FPEeXTRAIssue46

The Development of Sprinkler Protection Guidance for ESS

By: Tom Long

In the U.S., lithium-ion battery applications are expected to become a $90+ billion market by the year 2025. Since 2004, electronics were the main use of lithium-ion batteries from 2004 up until approximately 2015, when electronic cars and buses took over the top spot. At that point, energy storage systems (ESS) were just hitting the market. Today, ESS installations are growing but are still considered a minor contributor to the usage of lithium-ion technology but projections show the number growing exponentially. By the year 2025, it is projected that ESS’s will be the second-largest use of lithium-ion battery cells after electric cars.

Local AHJ’s (Authority Having Jurisdiction) and other stakeholders have been challenged by the lack of information regarding fire suppression for these installations.

Relationship between ESS and fire protection

Lithium-ion batteries and ESS have become popular topics amongst the fire protection community due to their potential volatility and the potential fire hazard. The expected growth only further increases the need and importance of fire protection guidance. Previous research projects conducted by the Fire Protection Research Foundation (FPRF) have tested single lithium-ion battery cells and household item sizes modules in different arrangements, but the most recent multi-phase project focused on lithium-ion battery energy storage systems (ESS). Previous work by the FPRF in 2016 “Fire Hazard Assessment of Lithium-Ion Battery Energy Storage Systems” identified additional research needs regarding ESS. The FPRF second phase project, which was supported by the Property Insurance Research Group (PIRG) was intended to determine sprinkler protection guidance for lithium-ion battery based ESS for commercial occupancies. As part of this project, FM Global conducted small-, intermediate-, and large-scale tests to develop sprinkler protection recommendations. Two reports were published in June 2019, the test report by FM Global “Development of Sprinkler Protection Guidance for Lithium-Ion Based Energy Storage Systems” and the FPRF summary report “Sprinkler Protection Guidance for Lithium-Ion Based Energy Storage Systems.”

Phase II Overview and Testing

The overall goal of the testing was to determine performance of water-based fire protection systems on ESS. All tests were conducted using donated battery modules of two different battery chemistries; lithium iron phosphate (LFP) and nickel manganese oxide cobalt (NMC). The predominant different between the hazards were the electrolyte chemistry and the overall energy density. The first three tests conducted were free burn tests in a small-, intermediate-, and large-scale, and the final test was a large-scale sprinklered test. Based on prior work conducted by DNV-GL “Considerations for ESS Fire Safety” , water was used in all suppression tests. These sprinklered tests aimed to provide real scale data to support the use of sprinkler systems as part of the protection strategy.

The small-scale tests were conducted to determine if thermal runaway could be induced. Intermediate-scale testing was conducted to determine the effect of system capacity and thermal exposure. The large-scale tests involved two racks each with 16 modules and were conducted to establish the overall hazard of the ESS. The full-scale sprinkler protected tests were used to determine the performance of a water-based fire protection system typically found in a commercial occupancy where an ESS could be installed.

The sprinkler tests had a similar set up to the large-scale free burn tests and included 16 battery modules in addition to a target rack with 16 modules placed to the left of the main rack to measure fire spread. The sprinklers used were K81 L/min/bar1/2 (K5.6 gpm/psi1/2), QR, nominal 74°C (165°F) temperature rated sprinklers. Using a worst case scenario design, the sprinklers were installed with 3 m x 3 m (10 ft x 10 ft) spacing and the sprinkler link was located 0.3 m (1 ft) below the ceiling. The design area included 49 sprinklers, four were active and could produce water if activated, the remaining 45 were used to indicate operation without discharging water. Additional details on the test set-up can be found in the FM Global report.

Every test level showed for both battery chemistries that ignition of a single module was sufficient to involve all modules within the rack tested. Comparing the two battery types, all stages of testing showed the LFP modules presented a lower fire hazard risk than the NMC modules. During the LFP test, a single sprinkler operated and was able to control the fire spread to the origin rack. In the NMC test, multiple sprinklers activated resulting in a demand area of over 230 m2 (2,500 ft2), and the fire spread from the origin rack to the target rack.

Based on the experimental results, the following conclusions were made:

  • The ESS comprised of LFP batteries under a 4.6 m (15 ft) ceiling was adequately protected by the target sprinkler protection. The water supply should be based on a minimum 230 m2 (2,500 ft2) demand area with a duration of at least 90 minutes. The conclusions are based on a single sprinkler operation controlling the fire to the rack of origin with no involvement of the target rack.
  • The ESS comprised of NMC batteries under a 4.6 m (15 ft) ceiling can be adequately protected by the target sprinkler protection. However, excessive ceiling sprinklers operated during the test conservatively representing a demand area > 230 m2 (2,500 ft2). In addition, fire spread from the origin rack to the adjacent target rack indicating that ESS racks installed side-by-side in a row could eventually become involved in the fire.
  • Large-scale free burn tests as described in Section 6.1 of the FM Global report are recommended to determine adequate space separation distances to prevent fire spread to nearby combustibles or damage to non-combustibles when sprinkler protection is not provided. Large-scale free burn testing is also necessary whenever there is doubt regarding the potential impact a change in an ESS design feature may have on the system hazard.
  • Large-scale sprinkler protected tests as described in Section 6.2 of the FM Global report are recommended to determine adequate space separation distances to prevent fire spread to nearby combustibles or damage to non-combustibles, as well as sprinkler protection design including discharge density/area and water supply duration.

To supplement the conclusions, general protection recommendations for lithium-ion battery based ESS located in commercial occupancies were developed through fire testing. The following recommendations are derived from the results of the specific tests:

For the tested LFP system:

  • Without fire protection, the minimum space separation from any part of the ESS is 1.2 m (4 ft) from non-combustible objects and 1.8 m (6 ft) from combustible objects.
  • With sprinkler protection, the minimum space separation from any part of the ESS is 0.9 m (3 ft) from non-combustible objects and 1.5 m (5 ft) from combustible objects. The sprinkler system water supply should be designed for a minimum 230 m2 (2,500 ft2) demand area and a duration of at least 90 minutes.

For the tested NMC system:

  • Without fire protection, the minimum space separation from any part of the ESS is 2.4 m (8 ft) from non-combustible objects and 4.0 m (13 ft) from combustible objects.
  • With sprinkler protection, the minimum space separation from any part of the ESS is 1.8 m (6 ft) from non-combustible objects and 2.7 m (9 ft) from combustible objects. The sprinkler system water supply should be designed for the total room area where the ESS is located, and the water supply should be calculated as 45 minutes times the number of adjacent racks.

Next Steps

The following is a suggested list of future work to further understand the overall protection requirements for ESS:

  • Determine fire hazard and sprinkler protection criteria for ESS multiple rack installations
  • Test effects of different thermal barriers installed between adjacent ESS racks
  • Test effects of rack design and materials of construction on fire development
  • Evaluate relationship between fire hazard and various battery/module properties
  • Test effects “in rack” sprinkler protection on modules in a rack storage configuration
  • Conduct tests varying sprinkler demand, area, water duration, and separation distances
  • Conduct tests varying manual hose stream flow and duration
  • Evaluate potential hazards for firefighting personnel utilizing manual hose streams
  • Evaluate potential environmental concerns associated with water runoff

A more detailed description of the fire tests outlined in this article will be provided in the Quarter 4, 2019 issue of Fire Protection Engineering magazine.

Tom Long is with Exponent

References

  1. Baes, K., Kolk, M., Carlot, F., Merhaba, A., Ito, Y., Future of batteries, Winner takes all? Arthur D. Little, (May 2018)
  2. Merlin, H. E., The lithium-ion battery end-of-life market – A baseline study, Global Battery Alliance, (2018)
  3. Andrew F. Blum, P.E. CFEI, and R. Thomas Long, Jr., PE, CFEI, “Hazard Assessment of Lithium Ion Battery Energy Storage Systems”, Fire Protection Research Foundation, February 2016
  4. Benjamin Ditch and Dong Zeng, “Development of Sprinkler Protection Guidance for Lithium Ion Based Energy Storage Systems”, FM Global, June 2019
  5. R. Thomas Long, Jr. and Amy M. Misera, “Sprinkler Protection Guidance for Lithium-Ion Based Energy Storage Systems”, Fire Protection Research Foundation, June 2019
  6. DNV-GL, "Considerations for ESS Fire Safety”, Consolidated Edison New York, NY, Final Report OAPUS301WIKO(PP151895), Rev. 4, 2017