents. It is important for the Review Board to hold code enforcement officials accountable for improper conduct, when it occurs. Allowing code enforcement officials to circumvent this type of evaluation by removing offending documents and replacing them, after the fact, with revised documents undermines the credibility of the review process.

I haven't seen the revised Atrium Smoke Control Report; its content, as I stated, is not relevant to my appeal. Nevertheless, I will comment of the following points:

1. The building model that was imported into the fire modeling software is that provided in the form of a Revit Architectural Model from STV. The fire model entails all accurate geometry shown by the architectural model.

It is still unclear from this description whether the building model used in the Atrium Smoke Control Report is actually modeling the open metal gratings that comprise the floor-ceiling assemblies of all the bookstack floors or the non-fireproofed roof girders. Stating that the model was created with Revit does not answer that fundamental question.

2. Fire Dynamics Simulator 6.7.0.0 (FDS) is a computational fluid dynamics (CFD) model for fire-driven fluid flow. FDS solves numerically a form of the Navier-Stokes equations appropriate for low-speed, thermally-driven flow with an emphasis on smoke and heat transport from fires.

This is interesting, but not relevant to the issues I have raised.

b. The smoke being exhausted from the smoke exhaust vents does not pose a threat to the rooftop occupants. Smoke passing by the occupants in an exterior space does not threaten to accumulate as there are no obstructions on the roof to maintain a smoke layer that would compromise tenability. Also, the smoke being exhausted from these vents is being pushed upward with a force and speed from the fire and the makeup air being provided in the atrium space.

It is disingenuous to claim that smoke released immediately adjacent to an occupied roof deck, and precisely at the level of the roof deck, would not present a hazard to art-gallery occupants during a fire event. Unpredictable and variable wind directions and wind pressures, especially prevailing winds from the north and northwest, can blow toxic smoke exhausted from the roof hatches directly south onto the open art-gallery space. These exhaust hatches were designed for the roof before an art gallery was envisioned for the roof-top space, and the architects and their consultants never bothered to substantially revise the smoke exhaust design geometry when an add-alternate for the art-gallery was ultimately included in the project. The art gallery assembly space is literally bounded by these smoke hatches on its northern edge, as can be seen in Appendix Figure 1.3 below.

Rand Hall roof deck showing smoke exhaust vents
Appendix Figure 1.3. Smoke hatches form the northern boundary of the roof-top art gallery assembly space.

c. Due to the early detection (VESDA) system that is in place, the smoke exhaust vents will be opened prior to the rooftop terrace occupants beginning to egress, and thus opening the exit access door. The smoke exhaust vents will allow smoke to be pushed upwards out of the building and when the exit access door does open the size of that door alone will not be enough to alter the fire smoke plume to the degree shown in Figure 7. The numerous smoke exhaust vents and the makeup air will continue to push the smoke upwards and out of the building and will dictate the direction of the flow of smoke.

First, the 2017 Atrium Smoke Report describes fire scenarios in which smoke vents would remain closed until the temperatures at the vents reached a specified threshold. Therefore, it is quite possible that smoke could migrate into the highest point in the atrium, i.e., where the unenclosed exit access stairway connects with the roof-top art gallery, before the smoke exhaust vents open.

Second, it is mere speculation to claim that the opening of egress doors has any fixed relationship to the opening of smoke exhaust vents. The smoke exhaust vents open according to some hard-wired program triggered by various fire scenarios, whereas the egress door opens whenever an occupant chooses to open it, either coming up from the lower levels or going down from the roof deck. And if the exhaust vents open before the egress door is opened, that just puts the roof-deck occupants at risk from smoke emanating from the roof vents, as I have described above. Either way, occupants of the art gallery are placed in a precarious situation, with smoke potentially entering the roof-top space through either the vents, through the open exit door, or through both sets of orifices.

d. There is limited combustible loading directly beneath the roof stair opening. All book stacks, desks, and furniture are located in adjacent spaces to the floor area directly below the opening. The stair opening for the roof is surrounded by solid structural members that act as a smoke curtain preventing any smoke from adjacent area from flowing directly into the stair area.

A smoke plume rises to the top of the main atrium ceiling and then begins descending at an average rate of 0.02–0.06 meters per second (2017 Atrium Smoke Report, p.10), depending on whether the smoke vents activate. That means that the smoke layer drops below the "smoke curtain" surrounding the opening containing the unenclosed exit access stair to the roof-top art gallery in approximately 6 seconds (if smoke vents do not activate) or 18 seconds (if the smoke vents activate) and then begins rising into the stair opening itself. The beams that create this smoke curtain are no different from any of the other non-fireproofed steel beams that support the rest of the roof. These beams and girders at the roof of the atrium are not even discussed in the Atrium Smoke Control Report, nor are they shown on the imported Revit model diagrams. Claiming that these beams are "preventing any smoke from adjacent area from flowing directly into the stair area" seems specious, if not deliberately deceptive, since if such beams actually prevented smoke from entering the stair area, they would also prevent smoke from entering the array of smoke exhaust vents! Appendix Image 1.4 shows that, even with the smoke exhaust vents open, the smoke layer has descended far below the roof beams that are allegedly acting as "smoke curtains" to prevent smoke from entering the high point of the atrium.

Rand Hall smoke image from Atrium Smoke Report
Appendix Image 1.4. "Smoke Image 2-5" from the 2017 Atrium Smoke report Appendix shows a smoke layer that has descended well below the position of any and all roof beams, which are not even shown on this Revit-imported model.

e. In the updated smoke control design, half of the 131 occupants on the roof are modeled through the exit access stair and the other half are assumed to utilize the enclosed exit stair which can be accessed from the roof terrace. Due to the slow speed of fire growth of the "worst-case" reception desk fire and the early detection (VESDA) system that is in place that activates the opening of the roof smoke exhaust vents, the results of the fire/smoke and egress models indicate that the occupants will be able to egress below the fourth floor or into an approved enclosed exit stair prior to the space becoming untenable, with a safety factor of over 40 seconds.

First, the assumption of 131 occupants is based on a fraudulent series of calculations, as I have argued in my discussion of Violation #2 above. The actual number of occupants that should be assigned to the roof gallery, based on the actual art gallery floor area of 1315 square feet and the actual "standing space" occupant load factor of 5, is 263 occupants, not 131.

Second, it is unreasonable to assume that only half the occupants will attempt to exit from the primary stair enclosure from which they entered the space. It is well known that the overwhelming majority of occupants choose to exit from the door they are most familiar with—in most cases, the door from which they entered the space. The Fifth Edition of the SFPE Handbook of Fire Protection Engineering puts it this way: "A key feature in any building is the main exit/entrance of the building, as it is well-established that people tend to move towards familiar exits in emergencies." (p. 2050–51)

Third, the idea that even 131 occupants can exit the roof-top gallery in a worst-case scenario "with a safety factor of over 40 seconds" is simply implausible. In a worst-case scenario, the smoke vents will not open; in fact, the 2017 Atrium Smoke Report describes a fire scenario where, "If any initial single or multiple smoke detectors within the floor plan area of the book stacks activate, or the book stacks fire sprinkler water flow zone is activated, the smoke vents shall not open, and the windows for makeup air shall not open until the electronic heat sensor at each individual smoke vent reaches 200 degrees F." Thus, the smoke vents remain closed for an unspecified period of time and may not open at all if any number of system components fail to operate as specified.

Fourth, the 2017 Atrium Smoke Control Report doesn't appear to consider the fact that plumes of hot and toxic smoke will not only enter the atrium space and rise to the atrium ceiling before beginning their descent but will also literally pass through the open steel gratings that comprise all of the exit access paths and occupied spaces on the bookstack floors. This particular hazard will be further examined in my discussion of Violation #9 below, but should also be noted in this context, since it adversely affects the ability of occupants to safely exit the building during a fire event. It must be emphasized that the fourth floor is not merely a space for 40 bookstack floor occupants, but is also acting as an unenclosed exit access pathway for what the architects call "Exit Stair B." In other words, egress from the roof-top art gallery via "Exit Stair B" involves first descending to the fourth floor through an unenclosed exit access stairway, then traveling more than 135 feet along an unenclosed exit access pathway that is directly under the atrium roof along the north wall of the building—precisely where the smoke layer could be moving downward at a rate of 0.02–0.06 meters per second—and then descending down a noncompliant unenclosed exit access stairway that passes through the fourth and third stories before it terminates on the atrium floor. At this point, occupants along this egress path must find the continuation of "Exit Stair B" tucked away in the north-east corner of the building in order to finally reach an exit discharge.

Fifth, the evacuation times assumed in the 2017 Atrium Smoke Control Report are, in part, "based upon the ability of occupants to see/smell/hear what is happening within the open atrium space." (p.8) Clearly, this ability to see/smell/hear an atrium fire is not possible for the hundreds of occupants attending a reception above the atrium in the roof-top art gallery.

f. The occupied roof is no different than any other occupied space in a building requiring two exits. If one exit becomes compromised occupants are expected to use the other exit. In Rand Hall, the occupied roof has one exit access through the atrium and one directly to an enclosed exit stair. Should the exit access through the atrium be compromised occupants on the roof can exit directly to the enclosed stair.

I'm happy that the architects have provided the roof-top art gallery with two exits, as required by the 2015 NYS Building Code. Of course, as I demonstrate in Violation #1, one of these two means of egress is noncompliant because it incorporates an exit access stairway that passes through more than one adjacent story. In any case, having two exits, as required, has nothing to do with the inadequacy of the Atrium Smoke Control Report, so it is not clear why it is brought up as an argument in support of that report.

A smoke report is meant to demonstrate that egress through the atrium remains tenable during a worst-case fire event. It is both irrelevant and inappropriate for the architects to suggest that it's perfectly alright for a descending smoke layer in the atrium to make egress untenable, since "occupants on the roof can exit directly to the enclosed stair." On a crowded roof deck, with smoke potentially obscuring the location of a second unfamiliar (and unconventional) exit that leads to an enclosed stair, and with occupants jostling each other in order to get into the exit access stair they are familiar with, this suggestion by the architects appears ungrounded in the behavior of real occupants experiencing real fire events, and certainly doesn't adequately model a worst-case scenario.

g. The space located on the north side of the atrium does not have a fire load large enough to be a concern. On the north side of the atrium there are metal framed chairs with minor upholstery. Based on the Society of Fire Protection Engineers (SFPE) Handbook, the heat release rate (HRR) of one of these chairs would conservatively be 1/6 of the HRR currently modeled for the reception desk on the south side of the atrium with a slow growth. It is GHD's professional opinion that the "worst-case" fire scenario for occupant tenability continues to be the south reception desk.

A "worst-case" fire scenario should be tested by evaluating all plausible "worst-case" events using available modeling software and accurate models, not by speculation unsupported by rigorous fire science. For example, carts of flammable books or journals are constantly being moved around the library space and could plausibly end up pretty much anywhere, posing a risk in the event of a fire. All sorts of furnishings could be brought into the space over time: this has already been the case in this library and is typical for all buildings. A "worst-case" scenario cannot be determined by excluding precisely such worst-case situations. In any case, my argument that the 2017 Atrium Smoke Control Report is noncompliant has nothing to do with any particular fire scenario modeled in the report itself: the report itself is noncompliant—irrespective of any specific flaws in its model or the choice of fire scenarios—because it neglected to consider the addition of a roof-top art gallery for the scheme, a gallery space that virtually doubled the building's occupancy load. The 2017 Report also neglected to provide "Smokeview" diagrams for Fire Locations #2 and #3; and it shows a maximum temperature, in all of its numerous "Temperature Slice" diagrams, of only 45 degrees C (or about 113 degrees F) during a fire event. To reiterate the crux of my argument: the 2017 report, which was included as a required part of the building permit application, concluded that the low-occupancy fourth floor bookstack story could be evacuated 28 seconds before a descending smoke layer made egress untenable. If we now add an additional 263 occupants to the roof-top art gallery (or even the fraudulently determined number of 131 occupants), most of whom will attempt to use the means of egress associated with the main entrance to the art gallery consisting, in part, of an unenclosed exit access through this same fourth story, it is simply implausible that they will be able to evacuate through the fourth floor before passage becomes untenable due to the descending smoke layer modeled in the report.

Appendix 3: Comments on revised Atrium Smoke Control Report

Comments on revised Atrium Smoke Control Report (Rev. 3—Sept. 23, 2020)
Prepared by Jonathan Ochshorn December 17, 2020


I just received a copy of the Rand Hall Atrium Smoke Control Report (Rev. 3—Sept. 23, 2020) on December 15, 2020 after submitting a FOIL request to the City of Ithaca. As I stated in Appendix 1, "any such ex post facto revisions cannot be used to defend a Building Division determination that was made on the basis of inaccurate and, therefore, noncompliant documents." Nevertheless, having looked through the revised Smoke Control Report, it appears that there are still major deficiencies, including the following:

  1. The revised Atrium Smoke Control Report doesn't indicate whether its underlying model accounts for the openings in the bookstack floors adjacent to the atrium.

  2. The revised Atrium Smoke Control Report now includes consideration of a "rooftop terrace" but assigns it an occupancy load of only 76, with only half of those occupants assumed to egress into the primary exit access stair. The actual occupant load for that rooftop art gallery should be 263, not 76, and far more than half should be assigned to the primary exit in order to model a realistic worst-case scenario (see my Exhibit 2 and corresponding comments in Appendix 1).

  3. The revised Atrium Smoke Control Report doesn't define its "egress destinations" so it is not possible to assess the significance of its computed evacuation times.

  4. The revised Atrium Smoke Control Report models an evacuation time for the rooftop terrace of only 230 seconds, which is less than that of the fourth-floor occupants (272 seconds), even though the rooftop occupants must first exit the roof level and then proceed through the unprotected fourth-floor level, and even though there are many more occupants on the rooftop level than on the fourth-floor level. This implies (but is never actually stated in the Report) that the "egress destination" for the rooftop occupants is modeled as the rooftop exit door leading to the open exit access stairway. If so, the additional evacuation time needed to reach the continuation of the exit access stairway at the eastern end of the fourth floor—through an unprotected access route immediately below the atrium ceiling—has not been counted or considered. And in that case, the uncounted portion of the evacuation route is precisely where the smoke layer at the top of the atrium constrains visibility after only 290 seconds and calls into question the Report's conclusion about the viability of egress from the rooftop art gallery.

  5. The revised Atrium Smoke Control Report continues to falsely claim (p.10) that its evacuation assumptions are "…based upon the ability of occupants to see/smell/hear what is happening within the open atrium space" even though occupants on the rooftop space are completely separated from the atrium below.

  6. The revised Atrium Smoke Control Report is inconsistent in identifying the various floor levels. For example, the evacuation timetable (p. 10) lists 2nd floor, 2nd floor stacks, third floor stacks, fourth floor stacks, and rooftop terrace; whereas the occupancy load table (p. 9) lists second floor, raised second floor, second floor mezzanine, upper mezzanine floor, and rooftop terrace.

  7. The revised Atrium Smoke Control Report claims (or appears to claim—the description in the text is not that precise) that steel beams at the roof level delay "smoke from rising to the higher roof" of the exit access stair that leads to the rooftop art gallery. As I wrote in Appendix 1: "Claiming that these beams are 'preventing any smoke from adjacent area from flowing directly into the stair area' seems specious, if not deliberately deceptive, since if such beams actually prevented smoke from entering the stair area, they would also prevent smoke from entering the array of smoke exhaust vents!"

Violations #6–#8

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Violation #6: Elevator too small for an ambulance stretcher.

Relevant code section: 3002.4

Explanation:

A prior code variance, no longer applicable to as-built library (see "Introductory Comments" above) waived the requirement for an elevator big enough to accommodate ambulance stretchers. Section 3002.4 (Elevator car to accommodate ambulance stretcher) states that "where elevators are provided in buildings four or more stories above, or four or more stories below, grade plane, not fewer than one elevator shall be provided for fire department emergency access to all floors. The elevator car shall be of such a size and arrangement to accommodate an ambulance stretcher 24 inches by 84 inches (610 mm by 21345 mm) with not less than 5-inch (127 mm) radius corners, in the horizontal, open position…" The architects argued that this provision should be waived on the basis of a fourth-floor plan that had only about 36 occupants. With the addition of a roof-top art gallery, there are now hundreds of additional occupants above the fourth story, information that was not part of the 2016 variance request. The correct occupancy of the roof-top art gallery is not 131 or 93, as claimed, although this number is high enough to invalidate the argument made for a waiver of the ambulance stretcher code requirement. As stated above in Violation #2, the real occupancy of the art gallery is 263, based on an actual area of 1,315 square feet and an occupant load factor of 5. Thus, there are 299 occupants on the fourth-story or higher, rather than the 36 occupants in prior versions of this project. Emergency medical technicians would therefore need to carry any injured or disabled occupant (i.e., any of the 263 occupants on the roof-top gallery) down four flights of stairs, putting all such occupants in danger while simultaneously violating the clear language in the New York State Building Code that is designed to protect health, safety and the general welfare.

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

Violations #6–#8. These responses will be discussed together, since they all rely on the same argument.

6. Item Identified in Appeal as Violation #6: The elevator is too small for an ambulance stretcher

a. Granted per Variance Petition 2016-0269, dated 10/11/2016.

7. Item Identified in Appeal as Violation #7: The allowable story height is exceeded for library occupancy without Type 1 Construction.

a. Granted per Variance Petition 2016-0269, dated 10/11/2016.

8. Item Identified in Appeal as Violation #8: The allowable floor area is exceeded at the second story

a. Granted per Variance Petition 2016-0269, dated 10/11/2016.

b. The appellant questioned whether the Variance Petition 2016-0269 remains in effect with the changes made to the building design

c. Cornell returned to the Syracuse Board of Review on June 15, 2017 asking for a determination whether the previously granted Variance Petition 2016-0269 remains in effect. Per the decision letter dated August 7, 2017 the Board of Review determined that it "remains in full force and effect".

The responses to Violations 6–8 all presuppose the continued applicability of the Variance Petition 2016-0269 granted by the Capital Region-Syracuse Board of Review to Cornell's Fine Arts Library proposal. It is true that the Board of Review declined to reopen that variance petition and refused to hear additional testimony at their June 15, 2017 hearing, thereby establishing that the variance remained in effect for the Fine Arts Library proposal as it existed in June 2017. However, as I stated in my appeal (Exhibit 2, page 3):

New York State Code variances only apply to the specific building proposal for which the variances were granted—and cannot be interpreted to give implied approval to a different proposal. All three Rand Hall variances (Petition Nos. 2013-0456, 2015-0432, and 2016-0269) contain the following written disclaimer: "Furthermore, it should be noted that the decision of the Board is limited to the specific building and application before it, as contained within the petition, and should not be interpreted to give implied approval of any general plans or specifications presented in support of this application." Since the library as built is substantially different from all prior schemes, the code variances that supported those prior schemes cannot be applied to the current building.

This conclusion remains unchanged: All of the variance decisions granted for the Fine Arts Library are no longer valid because they are "limited to the specific building and application … contained within the petition" and none of those specific building proposals for which the variances were granted correspond to the design of the current building. The difference between the current building, which is the subject of this appeal, and the various proposals for which variances were granted, are substantial. In particular, the occupancy load of the current building is approximately double that of any of the prior proposals, due to the addition of a roof-top art gallery. Furthermore, the current building contains five stories, whereas the building proposal for which the final variance was granted in 2016 (and reaffirmed in 2017) contained four stories. Both the increased occupancy load and the increased number of stories are critical factors in evaluating the compliance of this building with the 2015 NYS Building Code, an evaluation that should not be prejudiced by variances conditioned upon a "specific building and application" based on provisional design development drawings from 2016 that were ultimately discarded and substantially revised.

A library (A-3) occupancy in a sprinklered building with Type V-B construction—the actual occupancy and actual construction type corresponding to the project without recourse to Code variances—is limited to two stories. A library (A-3) occupancy in a sprinklered building with Type II-B (non-fireproofed) construction is limited to three stories; the same occupancy with Type II-A (fireproofed) construction is limited to four stories. Since the current building has five stories and non-fireproofed construction, notwithstanding the architect's characterization of the fifth story as a "mezzanine," not even the 2016 variance—which allowed a four-story building to be constructed without the required Type II-A fireproofed construction—would permit a library (A-3) occupancy in a five-story building: Table 504.4 of the 2015 NYS Building Code requires Type I-B construction. And the architects, in their response to Violation #4 (see above), have admitted that the roof-top art gallery space provides fire alarm and fire sprinkler coverage for "any enclosed, temporary roof structure built as part of the roof exhibition program of the College of Architecture, Art and Planning" above the fourth floor: in other words, these future enclosed pavilions would be constructed on a fifth floor. As implausible as is the contention that the fifth-story bathrooms and elevators constitute a mezzanine, it is literally impossible to imagine that these future fifth-floor enclosed pavilions—with absolutely no connection to any spaces below them—could be construed as "mezzanines" of either the atrium or of the fourth floor.

A timeline (included at the end of this Appendix) makes it clear that the Hearing Board's 2017 ruling on the 2016 Variance made no mention of the increases in occupant load and number of stories that were eventually included in the application for a building permit. The key dates are August 7, 2017, when the architects claim that the Syracuse Hearing Board issued a "decision letter" reaffirming the validity of the 2016 variance petition; and December 15, 2017, when the architects issued a so-called "conformed set" of drawings that—for the first time—included a roof-top art gallery as an add-alternate, that is, as a possible option that might be considered if the cost could be reconciled with a donor's potential gift. In other words, when the Hearing Board validated the 2016 variance petition, they had only seen a preliminary set of design development drawings for a 4-story building without an occupied roof deck, and they did not validate (and could not even have known about) subsequent increases in occupant load and number of stories.

By the architect's own admission, the Syracuse Hearing Board merely reaffirmed the validity of the 2016 variance, thereby validating only the design development drawings from 2016 that were never built. Moreover, this same petition stated clearly that the Board's ruling was "limited to the specific building and application before it." For that reason, the 2016 variance petition, and all the other variances granted to Cornell before that, cannot be applied to the substantially modified Fine Arts Library proposal for which a building permit was granted on February 16, 2018.

Violation #7: Allowable story height exceeded for library occupancy without Type I construction.

Relevant code section: 504.4, Table 601.

Explanation:

Without recourse to superseded variances, no longer applicable because the current building design is substantially different from prior designs for which variances were granted, the allowable height for an A-3 library occupancy in a sprinklered Type VB building is 2 stories (Figure 9). The current building, with new enclosed spaces on the roof, counts as a 5-story building (Figure 5) and so is noncompliant. Even if wood-framed Sibley Hall is entirely eliminated from consideration, the remaining hypothetical Milstein-Rand combined building, with Type IIB (non-fireproofed steel) construction, has an allowable height of 3 stories (Figure 9) and so remains noncompliant. Even if Rand Hall is considered to be a separate sprinklered building with Type IIB construction, the allowable height is still 3 stories and the building remains noncompliant. Even if the roof-top enclosed space is called a "mezzanine" (but see Violation #3 above) and Rand Hall is considered to be a separate sprinklered building, it would still be noncompliant, since the allowable height for a sprinklered Type IIB building is only 3 stories. Furthermore, the superseded code variances that permitted the building height to exceed 2 stories in 2013, and then to exceed 3 stories in 2015, were based on proposals that did not include a roof-top art gallery with hundreds of occupants (whether or not this level is improperly designated as a mezzanine).

Table 504.4 from 2015 NYS Building Code
Figure 9. Excerpt from Table 504.4, 2015 New York State Building Code, showing allowable number of stories for A-3 (library) occupancies.

Examining Table 504.4 (Figure 9), it becomes clear why the architects have attempted to frame this 5-story building as a 4-story building. Even if they added a fire wall separating Rand from Milstein Hall and added 1-hour fireproofing to every element of the "primary structural frame" and to all of the "floor construction and associated secondary members," the resulting Type IIA construction would still be noncompliant, since the allowable number of stories for an A-3 occupancy with Type IIA construction is only four. A five-story building with library occupancy requires Type I construction, and therefore requires 2-hour fire-ratings for the structural frame and floor construction (see Table 601, reproduced in Figure 10). The current proposal doesn't even provide 1-hour fire-ratings for all of the structural frame and floor construction, nor does it provide a fire wall between Rand and Milstein Hall. Only by building a fire wall between Rand and Milstein Hall and upgrading the construction of Rand Hall to Type I can Rand Hall be considered a separate five-story building conforming to the requirements in Table 504.4.

Table 601 from 2015 NYS Building Code
Figure 10. Table 601 shows that A-3 library occupancies in 5- or 6-story buildings requiring Type I construction need 2-hour fire-resistance ratings on structural frame and floor construction.

Violation #8: Allowable floor area is exceeded at the second story.

Relevant code section: 506.2; 506.3.

Explanation:Without recourse to superseded variances, no longer applicable because the current building design is substantially different from prior designs for which variances were granted, Rand-Milstein-Sibley Hall is a single building with Type VB construction and an actual second-floor area of 41,993 square feet. Assuming a frontage factor of 0.57 (per Section 506.3), the allowable single-floor area for a sprinklered Type VB building governed by Occupancy Group A-3 is 18,000 + 0.57(6,000) = 21,420 square feet, per Table 506.2 (Figure 11). Therefore, the allowable area is exceeded and the proposal is noncompliant. Even if wood-framed Sibley Hall is entirely eliminated from consideration, the remaining hypothetical Milstein-Rand combined building still has an actual second-floor area of 25,642 + 8,751 = 34,393 square feet. Assuming the same frontage factor, and even "upgrading" the construction type to IIB (non-fireproofed steel frame), the allowable per-floor area of 28,500 + 0.57(9,500) = 33,915 square feet is still exceeded by the actual area and so even this hypothetical combined building with reduced size and increased fire resistance remains noncompliant. Only by building a fire wall between Rand and Milstein Hall can Rand Hall be considered a separate building with floor areas conforming to the requirements in Table 506.2.

Table 506.2 from 2015 NYS Building Code
Figure 11. Excerpt from Table 506.2, 2015 New York State Building Code, showing allowable single-floor area (excluding the frontage factor) for A-3 (library) occupancies.

Violation #9: Vertical openings in bookstack floors.

Appeal (Exhibit 2)Text of, and my comments on, architect's response (Appendix 1)

Relevant code sections: 712.

Explanation:Section 712 permits vertical openings in floor-ceiling assemblies only when they are "in accordance with one of the protection methods in Sections 712.1.1 through 712.1.16." The three bookstack stories all have openings in their floor-ceiling assembles, not only the "6-inch flues" below each bookstack, but also throughout the floor-ceiling construction. These levels do not provide assemblies that are "continuous without vertical openings" as required by Section 711.2.2, but rather consist of perforated (open) metal floor panels made with metal bars spaced one half inch apart (Figure 12). Section 712.1.9 (Two-story openings) states that "a vertical opening that is not used as one of the applications listed in this section shall be permitted if the opening … does not connect more than two stories…" The bookstack floor perforations and larger openings under the bookstacks connect more than two stories and are therefore in violation of this section.

The fact that three bookstack stories are open to an atrium does not allow them to violate the requirements in Section 712 regulating vertical openings between stories. In fact, the whole point of fire-safe atrium design is to direct smoke—originating in an adjacent story that is open to the atrium—into the atrium space so that it can rise upwards through the atrium space and be safely exhausted at the atrium roof. The Rand Hall design, in contrast, does exactly the opposite: by having perforated, grated metal floors, in violation of Section 712, smoke will move literally through the occupied floors—which also serve as part of the exit access system—instead of directing smoke into the atrium where it can be safely exhausted. Yet even that simple requirement—to safely exhaust smoke through the atrium roof—is compromised in this building by placing occupied spaces on the roof immediately adjacent to the atrium smoke exhausts (see Violation #5).

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

a. The atrium is open to and connects all three book stack levels. The openings in the bookstack floors are irrelevant since they are all open to the atrium and to each other.

This explanation conflates Exception 3 in Section 404.6, which permits any three floors adjoining an atrium to be open to the atrium, with Section 712, which forbids vertical openings in floor-ceiling assemblies unless they are in accordance with one of the various protection methods listed in the section. In other words, the floors themselves cannot have any vertical openings since they do not conform to any of the available options listed in Section 712. The option listed in Section 712.1.7 for "atriums" applies only to the atrium itself—the atrium being defined as the "opening connecting two or more stories"—and not to the floors adjoining the atrium. It is precisely the purpose of these floor-ceiling assemblies—at least the three floors permitted to be open to the atrium—to direct smoke into the atrium so that it can be safely exhausted; having openings within the floors themselves completely sabotages the entire premise underlying atrium theory and design, since smoke originating in an adjoining floor open to the atrium, instead of being directed safely into the atrium space, will simply rise up through the occupied floors, placing all occupants of those floors at great risk.

The idea that atriums are vertical openings that are specifically designed and intended to exhaust smoke from occupied floors (and not to allow smoke to freely rise up through openings within occupied floors) is made clear in the 2015 ICC Code and Commentary discussion of Section 404.6 (Enclosure of atriums):

It is also recognized that some form of a boundary is required to assist the smoke control system in containing smoke to just the atrium area. The basic requirement, therefore, is that the atrium space be separated from adjacent areas by fire barriers and horizontal assemblies having a fire-resistance rating of at least 1 hour. … Exception 3 recognizes the desire to have at least some floors open to the atrium, and permits a maximum of three. … Essentially these spaces have simply increased the possible design fires that may send smoke into the atrium, thus threatening to send smoke throughout the building and other adjoining spaces. [Italics added]

In all cases, the fundamental idea underlying atrium design is to send smoke "into the atrium" or to contain smoke "to just the atrium area."

b. Section 712.1.7 allows for vertical openings within an atrium. The bookstack levels are contained in the atrium. There are no requirements for solid floors (without openings) within the atrium.

This is simply false. Section 712.1.7, as one of several options to create vertical openings, states only that: "In other than Group H occupancies, atriums complying with Section 404 shall be permitted." There can be no vertical openings "within an atrium" because the atrium, by definition, is itself "an opening connecting two or more stories…." (Section 202). The floors adjoining the atrium cannot be "in" the atrium, since they define the boundaries of the atrium by virtue of adjoining it, not by being in it. Atriums are openings within floors; atriums are not big rooms containing floors. Rather, floors define a boundary outside of and adjoining the atrium. In the specific case of Rand Hall's Mui Ho Fine Arts Library, the atrium is a vertical opening adjoining the bookstack stories: the bookstack stories, along with the building's exterior walls, define the boundary of the atrium and cannot, therefore, be in the atrium.

c. The 6" openings between book stack bays are provided to comply with sprinkler requirements for a book storage facility. This 'warehouse' sprinkler design is provided in addition to the standard sprinkler system designed based on area of head coverage.

Contrary to the assertions made by the architects, the "flue" openings in the floor-ceiling assembly between book stack bays are not a requirement of any building code and are, in fact, strictly prohibited by the 2015 NYS Building Code. Such openings cannot be made in floor-ceiling assemblies separating stories unless they comply with the provisions in Section 712 of the 2015 NYS Building Code. Rather, such 6-in-wide flue spaces are intended for various types of multi-tier rack storage systems placed within high-ceilinged warehouse-type spaces, as described in NFPA Chapter 12. They are not permitted for "library stack areas" in A-3 library occupancies. The fact that these sprinklers are "in addition to the standard sprinkler system" does not allow them to violate Section 711.3.2 of the Code, which states unambiguously that floor and roof assemblies "shall be continuous without vertical openings, except as permitted by this section and Section 712." Section 712, in turn, provides numerous instances where vertical openings can be used, none of which apply to the open floor grates and 6"-wide flue openings in the Mui Ho Fine Arts Library bookstack stories.

Historical digression: Such openings can be found in older multi-tier stack systems but are not used in multi-story libraries anymore because they increase the hazards caused by fires, as argued by Frazer G. Poole in a 1965 article titled "The Selection and Evaluation of Library Bookstacks":

Multi-tier installations consist of two or more levels of stacks in which each level supports the weight of those above. In an earlier era, the spaces between vertical units were left open to allow the circulation of air around the books. These openings, however, promoted vertical drafts and considerably increased the hazards caused by fires. Today, air-conditioning largely obviates the need for this circulation of air around the books and, as a result, the great majority of libraries are constructed with continuous, solid floors, each of which is capable of supporting, independently, the full load imposed by the stacks and the book collection. [source; italics added]

Such multi-tier stacks are also described in the more recent NFPA 13 ("Standard for the Installation of Sprinkler Systems," 2007 Edition, p.179), but they refer, not to library stack areas placed on separate stories as is the case in the Mui Ho Fine Arts Library, but rather to very specific multi-tier self-supporting systems within a single room or space that can accommodate the combined height of the multi-tier storage units. There is nothing in the 2015 NYS Building Code that permits a sprinkler design that violates the floor-ceiling assembly continuity required by Section 711 and 712; the claim that the openings are provided "to comply with sprinkler requirements" is simply false.

The idea that these book stacks constitute "warehouse" storage contradicts the architect's own assignment of an A-3 occupancy class to the library rather than an S-1 occupancy class; and contradicts the architect's own use of the "library stack area" functional category in Table 1004.1.2, rather than the "warehouse" category, to assign occupant load factors to the stacks. Moreover, the actual and intended use of the bookstack floors is not merely for book storage, as in a warehouse or annex facility, but rather as an active and occupied research and browsing space for students, faculty, and visitors. Yes, the occupancy load is low (100 gross square foot per occupant), but the mean fire load density for a library is extremely high—substantially higher than for any other common educational space (see Table 35.3, SFPE Handbook of Fire Protection Engineering, Fifth edition, p. 1138).

It is telling that when Chairman Richard T. Lafferty of the Syracuse Board of Review asked, during the hearing for Variance Petition No. 2015-0432, whether Cornell has "any other precedent for that kind of operation vertically," the only project that Cornell's Hugh Bahar could think of was the A.D. White Library, a high-ceilinged room in Uris Library built in 1891 that contains three levels of multi-tier bookstacks. This is typical of 19th-century book stack design, but wholly inappropriate as a model for modern Code-compliant libraries.

Open steel grate flooring in Mui Ho Fine Arts Library
Figure 12. Perforated floor panels violate basic principles of compartmentation by connecting more than two stories; larger "6-inch flue" openings also appear in each story under the bookstacks. Photo by Jonathan Ochshorn.

Exhibit 3: Code interpretation documentation.

Appeal (Exhibit 3)Text of, and my comments on, architect's response (Appendix 1)

Emails from Kimberly Paarlberg (ICC) and James Harding (DBSC) confirming my Code interpretations for Violation #1 (Unenclosed egress stair in atrium) and Violation #4 (lack of 1-hour horizontal assembly between atrium and roof-top spaces)

Feb. 14, 2019 email from Kimberly Paarlberg (ICC)

Feb. 27, 2019 email from James Harding and Kevin Clark (DBSC)

March 7, 2019 email from Kevin Clark (DBSC) clarifying Feb. 27, 2019 email

March 7, 2019 email from the petitioner responding to Kevin Clark's March 7, 2019 email

From: Kimberly Paarlberg
Date: Thursday, February 14, 2019 at 11:17 AM
To: Renee Testroet , Jonathan Ochshorn
Subject: RE: Sections 1006, 1019 - 2015 IBC

Jonathan,

 

Re:  2015 IBC Sections 404.6, 1005.3.1, 1017.3.1, 1019, 1023.1, 1023.2,

 

Question 1:  Can the path of egress travel to an exit pass through more than one adjacent story in an atrium?

 

Answer 1:  There is nothing saying someone could not follow the stairway in the atrium as a path of egress travel for as many stories as they wanted.  However, if the stairway in the atrium is a considered a required exit access stairway, the egress path for the required means of egress (i.e. number of exits off of a floor) can only go down one story till occupants could choose to move to the required exits (1006.3.1, 1017.3.1, 1019).  If the stairway in the atrium is considered a required exit stairway, while Section 1023.2 Exception 2 does allow for the atrium to meet the construction requirements for the exit stairway, the atrium must meet all the remaining provisions for and exit stairway in Section 1023, including termination at the exterior, not be used for any other purpose other than exit (e.g. no uses on the ground floor of the atrium) and no paths that go through the atrium to get to the 2nd exit.

 

Question 2:  Can an unenclosed interior exit stairway, as permitted in Section 1023.2 Exception 2, terminate in the middle of an atrium floor that is above the level of exit discharge.

 

Answer 2:  No.  Exit stairways must discharge directly to the exterior (1023.1) and cannot stop at an upper floor.  There are the options for the stairway to discharge through a lobby or vestibule (1028), however, this lobby cannot be within the stairway/atrium enclosure and it cannot be at other than the level of exit discharge.

 

Question 3:  Does an occupied roof (Group A-2 assembly) above an atrium need to be separated from the atrium with a 1-hour horizontal assembly?

 

Answer 3:  Yes, an atrium must be separated from occupied spaces.  While a roof is not a story, it is an occupied space, so it must be separated from the atrium where the floor of the occupied roof is over the atrium.  Where an occupied roof floor is around the atrium, a separation would not be required.

 

Code opinions issued by ICC staff are based on ICC-published codes and do not include local, state or federal codes, policies or amendments. This opinion is based on the information which you have provided. We have made no independent effort to verify the accuracy of this information nor have we conducted a review beyond the scope of your question. This opinion does not imply approval of an equivalency, specific product, specific design, or specific installation and cannot be published in any form implying such approval by the International Code Council. As this opinion is only advisory, the final decision is the responsibility of the designated authority charged with the administration and enforcement of this code.

 

I hope that this answers your question in full.  Please feel free to contact me again if you have any additional questions on this issue.

 

"Copyright © 2018 International Code Council, Inc. All rights reserved."

   

Kimberly Paarlberg, RA
International Code Council
Codes and Standards, Senior Staff Architect
5332 Woodfield Drive, Carmel, IN
888-422-7233, Ext. 4306


From: Kevin Clark Codes@dos.ny.gov [Signed by James Harding]
Date: Wednesday, February 27, 2019 at 2:06 PM
To: Jonathan Ochshorn
Cc: "Harding, James L (DOS)"
Subject: RE: Code interpretations

Jonathan:

 

Your questions were forwarded to me to respond to. I discussed the following answers with Kevin and provide them as our consensus opinion on these issues. We reached the same conclusions as did the ICC representative, Kimberly Paarlberg.

 

Question 1: Can the path of egress travel to an exit pass through more than one adjacent story in an atrium?

 

No.

 

The requirements for exiting from all areas of a building are provided in chapter 10. Any portion of a building that serves as part of the required means of egress system must comply with all applicable provisions of chapter 10. Referring to Section 1006 for exit access, travel through more than one story of the building is prohibited:

 

1006.3 Egress from stories or occupied roofs.

The means of egress system serving any story or occupied roof shall be provided with the number of exits or access to exits based on the aggregate occupant load served in accordance with this section. The path of egress travel to an exit shall not pass through more than one adjacent story.

 

The following two definitions are provided to clarify the fact that vertical travel within an atrium is considered to be through a story or stories of the building even though the atrium is separated from the building stories.

 

ATRIUM. An opening connecting two or more stories other than enclosed stairways, elevators, hoistways, escalators, plumbing, electrical, air-conditioning or other equipment, which is closed at the top and not defined as a mall. Stories, as used in this definition, do not include balconies within assembly groups or mezzanines that comply with Section 505.

 

STORY. That portion of a building included between the upper surface of a floor and the upper surface of the floor or roof next above (see "Basement," "Building height," "Grade plane" and "Mezzanine"). A story is measured as the vertical distance from top to top of two successive tiers of beams or finished floor surfaces and, for the topmost story, from the top of the floor finish to the top of the ceiling joists or, where there is not a ceiling, to the top of the roof rafters.

 

Also note that travel within an enclosed exit stairway or ramp, even if such stairway or ramp was entirely within the atrium, would not be considered travel within the atrium. This would be considered travel within the stairway or ramp enclosure.

 

Question 2: Can an unenclosed interior exit stairway, as permitted in Section 1023.2 exception 2, terminate in the middle of an atrium floor that is above the level of discharge?

 

NO.

Section 1023 addresses interior exit stairways and ramps. Although exception 2 of section 1023.2 permits an "interior exit stairways within an atrium enclosed in accordance with Section 404.6," section 1023.3 requires the termination to be as follows:

 

1023.3 Termination. Interior exit stairways and ramps shall terminate at an exit discharge or a public way.

Exception:

A combination of interior exit stairways, interior exit ramps and exit passageways, constructed in accordance with Sections 1023.2, 1023.3.1 and 1024, respectively, and forming a continuous protected enclosure, shall be permitted to extend an interior exit stairway or ramp to the exit discharge or a public way.

Thus, unless the stairway ends at an exit discharge or a public way, it must end at an enclosed exit element as described in the exception. This guarantees that the level of protection for occupants is not diminished along the exit path, which is required for all exits by Section 1022.

There are two exceptions provided in section 1028 to the requirement that the exit discharge be to the exterior of the building, a lobby and a vestibule. Since both of these exceptions are for such areas at the level of exit discharge, they may not be applied to the situation described.

 

A final consideration regarding this question is whether the atrium might also be classified as an enclosed exit element (stairway, ramp or passageway). No. Section 1022 also provides that an exit, "shall not be used for any purpose that interferes with its function as a means of egress." An atrium is provided to for ambiance and other purposes as needed and would not be considered exclusively for use for exiting.

 

Question 3: Does an occupied roof (Group A-3 assembly) above an atrium need to be separated from the atrium with a 1-hour horizontal assembly?

 

Yes.

Section 406. provides that the atrium be separated from adjacent spaces, with two exceptions, as follows:

 

404.6 Enclosure of atriums. Atrium spaces shall be separated from adjacent spaces by a 1-hour fire barrier constructed in accordance with Section 707 or a horizontal assembly constructed in accordance with Section 711, or both.

Though not explicitly stated, space in this context includes any area that may be occupied including the roof as described.

 

Please respond to this e mail to confirm that the above information satisfies your request for a code interpretation and a formal response other than this e mail is no longer requested.

 

Thank you,
James L. Harding, PE, CEO
Division of Building Standards and Codes

 

New York State Department of State
One Commerce Plaza, Albany, New York 12231
(518) 473-7761
James.Harding@dos.ny.gov
www.dos.ny.gov


From: "Duerr–Clark, Kevin (DOS)"
Date: Thursday, March 7, 2019 at 8:07 AM
To: Jonathan Ochshorn , "MNiechwiadowicz@cityofithaca.org"
Cc:: "Harding, James L (DOS)"
Subject: RE: Code interpretations

All:

 

Please be advised the responses below and those provided in the original email are not project specific and have not taken into account any specific design, arrangement, approvals, variances, etc. nor were they intended to overrule or nullify any determinations previously made on a particular project by a local building department, authority having jurisdiction (AHJ), or Board of Review.  The specific building being discussed was not evaluated in this response, nor will it be.  Additionally, it should be noted the questions answered were done so with the assumption that we were discussing new construction.  Existing buildings and modifications to existing buildings would need to be evaluated based on their original construction or compliance path through the NYS adopted 2015 International Existing Building Code as applicable.

The questions were asked and answered in a simple yes/no format with very little additional details (as is requested by this office).  Each building and scenario will need to be evaluated individually for compliance.  The local building department/AHJ has jurisdiction over specific buildings and projects.  Where any aggrieved person feels as though a determination has been rendered incorrectly by the local department, that individual may appeal the determination to a Board of Review (please see our website here).

 

Upon further review of the questions and answers provided, I believe the responses initially rendered by Jim Harding to be correct.  Question No. 1 is clearly addressed in Section 1006.3 of the 2015 IBC where it is indicated that "The path of egress travel to an exit shall not pass through more than one adjacent story."  Per this section, if using the atrium as part of the path of egress travel as the question stated, it can only pass through one adjacent story or once the traveler as reached an exit.  Per the definition in the 2015 IBC, an exit is  "That portion of a means of egress system between the exit access and the exit discharge or public way. Exit components include exterior exit doors at the level of exit discharge, interior exit stairways and ramps, exit passageways, exterior exit stairways and ramps and horizontal exits." Whether or not this portion of the specific building discussed in subsequent emails is the path to an exit, part of the exit, etc. and therefore potentially allowed to pass through more than one story, was not a part of the answered provided and will need to be evaluated by the local building department for compliance. 

 

One of the arguments made that is building specific is that the atrium for the building in question is to be considered part of the interior exit stairway, making it part of the exit by definition, and that it terminates at an exit passageway (addressing question No. 2).  Please note, the stairways would need to meet all of the code requirements for the applicable sections.  Some of which are Sections 1023.1 and 1023.3 which state in part that "An interior exit stairway or ramp shall not be used for any purpose other than as a means of egress and a circulation path" and "Interior exit stairways and ramps shall terminate at an exit discharge or a public way".  Again it is not clear what the proposed use of the atrium is in your specific building, but these provision would need to be met.

 

As you can see, formatting questions as originally noted below may not encompass each and every scenario for each and every building in a simple yes/no format.  A detailed review of how these questions and code sections apply to a specific building is the responsibility of the applicant and their design professional to show to the satisfaction of the local building department/AHJ.

 

Please reach out to me directly with any additional questions.

 

Thank you

 

Kevin Clark

Kevin Duerr–Clark, P.E., CEO
Assistant Director for Technical Support
Manager Building Standards and Codes
Division of Building Standards and Codes

 

New York State Department of State
99 Washington Ave, Albany, NY, 12231–0001
P: 518–474–4073 | C: 518–441–4109 | F: 518–486–4487 
Kevin.Duerr–Clark@dos.ny.gov
www.dos.ny.gov

 

From: Jonathan Ochshorn
Date: Thursday, March 7, 2019 at 11:15 AM
To: "Duerr–Clark, Kevin (DOS)"
Cc: "Harding, James L (DOS)" , "MNiechwiadowicz@cityofithaca.org"
Subject: Re: Code interpretations

Dear Kevin,

 

Thank you for the clarification/disclaimer. I will be first submitting a "Title 19 (NYCRR) of the Rules And Regulations, Department of State" complaint to the City of Ithaca Building Division as soon as I obtain information about the latest iteration of this project (I have been basing my concerns on working drawing sheet numbers LSP–100 – LSP–103 prepared by STV Architects that were submitted on Feb. 14, 2018, and I have not yet been able to confirm whether these drawings have been superseded by later drawings). If I am not satisfied with the outcome of the Title 19 complaint, I will proceed with a complaint at the New York State level, per your website instructions.

 

However, there are three issues mentioned in your email which can be addressed immediately:

 

First, this project counts not just as an "alteration," but primarily as an "addition" since, per the definition in Chapter 2, it is "an extension or increase in floor area, number of stories, or height of a building or structure." As such, it "shall comply with the International Codes as adopted for new construction" (Section 1101.1 of the 2015 Existing Building Code). This much is not contentious: all the construction documents for this project reference the Building Code for new construction, and not the Existing Building Code, in any case.

 

Second, none of the three code variances granted for prior iterations of this project are valid for the current proposal. This is because the current proposal for a Rand Hall library is substantially different from prior proposals for which these code variances were granted. Specifically, the new proposal contains—for the first time—a roof–top art gallery (A–3 assembly occupancy) that greatly increases the building's occupant load compared to all prior proposals. Code variances only apply to the specific building proposal for which the variances were granted—and cannot be interpreted to give implied approval to a different proposal. All three prior Rand Hall variances (Petition Nos.  2013–0456, 2015–0432, and 2016–0269) contain the following written disclaimer: "Furthermore, it should be noted that the decision of the Board is limited to the specific building and application before it, as contained within the petition, and should not be interpreted to give implied approval of any general plans or specifications presented in support of this application." Since the current proposal is substantially different from all prior schemes, the code variances that supported those prior schemes cannot be applied to the current proposal.

 

Third, the use and occupancy of the atrium floor is clearly indicated on the construction documents: it is a library reading room (occupancy group A–3), with an area of 4,928 square feet and, per Table 1004.1.2, an occupant load factor of 50. In other words, it functions as an assembly space with 99 occupants.

 

Best,

 

Jonathan Ochshorn, R.A.
Registered Architect, New York State, since 1979 (#14264)
ICC Membership Number: 8069560

To provide context for my responses, I have inserted my comments from November 23, 2020 (shown in black font) directly into the architect's response (shown in red font).

Mr. Ochshorn attached as Exhibit 3 email correspondence from the ICC and DSBC [sic] that provided advisory opinions in response to questions he posed. Mr. Ochshorn's appeal alleges that these advisory opinions support his belief that Rand Hall does not comply with the Building Code. The core problem with this argument is that Mr. Ochshorn did not provide the ICC or DSBC [sic] with information specific to Rand Hall; instead, he asked general questions that resulted in advisory opinions that do not address the conditions at Rand Hall and are not relevant to this appeal.

Requests for code interpretations directed to the NYS Division of Building Standards and Codes "must be written so that they can be answered 'yes' or 'no.' " and will not be considered if they involve "the acceptability of a design, installation, or product' or 'the review of construction documents' " (source). My requests for code interpretations complied with these NYS requirements. Contrary to the assertion of the architects, these requests for code interpretations correspond precisely to the conditions at Rand Hall and are therefore relevant to this appeal.

His first question did not mention that the exit access stair was in an atrium with an egress path to an exit at every level.

The first code interpretation question asked if the path of egress travel to an exit can pass through more than one adjacent story in an atrium. The answer from both the ICC and the NYS DBSC was "no." The assertion by the architects that one can find "an egress path to an exit at every level" is not relevant, since such exits at each floor level are associated with a second (different) means of egress system. Table 1006.3.1 in the 2015 NYS Building Code requires two exits, or access to exits, per story, and Table 1006.2.1 limits the common path of egress travel to 75 feet. In other words, after traveling no more than 75 feet from any point in a sprinklered Assembly occupancy, an occupant must have "separate and distinct access to two exits or exit access doorways" (Chapter 2 definition of Common Path of Egress Travel). In Rand Hall, one of those exits is an interior exit stair labeled "Exit Stair A" and the other one, at the floor of the atrium, is labelled "Exit Stair B." It should be self-evident that where a building occupant is required to have access to two discrete exits—after the common path of egress travel distance is exceeded—both egress travel pathways must be compliant, not just one.

In my code interpretation request, I didn't mention the fact that there is a second means of egress in compliance with the 2015 NYS Building Code—providing "an egress path to an exit at every level"—because having one legal means of egress is not germane to the question of whether the other means of egress contains an exit access passageway that passes through more than one adjacent story. In other words, both means of egress systems must be Code-compliant, and it is entirely proper to focus on each system separately in this context. There was no exculpatory information withheld from the ICC or DBSC; the compliance of one means of egress system has no bearing on the noncompliance of the other means of egress system. The architects are improperly suggesting that the enclosed interior exit stair ("Exit Stair A") can be used for both means of egress!

His second question failed to mention that the egress path along the exit access stair continued for an additional story through a 1-hour fire rated exit enclosure that terminated directly at the exterior of the building.

The second question is no longer relevant to this appeal, since it asks whether the unenclosed atrium stair in question can be considered an "interior exit stair." The architects, in their response to my appeal, have clarified that they do not consider the stair in question to be an "interior exit stairway" but instead classify it as part of an exit access pathway. I asked the question because, at the time, both code interpretations were being put forward by the architects and by the City of Ithaca Code enforcement officials. Therefore, I wanted to confirm that both interpretations were flawed—and, in fact, both interpretations are flawed, per the ICC and DBSC. Even so, the architect's argument should be refuted. First, I was not asking about an "exit access stair" in my second question, as the architects claim, but rather about an "interior exit stairway." Second, the fact that there is an enclosed interior exit stair leading from the second floor (the floor of the atrium) to an exit discharge at the exterior of the building is not relevant to my question, since this enclosed interior exit stair is neither connected to, nor continuous with, the unenclosed interior exit stairway that I was asking about. This unenclosed interior exit stairway does, in fact, "terminate in the middle of an atrium floor that is above the level of discharge." As such, it is noncompliant and, as before, there was no exculpatory information withheld from the ICC or DBSC.

His third question failed to mention that the occupied roof was not enclosed.

An occupied roof is, by definition, unenclosed. Otherwise, it would not be a roof. The 2016 NYS Building Code discusses "occupied roofs" in, for example, Section 1006.3 (Egress from stories or occupied roofs). To be clear: there is no need to stipulate that an occupied roof is not enclosed, since the Code itself presumes that occupied roofs are not enclosed; otherwise, they would count as stories.

His failure to include all material information specific to Rand Hall undermines the application of those opinions on this appeal. This is further supported by an investigation done by the Department of State, Division of Building Standards and Code. The Division of Building Standards and Code investigated the same allegations raised on this appeal in a separate complaint that Mr. Ochshorn made against the City of Ithaca Building Commissioner, alleging that he did not do a proper review of the Rand Hall project and should not have issued the building permit. In a letter dated October 22, 2019 Gary Traver, Assistant Director, Division of Building Standards and Codes Oversight Unit stated "Based on the review of the information provided by Mr. Ochshorn, documentation submitted to the City of Ithaca Building Department, review of plans and correspondence provided by the City of Ithaca Building Department, interviews and correspondence with you (Building Commissioner Niechwiadowicz), the Designers of Record, as well as site visits by the Division of Building Standards and Codes Staff, it is evident that the allegations of 'not upholding your code enforcement duties' are not supported". After a comprehensive review and looking at the specific conditions at Rand Hall, DSBC reached a different conclusion than what was stated in the advisory opinions based on Mr. Ochshorn's general questions. DSBC [sic] concluded that the building design does meet the requirements of NYS building code and that the building permit was issued appropriately.

The architect's summary of the City of Ithaca and DBSC Oversight Unit determinations is correct. This is precisely why I have filed an appeal. The responses from the City of Ithaca and the DBSC presented conclusions without providing a single argument refuting any of the evidence I submitted. The same pattern is apparent in the architect's current response to my third question. His only argument—that I neglected to stipulate that an occupied roof is not enclosed—is both trivial and ill-informed. Certainly, it cannot be considered a serious refutation of the Code-based argument I made, an argument that was validated through a rigorous code interpretation process implemented by both the ICC and DBSC.

Timeline (included in Appendix 1)

  • July 18, 2013: The Capital Region-Syracuse Hearing Board sustained my appeal—which contended that a temporary move of the Fine Arts Library to the third floor of Rand Hall (which had recently been combined with Milstein and Sibley Halls) was noncompliant—and reversed the determination of the City of Ithaca code enforcement official, Michael Niechwiadowicz. Niechwiadowicz's flawed determination was supported at the hearing by Thomas Hoard, Code Analyst for HOLT Architects and former Building Commissioner for the City of Ithaca.

  • November 21, 2013: Variance Petition No. 2013-0456 determined that the library in Rand Hall was permitted to exceed floor area and height limits for A-3 occupancies in Type V-B construction but continued to classify Rand-Milstein-Sibley Hall as a single building with Type V-B construction.

  • July 31, 2015: 50% Design Development drawings issued for Mui Ho Fine Arts Library.

  • August 12, 2015: Variance Petition No. 2015-0432 determined that 2-hour fire barriers, instead of a 3-hour fire wall, would be allowed, effectively creating a separate "Rand Hall" building with Type II-B (non-fireproofed steel) construction for what—at the time—applied only to a library proposal with a total building height, including the ground-level F-1 shop, of three stories. After granting Cornell's request to use a 2-hour fire barrier instead of the required 3-hour fire wall, the Hearing Board Chairman Richard T. Lafferty remarked: "So back to this, basically as long as you don't come back to us ever, we will be happy to fix this wall." (Hearing transcript, p. 36–7) Of course, this condition was not enforced: Cornell returned to the Hearing Board with another variance request one year later.

  • Early 2016: HOLT Architects—whose Code Analyst, Thomas Hoard, defended HOLT's noncompliant design for Cornell's temporary Fine Arts Library in Rand Hall at the 2013 Appeal Hearing—was selected to plan and design Cornell's Gannett addition.

  • September 15, 2016: Variance Petition No. 2016-0269 determined that a proposed four-story A-3 occupancy in Type II-B construction (otherwise limited to two stories without the variance; and limited to three-stories if only the earlier variance permitting Type II-B construction were in effect) would be permitted.

  • November 16, 2016: 100% Design Development drawings were issued for a four-story library.

  • June 15, 2016: The Syracuse Hearing Board validated its 2016 Variance Petition No. 2016-0269 (and validated it again, according to the architects, in a "decision letter" on August 7, 2017).

  • March 24, 2017: I alerted Cornell's Project Manager for the Mui Ho Fine Arts Library, Hugh Bahar, and Cornell's Dean of Architecture, Art, and Planning, Kent Kleinman, that Code Section 1006.3—which appeared for the first time in the 2015 NYS Building Code—rendered the unenclosed exit access stairway in the atrium noncompliant.

  • December 15, 2017: A "conformed set" of bid documents was issued for Mui Ho Fine Arts Library that included, for the first time, an "add-alternate" for a roof-top art gallery.

  • February 14, 2018: A building permit application was filed for Mui Ho Fine Arts Library in Rand Hall. On the same day, Cornell's architects of record for the Mui Ho Fine Arts Library, STV, issued a 4-sheet set of Life Safety Drawings (LSP-100 through LSP-103). The LSP set was revised several times after that, the latest revision date on the drawings provided by the architects being June 27, 2018.

  • February 16, 2018: A building permit was issued for the Mui Ho Fine Arts Library in Rand Hall. Yes, the permit was issued two days after the application was filed!

  • April 1, 2019: I filed a Code complaint with the City of Ithaca Building Division under Title 19, which was resolved on May 3, 2019 with a statement from the City of Ithaca Director of Code Enforcement, Michael Niechwiadowicz, stating that the "proposed work is in compliance with applicable codes." Not one of my detailed Code-based objections was discussed or refuted.

  • June 10, 2019: I filed a Code appeal (Complaint #4660) with the NYS Division of Building Standards and Codes Oversight Unit, which was "closed with prejudice" on September 26, 2019. Not one of my detailed Code-based objections was discussed or refuted. However, on November 5, 2019, Brian Tollisen of DBSC told me that the DBSC was "finishing up a 'response letter' " to me and that he would wait for receipt of the letter before filing an appeal. On December 3, 2019, Brian Tollisen told me that the case had been "re-opened" and that they were not "leaving any stone unturned." I never received a "response letter" from the DBSC.

  • December 4, 2019: A certificate of occupancy was issued for Mui Ho Fine Arts Library in Rand Hall.

  • February 11, 2020: In spite of "closing" the case on September 26, 2019, Gary Traver of DBSC emailed me to say that my complaint was still being reviewed: "Brian and the other staff members that were reviewing the technical aspects of your complaint have been deployed to Puerto Rico to assist with damage assessments since February 2, 2020. I know that there has been a request by DBSC for additional information from the consultants that provided the smoke control modeling for the project."

  • April 24, 2020: Michael Niechwiadowicz—several of whose determinations for the Milstein Hall addition to Rand and Sibley Halls were overturned by the Syracuse Hearing Board, and whose determinations for the Mui Ho Fine Arts Library are the subject of this appeal—retired from his position as City of Ithaca Building Commissioner after 31 years with the Division and accepted a new job. Mr. Niechwiadowicz is now working for Cornell University as "Engineer Architect II" in Facilities and Campus Services. In this new position, he is asked to "serve as technical mentor for other staff" and "develop, review, and analyze proposed designs and/or construction plans." (source)

  • August 19, 2020: Gary Traver of the DBSC sent me, via email, a summary and status report in which he reiterated that my complaint was discontinued with prejudice on September 26, 2019. However, he also disclosed that the DBSC continued to hold discussions with the City of Ithaca Building Division on this matter: "On August 17, 2020, DBSC staff followed up with the City of Ithaca's Building Department. DBSC was advised that the City now has a new Director of Code Enforcement. The City's new Director of Code Enforcement advised DBSC staff that the City has not received the additional information from Cornell University and/or from the project's Designers of Record relating to the Rand Hall project, but indicated that the City will be following up regarding the same."

  • August 19, 2020: Two hours after I heard from Gary Traver, I received an email from Brian Tollisen of DBSC stating that my complaint "was not reopened," that the DBSC was continuing to study the complaint, but that I should proceed with my appeal: "The complaint investigation regarding NYCRR 1208 was not reopened. The Division was looking closely at all of the code related sections associated with your complaint to make sure we fully understood the technical matters and the methodology employed by the City of Ithaca. I asked you to wait on your appeal in the event that we found a technical matter that would need to be addressed and, possibly, same [sic] you some effort. We have not discovered any technical matters that should be addressed and you should proceed with your appeal if you intend to do so."

  • September 21, 2020: I sent the formal appeal application (dated September 17, 2020), to which this document is an Appendix, to Thomas DiTullio, Division of Code Enforcement, Syracuse, NY.

  • November 17, 2020: Thomas DiTullio, Senior Architect with the DBSC, sent me an email stating: "The Dept. of State DBSC has received additional information from Cornell University and with their permission we forward the documents to you. The DBSC is still in the reviewing and investigating your appeal items and the available board schedules for the soonest hearing date. Should you have any questions please feel free to contact me." Attached to the email were three files: (1) a response to my Code Appeal by David Miles Ziskind of STV Architects which I have reproduced, in red font, in this Appendix; (2) a 4-sheet set of Life Safety Drawings prepared by STV (LSP-100 through LSP-103) whose last revision date is June 27, 2018. I had included the same drawing set as Exhibit 4 of my appeal filed with the DBSC on September 21, 2020 (although my version was made by cobbling together photographs taken of the drawing set); and (3) a copy of my Title 19 complaint to the City of Ithaca, dated April 1, 2019.

© 2020 Jonathan Ochshorn. First posted 22 September 2020. Last updated: 17 March 2021