One of the issues that we frequently must deal with for our projects involving sprinkler protection of storage in warehouses is what to do about sloped ceilings.

From Chapter 12, the chapter on General Requirements for Storage, of NFPA 13 2016 edition, paragraph 12.1.2 states:

12.1.2 Ceiling Slope. The sprinkler system criteria specified in Chapter 12 and Chapters 14 through 20 are intended to apply to buildings with ceiling slopes not exceeding 2 in 12 (16.7 percent) unless modified by a specific section in Chapter 12 and Chapters 14 through 20.

Currently, Chapters 12 and Chapters 14 through 20 do not contain any provisions for modifying this requirement. So, what can one do when asked to provide code complying storage sprinkler systems when the slope exceeds 2-in-12.

We can tell the owner or their architect that the ceiling/roof slope must be limited to 2-in-12 or a flat ceiling must be installed below the roof.

Unfortunately, this is another situation in storage protection design where the requirements for the sprinkler system may dictate the architectural/structural building features.

Now if you are using Factory Mutual criteria there are additional options. Table 14 from FM Data Sheet 2-0 Installation Guidelines for Automatic Sprinklers, January 2014 indicates that for a sprinkler system using quick response sprinklers or standard response sprinklers and no in-rack sprinklers, the maximum slope can be 10 degrees, which is approximately 2 in 12. For standard response sprinklers with in-rack protection, the maximum ceiling slope can be 20 degrees, approximately 4 in 12. (Ugh, here we go again with another thing forcing us toward in-rack sprinklers). In cases where these slopes are exceeded, paragraph 2.2.1.6 of FM Data Sheet 2-0 gives us two options.

• Option 1: Install a flat ceiling.
• Option 2: If rack storage is present, then protect the storage based on the presence of excessive clearance between storage and ceiling/roof in accordance with FM Data Sheet 8-9 (more in-rack sprinklers plus horizontal barriers).

You may have to wait awhile, but there is still hope. The Fire Protection Research Foundation (FPRF), the nonprofit research affiliate of NFPA, is looking into the issue. Thus far, the first two phases of the three-phase research project have been completed.

The Phase 1 report, Protection of Storage Under Sloped Ceilings – Phase, November 2015, contained the following:

• A survey, which looked at existing storage warehouses to determine ceiling types and slopes as well as, storage commodities, storage arrangement, ceiling heights and clearance heights above storage to ceiling.
• CFD simulations to evaluate sprinkler activation times under unobstructed sloped ceilings.

It was curious to me that in the survey it did not try to identify the percentage of unobstructed versus obstructed ceilings present in storage warehouses.

The Phase 2 report, Protection of Storage Under Sloped Ceilings – Phase 2 – Full Scale Test Matrix, October 2017 contained:

• CFD simulations for various obstructed slope ceiling configurations, addressing sprinkler actuation times and sprinkler spray pattern distribution.
• Recommendations for a full-scale testing program (Phase 3) aimed at developing design guidance for sprinkler systems.

Sprinklers actuation simulations were performed with two types of sprinklers. A quick response sprinkler with an ordinary temperature rating (QR/OT) and a standard response sprinkler with a high temperature rating (SR/HT).

The reports can be downloaded from the NFPA website. There is also a companion study that contains a little more detail of the simulations. This study, performed by FM Global Research (and available on its website for download) is titled, Numerical Simulations of Sprinkler Activations and Spray Transport under Obstructed, Slope Ceilings, September 2017.

This FM report provides the following conclusions regarding sprinkler activations. Note that in the simulations purlins were perpendicular to the slope run with girders parallel to the slope:

For QR/OT sprinklers, the following observations are made:

• For horizontal ceilings (0°) and purlin depths of up to 0.6 m (24 in.), a marginal increase (maximum 4 s) in the average activation time is observed for the four sprinklers immediately adjacent to the fire source.
• For a ceiling inclination of 9.5° and purlin depths of up to 0.3 m (12 in.), the average activation time for the four sprinklers immediately adjacent to the fire source is similar to that of the case of horizontal ceilings for the same purlin depths. The average activation times are also comparable to those of the 9.5° smooth ceiling.
• For a ceiling inclination of 9.5° and a purlin depth of 0.6 m (24 in.), considerable activation delay is observed for the non-elevated sprinklers when compared to the average activation time for a smooth, horizontal ceiling. This delay may adversely impact suppression performance.
• For a ceiling inclination of 18.4° and purlin depths of up to 0.1 m (4 in.), the average activation time compares favorably with the smooth ceiling results.
• For a ceiling inclination of 18.4° and purlin depths of 0.2 m (8 in.) and larger, considerable delays in non-elevated sprinkler activations are also observed which may affect suppression effectiveness.
• The presence of a ridge marginally affects the activation times of the four sprinklers surrounding the ignition location. Activations near the ridge are also affected with slightly earlier activation times observed on the near ends of the ridge.
• In the presence of the ridge, less skewness of the sprinkler activation pattern towards the elevated side of the ceiling is also observed for reduced sprinkler stand-off distance.

For SR/HT sprinklers, the following observations are made:

• At inclination angles >9.5° and a purlin depth of 0.3 m (12 in.), highly skewed activation patterns are observed between the elevated and non-elevated sides. As sprinklers on the non-elevated side do not activate, suppression effectiveness will reduce.
• For an inclined ceiling with the ridge located 6.1 m (20 ft) from the CUP array, significantly more SR/HT sprinklers activate near the ridge compared to a ceiling in absence of a ridge. The effect of the ceiling ridge on the sprinkler activation patterns appears more significant for SR/HT than for QR/OT sprinklers.

From the observations in the FM report, I have come up with my own set of conclusions. (Disclaimer:  Assume that I am wrong).

1. For quick response sprinklers, in the case of a maximum 2-in-12 slope with purlins 12-inches deep or less, and for a 4-in-12 slope with purlins 4-inches or less in depth, the current criteria in NFPA 13 is adequate.
2. At some slope greater than 2-in-12, quick response sprinklers may not be effective in controlling the fire if purlins are greater than 24-inches deep. Fortunately, in the type of buildings I deal with purlins are rarely greater than 12 inches deep.
3. For quick response sprinklers, in the case of a maximum 4-in-12 slope with purlins 4-inches deep or less, the current criteria in NFPA 13 may be adequate.

I note that the Phase II report from FPRF did not contain the FM conclusion regarding quick response sprinklers, with a slope of 2-in-12 and purlin depth of 24-inches.

Regarding SR/HT simulations, the FM report indicates that “for the 2-in-12 and 4-in-12 slope simulations, “none of the sprinklers on the (low side of the slope from the location of the test fire) activate”. The report goes on to state that “(for slope ceilings) SR/HT sprinklers would provide a lower suppression performance compared to the QR/OT sprinklers for a given purlin depth.”

It seems to me that what this report is saying is there may be issues with the use of SR/HT sprinklers for ceiling-only sprinkler protection in storage warehouses where the ceiling slopes are 2-in-12 or greater.

The reports do not address standard response sprinklers with an ordinary temperature rating. Would they perform in a more acceptable manner than the high temperature sprinkler? Perhaps this is moot, as given the advantage NFPA 13 gives one for high temperature sprinklers over ordinary temperature sprinkler when using the control-mode-density-application approach, I suppose few warehouses are equipped with ordinary temperature sprinklers.

One of the interesting findings in that came of the CFD simulations on sprinkler sprays was that though NFPA 13 requires sprinkler deflectors to be oriented parallel to the slope of the ceiling, the simulations indicated that sprinklers oriented parallel to the floor resulted in an increase in the “delivered water flux to the commodity.” This deflector oriented parallel to the floor will be examined in the full-scale testing.

Let us hope that the results of the Phase 3 full-scale testing provide clarity and a simple way to deal with ceiling slopes greater than 2-in-12. My skeptical side tells me this research has opened a can of worms (ah, ignorance is bliss) and that the standards will be very stringent with slopes greater than 2-in-12, as well as, provide additional requirements for certain conditions where slopes are less than or equal to 2-in-12 (can you say retroactive?). More in-rack sprinklers and horizontal barriers may be on the way.  That flat suspended ceiling is looking pretty good now.