Wendel’s Sprinkler System Design Approach
Many transit facilities in the Northern part of the country operate their buses indoors for many services. Indoor maintenance, fueling, washing and parking to name a few. Due to the inclement weather during the winter months, indoor operations are nearly a necessity in these locations. As we see more and more facilities converting fleets to Battery Electric Buses (BEB’s), there is an increased focus on providing a safe environment with this new fuel type operating indoors. On one hand there is the safety of all the people working in the building and on the other hand there is the protection of the buses, building and the investment made on the equipment.
With this growing use of electric buses, the industry has taken a much closer look at the lithium-ion (Li-ion) batteries within the buses and how they will impact a bus fire. Although electric vehicles are no longer a new technology, additional fire protection requirements for the use of these powerful batteries are still very much undefined. Currently, testing is still being conducted and standards are still being developed, so there is no clear code, standard or path as to how to protect an electric bus from a battery fire. At this time of this writing, we have performed a thorough review of the current Building Code, Fire Code, National Fire Protection Association (NFPA) codes, Factory Mutual (FM) Global published documents, Proterra’s published documents, first responder information provided by Tesla and current industry standards. Below we will outline our findings of current requirements, the minimum level of fire protection required, and our recommended level of fire protection, where electric bus charging, maintenance, and parking will occur.
The Li-ion batteries being used in buses allow for a water-based fire protection system because they do not contain metallic lithium. Based on large scale, live fire tests of Li-ion batteries, large amounts of water are the recommended criteria for extinguishing a Li-ion battery fire. Small amounts of water should be avoided as it is not enough to properly cool the battery and extinguish the fire. The first responder information provided by Tesla warns that a battery fire could take up to 3,000 gallons of water and 24 hours to fully extinguish. Although water is considered an electrical conductor, the conductance is rather low, and tests have proven that the use of water does not present an electrical hazard to firefighting personnel. Once a battery fire is believed to be extinguished, industry recommendations suggest the use of a thermal imaging camera to help identify any areas that may still be burning or overheated. If adequate sprinkler protection is not provided, it is expected that significant battery involvement will occur, which may result in thermal runaway. Once thermal runaway begins, cascading thermal runaway reaction to adjacent cells is possible and may cause an uncontrolled fire. Where the batteries on a vehicle are in the chassis and housed in a shell, water may not be sufficient for full extinguishment, but rather the water may serve as a medium to cool the battery and cell components as thermal runaway subsides. While it may be difficult to prevent cascading of batteries within the chassis, it would be the bigger goal to suppress the fire and prevent the cascading of buses within the garage.
NFPA and FM Global are currently partnered and doing tests to determine how Li-ion batteries can best be protected. One scenario these published documents refer to is Li-ion batteries in cell phones and laptops. These devices are clearly smaller than a bus but have a similar characteristics. The shell of the bus will help to slow the burning of the batteries and the time it takes for any adjacent batteries or buses to catch fire. FM Global recommends that the protection of these items be based on the commodity classification of the product and not the battery. In our case, the bus acts like the equipment containing the battery, and FM Global is essentially recommending neglecting the battery from this situation and protect the vehicle as you typically would.
NFPA 13 defines five (5) standard hazard classifications that most commodities and room types fall into. These are Light Hazard, Ordinary Hazard Group 1, Ordinary Hazard Group 2, Extra Hazard Group 1 and Extra Hazard Group 2. This list start with the smallest hazard types and ranges up to the largest. Light Hazard occupancies are typically office space settings, while Extra Hazard Group 2 occupancies include things like flammable liquid spraying, varnish dipping and plastic manufacturing. If the battery is neglected, as currently proposed by FM Global because it is located within the bus, we would just look at NPFA 13 with the concept of fire protection for bus parking and maintenance. Bus parking would be a hazard classification of Ordinary Hazard Group 1, and Repair Garages falls under Ordinary Hazard Group 2. This would require a density of 0.15 gallons per minute (GPM)/square foot (SF) over a design area of 1500 SF, and 0.20 GPM/SF over a design area of 1500 SF, respectively. This solution is the minimum system required to meet the code at this time for any type of bus parked indoors.
Since the Li-Ion batteries are a concern, a more cautious approach is outlined below. The idea is to keep a battery fire within a bus contained, preventing the spread of fire to adjacent buses and the building. Although the buses are parked very close together, and a bus fire can get extremely hot, a larger density of water may be able to better prevent the fire from growing. Preventing an adjacent bus from catching fire, would eliminate the possibility of more batteries catching. While the specific material composition for each manufacturer differs, we can perform this code review with the assumption that it is some type of Poly Carbon or plastic. Plastic is categorized as the higher hazard commodity according to NFPA 13. We are making the interpretation that buses are being stored, their outer shell is produced of a Group A plastic, and this will likely lead us to a greater protection requirement. Looking at the 2019 NFPA 13 decision tree in figure 21.3.1, the following path is determined: Plastic ⇨ Group A ⇨ Nonexpanded ⇨ Stable ⇨ Exposed.
Expanded plastic, when produced, is typically injected with foam or air bubbles to give it more insulation ability and make it lighter. These types of plastics burn hotter than a denser, unexpanded plastic. Vehicle parts are typically solid, unexpanded plastic materials. The bus would be considered stable as a fire would not cause it to tip over. The wheelbase is wide, and a bus fire would typically just burn down in place. Lastly, we consider the plastic exposed as it would not be covered in any other material that would slow or prevent the fire. This outer shell is essentially our carton for the batteries. Based on the result of this decision tree, NFPA refers to Table 21.3.3(a) or (b) Column E.
With a maximum storage height of 12 feet, since buses tend to range in height from 9 feet to 11 feet, and a building height between 20 feet and 32 feet, column E requires a density of 0.7 GPM/SF over an area of 2,500 SF.
This would exceed the standard high end, Extra Hazard Group 2 classification discussed above, which requires 0.40 GPM/SF over an area of 2,500 SF.
While it is uncertain exactly how the bus and batteries will react in a fire, and how the fire will burn, we believe that designing to the Group A plastic requirement determined above would significantly improve the chance of preventing an adjacent bus from catching fire. The additional water flow would help suppress the fire and keep the adjacent surfaces cool to help prevent cascading.
Although the 0.7 GPM/SF system would be more robust and much safer than an Ordinary Hazard Group 1 system, this is a significant water demand and comes with a cost. While some locations with good local water pressure may be able to handle this demand, many facilities will require the use of a fire pump. A result of the preliminary calculations we have performed show that we have utilized 16.8K CMSA (Control Mode Specific Application) sprinkler heads in our design to help reduce the overall requirement of the system. This approach allows for the calculation of the most remove 15 heads, instead of the maximum required area that a lower K-factor head would require.
It is not our belief that this level of protection is necessary everywhere within a facility. The greatest potential of a battery fire is when energy is being transferred, which means buses are charging , or when there is potential damage, and the bus is being repaired. Charging areas and maintenance are the location that poses the largest risk, and we believe should be protected to this level, until more testing and published code data takes over. Where we have a higher level of protection and a lower level of protection within the same open area, the high level extends beyond the hazard 15ft in all directions. This design approach is in line with current NFPA requirements for multiple hazard types.
We are currently in the middle of design for several facilities in the Northeast area of the country. This approach has been reviewed and agreed upon by our clients, local code officials, State-level code officials and FM Global.
In all cases, we recommend getting the local code officials involved early in the project and developing a solution that everyone is comfortable with.
Still have questions?
Please feel free to reach out to us to discuss your specific needs or application.
Chris Colvin
Alternative Fuels Systems Engineer
p: 716.688.0766 ext 1118
ccolvin@wendelcompanies.com
John Havrilla
Director of Alternative Fuels
p: 716.688.0766 ext 1258
jhavrilla@wendelcompanies.com