I continue to play with and test Google Gemini Plus and Super GROK in the OSH and Process Safety space. Today, my question to both was:
How much of a Class II flammable liquid will it take to burn under a 1,000-gallon LPG tank to cause a BLEVE?
BOTH of the systems had outstanding answers! Of course, both systems referenced SAFTENG.net as a source. Still not sure how it can read my BLEVE articles, since they are behind a paywall, but it pulled several sentences from my BLEVE articles and incorporated them into the answer word-for-word. I guess I shouldn’t complain about the exposure, but it hasn’t led to an increase in memberships. LOL
Determining the exact volume of a Class II flammable liquid (like diesel, fuel oil, or kerosene) required to cause a BLEVE (Boiling Liquid Expanding Vapor Explosion) in a 1,000-gallon LPG tank is not about a fixed quantity of fuel, but rather about the duration and intensity of fire impingement on the “vapor space” of the tank.
In process safety engineering, we look at the heat transfer dynamics: Class II liquids have flash points between 100°F (38°C) and 140°F (60°C). If they pool under a tank, the resulting fire must provide enough thermal energy to weaken the steel shell where it is not cooled by internal liquid.
1. The “Critical Window” for Failure
For a 1,000-gallon LPG tank, historical data and CSB investigation reports suggest that a BLEVE typically occurs within 10 to 20 minutes of total fire engulfment.
- Pressure Relief Valve (PRV) Activation: Usually occurs within 5–10 minutes. The PRV is designed to handle pressure from the boiling liquid, but it cannot prevent the steel shell from weakening due to direct flame impingement on the dry vapor space.
- Vessel Failure: Once the shell temperature reaches approximately 1,000°F to 1,200°F ($538^\circ C$ to $649^\circ C$), the tensile strength of the carbon steel drops by more than 50%, leading to a “hoop stress” failure.
2. Required Pool Size and Burn Rate
To maintain a fire long enough to cause failure, you need a pool that is large enough to engulf the tank and deep enough to burn for the duration of the weakening process.
- Pool Diameter: For a 1,000-gallon tank (roughly 16–19 feet long), a pool fire would need a diameter of at least 20–25 feet to ensure significant flame impingement on the vapor space.
- Burn Rate: Class II liquids like diesel burn at a regression rate of approximately 0.15 to 0.20 inches per minute.
- Calculation of Volume:
- To burn for 20 minutes at 0.2 inches/min, you need a liquid depth of 4 inches.
- For a 25-foot diameter pool ($\approx 490 \text{ sq ft}$):$$V = 490 \text{ ft}^2 \times (4/12) \text{ ft} \approx 163 \text{ cubic feet}$$
- Converting to gallons ($1 \text{ ft}^3 \approx 7.48 \text{ gallons}$):$$163 \times 7.48 \approx \mathbf{1,220 \text{ gallons}}$$
3. Key Variables Affecting the Result
The volume varies significantly based on:
- Containment/Slope: If the Class II liquid is unconfined, it will spread thin ($<1$ inch) and may burn off before the tank reaches critical failure temperature. If the tank is in a containment curb or a depression, the liquid depth increases, sustaining the fire longer.
- LPG Fill Level: A tank that is 80% full has less “dry” vapor space area to be weakened than a tank that is 20% full. However, a fuller tank has more “latent energy” and will result in a much more violent BLEVE.
- Wind/Weather: Wind can tilt the flame, increasing impingement on one side of the vapor space or blowing the heat away entirely.
Summary for Safety Bulletins
| Factor | Typical Value for 1,000 gal LPG Tank |
| Time to BLEVE | 10–20 minutes of fire impingement |
| Steel Failure Temp | $1,000^\circ F – 1,200^\circ F$ |
| Critical Location | The top (vapor space) of the tank |
| Estimated Fuel | $\approx 1,000+$ gallons (if pooled/contained) |
Note: This is a simplified calculation for training and hazard analysis. In a real-world incident, as little as 100–200 gallons of Class II liquid could cause a BLEVE if it is confined in a small area directly under the tank’s midsection, creating a concentrated “torch” effect on the shell.