Message Boards

Google FTO

Download and Distribute the Flyer!

The MIVEC Engine and Fuel Octane Levels - Part I

1995 Mitsubishi FTO GPX

September 2002

Article by:


Being a Japanese import, the FTO's engine management system expects to run on the Japanese 100 octane fuel recommended for this vehicle. Unfortunately, Australian petrol stations don't have a 100RON fuel bowser.

A couple of years ago, however, domestic petroleum suppliers identified a growing demand for higher quality fuels, and responded with products such as BP Ultimate 98 and Shell Optimax.

According to their marketing material, BP's offering is guaranteed to be at least 98RON. It is not unusual, apparently, for this fuel to test at 99RON - very close indeed to the type of fuel for which the FTO MIVEC powerplant was designed. But is it enough?

And what happens if you don't give the MIVEC engine suitable fuel on which to run?

The Evils of Poor Quality Fuel

First and foremost, running an engine on poor quality fuel increases the risk of pre-ignition. Once in each 4-stroke cycle, you have a highly volatile mixture being compressed in a very hot chamber. The FTO 2 litre MIVEC engine has a compression ratio of 10:1, which is quite high for a naturally aspirated powerplant.

If you use fuel with too low an octane rating, the mixture is likely to spontaneously ignite as it is compressed - before the engine management system gets a chance to fire its spark. This is A Bad Thing. Not only is there a flame moving through the chamber in an unexpected manner, but it is likely to meet another flame front moving in a different direction from the subsequent spark plug ignition.

This is commonly known as detonation (yes, it's as bad as it sounds). It is also known variously as knock, pinging or pinking. The "correct" octane fuel is one that resists pre-ignition during the compression stage, thus ensuring the spark plug is the only combustion trigger. Diesel is a different matter, but let's not go there.

A secondary issue (although still topical) is the rate at which different RON fuels burn. Not only is a lower octane fuel more prone to pre-ignition, but it will also burn faster once ignited. The higher the octane level, the slower the burn rate of the fuel-air mixture. If the burn rate is significantly slower than the engine designers expected, there will inevitably be a loss of efficiency. An engine running on fuel with a higher octane rating than optimal can actually show a reduction in torque.

How Can Engine Management Prevent Detonation?

Given all the different fuel types out there, how does the Engine ECU keep things running smoothly?

That's the job of the knock sensor. Through this, the engine ECU "listens" for any pre-ignition occurring. If detected, the spark timing is adjusted to ignite the fuel-air mixture a little earlier in the compression stage. This retarded timing prevents serious piston damage due to detonation. Unfortunately, engine performance is compromised in the process.

So what happens if the sensor malfunctions? The FTO GPX Workshop Chassis Manual documents the following fail-safe mode:

"Switches the ignition timing from ingnition timing for super petrol to ignition timing for standard petrol."

Really? Well, let's give it a try!

I disconnected the knock sensor (see right), then took the GPX out for an on-road dyno. For full details on this marvellous piece of kit, see the Article in question.

This is what the ECU did in response:

  • MIVEC cutover disabled.
  • 6000rpm rev limiter enforced.
  • Conservative ignition timing used.

Here is the performance graph, comparing a typical run against the trial with the knock sensor disconnected...

This is "at-wheels" data, by the way - the calculations did not include any upward correction for transmission loss.

As you can see, the Engine ECU is doing everything possible to protect against detonation. Since it can no longer tell when pre-ignition is occurring, it has simply made sure the ignition timing will be safe for "standard" fuel. It also takes all the fun out of life by disabling the MIVEC system and limiting revs!

We can also infer something else from this...

The blue baseline torque curve was taken using BP Ultimate 98 fuel. It's obvious that the engine computer is working much more efficiently than its "lowest quality fuel" comparison. So if you're still paranoid about using 98RON instead of Japanese 100RON, relax! It's obviously a happy engine compared to "worst case" fuel quality.

But if that's what the torque curve would look like using regular 91RON unleaded, I'm not touching it with a ten foot barge pole. Don't ask me to test it either...

Part II of this article covers some experiments with different quantities of octane booster. Can it really make that "typical" torque curve look healthier? We'll see...

(c) 2002, All Rights Reserved