Burning and Learning

“How can we ensure that the fire protection solutions we specify actually work for their intended purpose?,” I asked Phil Friday. He studied Fire Protection Engineering at the University of Maryland at the same time as me 25 years ago, and we have kept in touch since then. “The reassurance comes from carefully designed large-scale fire testing,” he replied. This conversation inspired him to write the following piece for the Burning Matters newsletter.

My first real experience with large scale fire testing happened many years ago. I was a consulting fire protection engineer, mainly focused on warehouse and industrial facility protection. We received a call from large building developer, who was being challenged by the local fire official. The ceiling early suppression, fast response (ESFR) sprinkler system they installed was not listed/approved for the aisle widths of the tenant’s proposed racking. The ESFR protection for the building was tested and certified for slightly larger aisles than the tenant wanted and slightly taller ceilings than the building. The contractor could not lower the roof (to everyone’s surprise 🙃), and moving the racks farther apart meant the tenant would lose a significant amount of storage space. Thus, the contractor was stuck between an angry tenant and a fire official who was (rightfully) calling the situation out.

I, however, had an idea: Why don’t we run a large-scale fire test to prove it will work? After all, the ceiling was only slightly taller (about a 1 ft [305 mm] and the aisle width slightly narrower (1.5 ft [457 mm]) than the listing. Wouldn’t these variables offset one another such that it would not negatively impact the ESFR systems performance?, I asked myself.

Oh boy, how wrong can one be? The test did not go well and the developer had to install a lot of in-rack sprinklers to make things right. This profoundly impacted my thinking and caused me to question what I was convinced that I knew. At the end of the day I realized that I was engaged in an all-too-common tendency among engineers – trying to make a prediction beyond the observable range of data; also, known as extrapolation.

That large-scale fire test taught me a lot many years ago:

• Small differences matter.

• Aisle width and ceiling height matters.

• Pressure, flowrate, and discharge pattern (distribution) matters.

Basically, Burning Matters. (See what I did there? 😃 )

In hindsight, I am grateful to have learned that lesson when the only thing at stake was money; and not lives and property.

Let us take a step back and see ‘how the world works’.

The Codes and Standards are Largely Based on It

Have you ever wondered where all of the storage fire protection criteria in NFPA 13 or FM Data Sheet 8-9 (or any number of other fire protection standards) came from? Or all the other fire sprinkler protection criteria found in the various codes and standards? The answer is that much of it grew out of large-scale fire testing. In fact, the two largest certification agencies (UL Solutions & FM Approvals) for storage sprinklers, require large-scale testing as part of the listing/approval process.

What does large scale testing involve?

Large-scale fire testing can be done for exploratory reasons (i.e., research and development) or for the purposes of obtaining a certification (i.e., standard-based), or both. Large-scale fire tests are conducted in a lab that is constructed and operated to accommodate large fires without incurring damage to the facility. This includes fire resistant construction, specialized ventilation and emissions treatment systems, climate control for storing and conditioning test commodity, and backup fire protection systems in case things go wrong. Appropriate instrumentation is needed to monitor and record the test results (e.g., thermocouples, heat flux gauges, visual and thermal imagery, pressure transducers, and calorimetry). And let’s not forget the trained on-site firefighters who monitor each test and serve as the first and last line of defense in case the fire gets out of control.

Large-scale tests for fire protection product (here: sprinklers) evaluation and certification are performed with ‘standard commodities’. One example is the ‘standard plastic commodity’. The standard plastic commodity consist of polystyrene cups in cardboard cartons. Other standard commodities are exposed (uncartoned) unexpanded plastic, or plastic pallets, and then there are Class I through IV standard commodities. In some instances, wood pallets and or and ignitable liquid are used to evaluate protection system performance. Figures 1 through 7 illustrate the standard commodities used in many large scale rack storage tests.

Of course, testing with standard commodities is not always appropriate. For example, in order to learn how to protect vehicles, hazardous materials (e.g., ignitable liquids, aerosols and oxidizers), or lithium-ion batteries, you need to burn those items specifically. Furthermore, unique storage configurations, like automatic storage and retrieval systems (ASRS), may need to be tested specifically in order to determine whether a specific protection strategy will work.

The chosen commodity is typically either stacked on the floor or loaded into racking and then a series of tests are conducted at various storage heights, ceiling heights, aisle widths, and ignition locations. Figures 8 through 10 illustrate a standard test set-up for storage.

What are the Pass/Fail Criteria?

Whether a test is considered a passing test or failure depends on several factors:

• The standard to which the test is performed (e.g., UL199 or FM AS 2000)

• The performance objectives of the system being tested (e.g., to achieve control, suppression, or extinguishment)

This is probably a good time to define what is meant by ‘fire control’, ‘fire suppression’, and ‘extinguishment’.:

• Fire Control - Limiting the size of a fire by distribution of water so as to decrease the heat release rate and pre-wet adjacent combustibles, while controlling ceiling gas temperatures to avoid structural damage. (NFPA 13, 2022 ed., Section 3.3.78).

• Fire Suppression - Sharply reducing the heat release rate of a fire and preventing its regrowth by means of direct and sufficient application of water through the fire plume to the burning fuel surface. ((NFPA 13, 2022 ed., Section 3.3.81).

• Fire Extinguishment – This one is technically a subset of Suppression. Fire extinguishment is fire suppression that requires no application of hose stream water to ensure the fire is out. It is a rare result in large-scale fire testing and in the real world.

Figure 12 represents a simplified fire control versus extinguishment analogy.

In general, for ESFR (or storage) fire sprinkler protection, the large-scale tests are evaluated based on three variables:

1. Maximum steel temperature – A 4 ft (1.2 m) long 2 x 2 x 0.25 in. (50 x 50 x 6 mm) steel angle iron is fixed at the ceiling over the ignition location. Five (5) Type K thermocouples are embedded at even spacing in the steel angle iron. The maximum one-minute average temperature of any thermocouple must not exceed 1,000 ^◦ F (538 ^◦ C). [Next time I will have some values in BTUs/fortnight, I promise. 🥳]

2. Fire spread – Fire cannot spread to either end of the main array  or to the outer side of either target array.

3. Maximum number of sprinklers permitted to activate – For a design that intends to use 12 sprinklers, not more than 8 sprinklers are permitted to activate in a test. This provides a 50% safety factor for properly designed and installed systems.

Note: There is nuance between how UL or FM evaluate large-scale fire test results. Consult the agency specific standards for the specific requirements.

Burning Questions

Now that we know a little more about large-scale testing and what it involves, let’s discuss a couple common questions:

Question 1: What is a 150 MW test (or some other HRR) and does it mean the system suppressed or controlled a 150 MW fire?

Answer 1: For some large-scale fire tests estimating the reduction in heat release rate can be useful to evaluate whether a fixed fire suppression system performed successfully. The stated heat release rate is the potential total heat release rate of the fuel load. Unless it is a flammable liquid fuel fire, the fire size when the system activates is usually much lower. One example where large-scale tests are defined by potential total heat release rate are tunnel fire tests. In a tunnel, the ventilation system is critical to the safety of occupants and first responders. The estimated heat release rate is used by fire engineers as input to the CFD-based fire models used to evaluated tenability conditions at different tunnel ventilation rates.

Question 2: Why can’t we just use fire models in lieu of large-scale tests?

Answer 2: There are numerous factors that hamper our ability to predict sprinklered large-scale fire test outcomes accurately:

• Fire spread in nonuniform fuel arrays is difficult to model.

• The interaction between fire and water droplets is complex and also difficult to model.

• The fluid dynamic flows become extremely dynamic after the first sprinkler activates.

The current state of fire modeling is comparable to that of weather modeling. It is good, but not good enough to predict exactly where a tornado will touchdown, or a hurricane make landfall, until it is too late. The level of accuracy and precision will need to improve significantly before fire modeling can replace large-scale testing as a means of evaluation the effectiveness of a particular fixed fire protection strategy.

Question 3: Please ask us and we will add it.

More discussion will follow on this topic, we are sure, so we will stop for now and await your questions to update the Newsletter with more Burning Questions and Answers.

Phil and Grunde (editing)

The Burning Matters team would love to hear your thoughts! Reach out through the Burning Matters feedback form.