Testing the octane rating of aviation fuel

A wise man once said, “If you can keep your head when all about you are losing theirs, then you have evidently not grasped the gravity of the situation.”

When I first started working at Shell back in the late 1960s, we did a lot of work on octane requirements.

Due to the poor correlation between the Research Octane Number (RON) and Motor Octane Number (MON) and the real world, we developed ways to rate the octane of actual fuels in actual cars. This was done by running the car at a wide open throttle condition from one set rpm to another. This process was then repeated again and again, with us advancing the timing a few degrees each time until the engine would knock.

The number of degrees between basic timing and the determined setting could then be correlated back to reference fuels with known octane quality.

A similar system is used to determine the octane requirement of aircraft engines. In the old radial engines, they would increase the boost pressure in small increments until they found knock. They would adjust back to standard conditions and set the limit for takeoff boost to provide a safety margin that would ensure against knocking in the normal operations.

An interesting side note is the engines were so loud that they could not hear knocking, so they would test the engines in the evening or at night and determine knocking by the smoke rings in the exhaust. I have never done this, but it sounds like an exciting test method.

In more modern times, I understand that they put pressure sensors in the spark plugs and watch the pressure vs. crank degrees to look for knock. Knock will appear as pressure waves at and around the peak pressure in the power cycle.

In the last few years, they are using accelerometers mounted under the spark plugs to determine knock. The procedure with most engines today is to advance the timing to determine the “safety margin” for octane required for each particular engine.

Unfortunately, there is a problem with these tests. The tests in the past were not run with standard reference fuels that had exactly a 100 lean rating and a 130 rich rating. Instead, the engine manufacturers would use the fuel they had on hand.

Back when I was working, I looked at the octane of a number of 100/130LL and 100/130 high lead samples. The typical 100/130LL samples had around 104-105 lean rating and around 132-136 rich rating. The 100/130 high lead fuels usually had lean ratings of 100-102 and rich ratings of 138-140.

Therefore, if an engine was certified on 100/130LL, the lean rating of the qualified fuel would have been four to five numbers higher than the “magic” 100 that the unleaded fuels are shooting for.

When this is combined with the lead bonus — because lead fuels rate better in real world engines — you can quickly see that the margin of knock safety that was built into the qualifications has been greatly reduced.

Another point of concern is that if you look at the typical octane ratings for those fuels, you’ll note that the lean rating went up when going from 100/130 high lead to 100/130LL, but the rich rating was typically lower.

If you use the logic that the lean rating and magic 100 is the only criteria, you would have a hard time explaining that the number of knock complaints in the field went up significantly when the switch to 100/130LL was made in the field.

So, the bottom line is that general aviation may be going from a proven fuel that works to an unproven fuel that has probably a six to eight number lower actual lean rating and a rich rating that cannot be measured. What could possibly go wrong?