flynhi

Here in Texas, my Suburban runs on 87 octane.
I just got back from a trip to Colo and found regular as low as 85 octane. I used some and had no ill effects. In fact, my mileage went up!
Can someone explain why the octane rating goes down as elevation goes up?
Thanks.

SWCDuke

The lower air density at altitude means lower density at high throttle settings, so there is less tendency to detonate, and you can get away with reduced octane rating. Of course you loose power too, about 20-25 percent at 7000 feet.

Because of the lower air density there is less aero drag at highway speed, and since most modern EFI or TBI systems compensate for altititude (unlike carbureted cars that run rich because they meter based on volume, not mass) either with a MAF or via the O2 sensor, it's usual to see reduced fuel consumption.

Duke

JerryL

SWCDuke has it right. It boils down to cylinder pressure, the lower the pressure the less octane is needed.

MoMo

For the benefit of those who might not know, Octane does not boost horsepower. It slows down the flame front, which reduces the tendency for pre-ignition. Higher octane allows for higher compression engines.

I have two cars that run 10:1 on iron heads in Colorado on 91 octane "premium" unleaded. I have no detonation as long as I keep the engine temps down around 200 or lower. But when I took one of them down to 800 feet ASL, I had real problems with detonation, even with the 93 octane premium and two bottles of octane boost (NOS brand).
I was forced to back off my timing until I got back to Denver.

So it is true, the higher the altitude, the less you need the octane. In my opinion, the octane still isn't high enough. I'm right on the edge in Denver and I'm still okay on pump gas. But at low altitudes, forget it.

SWCDuke

No, you've got some misconceptions here. Normal combustion means that the fresh charge is consumed at a predicatable rate. In a quienscent charge, flame front speed is about 300 ft/sec for common hydrocarbon fuels. In an engine, turbulence increases the effective flame speed, and in a good design the number of degrees to complete combustion will be about the same at any engine speed and split evenly on either side of TDC. If we light the fire at 38 degrees BTC combustion will be complete at 38 ATC. At 6000 revs, combustion is completed in half the clock time than at 3000 revs. This is why detonation is more prevalent at low revs. The more time the last of the unburned charge is at high temp. and press., the more likely it is to detonate.

Detonation and pre-igntion are two different things. Pre-igntion is the ignition of the charge as if there were another spark plug. It is equivalent to increasing the timing, and this can lead to detonation.

Detonation is "abnormal combustion". As temperature and pressure rises, the charge suddenly reacts before it is consumed by normal flame propagation, greatly increasing temperature and pressure. The shock waves from detonation bounce around the combustion chamber and substantially increase the heat transfer rate to the combustion chamber boundaries. That's why valves and pistons break. They overheat, weaken, and fail. Also the increased rate of heat transfer can create hot spots that act as preigntion sources, which makes the detonation worse. It's a positive feedback loop!

Regular and premium fuel burn at about the same rate and have about the same heating value, so they will produce about the same HP as long as there is no detonation, but premium fuel is more resistant to detonation as described above and will support normal combustion with higher CRs, and higher CRs increase thermal efficiency, which increases power while reducing fuel consumption.

The bottom line is that an engine designed and tuned for regular fuel will not make more power or get better fuel economy on premium. A modern engine designed for premium fuel will usually perform satisfactorily on regular because the knock sensor will retard the advance enough to keep it out of signficant detonation, but there will be less torque, most noticeable at low revs, and increased fuel consumption, especially under load.

On a vintage engine you need enough octane rating to keep the engine out of detonation. If you have detonation you either have to increase the octane or retard/slow down the spark advance curve. Once you have detonation free operation, increasing the octane rating has no value.

Duke

MoMo

Thanks Duke.
I must have read your post 3 times to find the difference between detonation and preignition, but it still looks like the same thing.

Other than that, you're basically saying what I just did. Octane does not make horsepower, the engine does. Octane just allows for it if the engine needs it. In very basic terms. But thank you for your excellent explanation.

Now on the difference between detonation and pre-ignition...
Pre-ignition does act like there's another spark plug that creates a flame front other than the spark plug's flame front. This can be caused by hot spots, as you said, such as sharp edges, and excessive carbon build up.
Smoothing out the sharp edges and ceramic coating the piston face can help. Also, increasing swirl and tumble can help, which is part of the head design. I understand all that.

How is that different from detonation? Detonation is also a different-from-intended flame front, and yes, the pressure waves create heat and stress on the piston and valves...and rods. Is it one of those "all cactus are succulents but not all succulents are cacti" situations? If so, in what way?

The way I've always pictured (okay let's use a more generic term, so that I don't get hit on technicalities) engine knock or ping...
As the piston is coming up, compressing the air/fuel mixture, there is a premature explosion which is like a mallet hitting the top of the piston while it's still coming up, against the force of the crankshaft. That, plus the heat, is what cracks, melts, and blows holes in pistons.

So please, for the benefit of me and others, what exactly is the difference between detonation and pre-ignition, or are they, as I've always thought, different names for the same event that causes knock and ping?

MoMo

Wait a minute. Maybe I understand now.
Are you saying that detonation is like dieselling, in which the fuel, because of it's higher flammability from lower octane, self-ignites?
And then pre-ignition does the same thing, but is caused by hot spots instead of the octane level?

That brings up another question. So octane can control detonation, but the hot spots would still be there, right? Would octane have any effect on pre-ignition caused by hot spots?

As you know, when you get a crate engine, already built up for you, the builder does not concern himself with the sharp edges cut into the piston for valve clearance. So you get this engine, already assembled, with knife-sharp edges on the piston face, and because you don't want to contaminate the engine with aluminum particles, you don't grind it once it's built up. That could be a source of hot spots. Would higher octane have any affect on controlling that?

SWCDuke

Detonation and pre-ignition are NOT the same thing, but are closely interrelated. Lets say that we have an engine on a dyno running under load and combustion is normal i.e. detonation free. With normal combustion the flame front propagates at a predictable rate.

Now we begin to advance the timing until we hear knock, which is detonation. Advancing the timing increased peak temperature and pressure, but at some point it becomes too much for the fuel's octane rating and abnomal combustion (detonation) begins. In this case PART of the fresh charge is consumed at the predicatble and steady rate, but then the remaining unburned charge explodes like a stick of dynamite. There is a sudden and significant rise in temp. and press. and shock waves bounce around the chamber. This is the "knocking" sound we hear. Light knock means that only a small part of the unburned charge is consumed by detonation. Heavy knock means that a larger portion of the unburned charge is consumed by detonation.

The above is detonation, but pre ignition is not involved.

Let's say we continue to operate the engine under detonation for a minute or two and then retard the timing to the intial setting that did not produce detonation, but we find that the engine continues to detonate. We know that the combustion chamber surfaces have significantly increased in temperature, because the shock waves produced by detonation increase the rate of heat transfer to the combustion boundaries about an order of magnitude (factor of 10), and a "hot spot" has developed such as a hot exhaust valve or hot carbon buildup on the piston crown that ignites the the fuel before the spark discharges, which is equivalent to having advanced timing. This "preigntion" results in normal combustion for part of the charge but then it detonates before it is consumed by normal combustion. In original case we initiated detonation from excessive spark advance, the subsequent overheating of the combustion chamber boundaries got the engine in to preignition, which perpetuated the detonation. This engine will have a failure before I complete this post. It's like a flat spin in and airplane. There is no recovery unless I shut the engine down or remove the load.

Ever had an engine loose all its coolant and begin knocking. What happens is the loss of coolant overheats the combustion chamber boundaries, and a hot spot becomes a preigntion source. Since this is equivalent to advancing the timing the engine begins to detonate. We know when detonation occurs. We can hear it, but we don't necessarily know if it's initiated from too much timing, insufficient octane, or preigntion.

As I said, they are different, but closely related. Detonation can lead to preigntion, and preignition almost invariably leads to detonation. Preigntion, too much timing, or insufficient octane are causes. Detonation is the effect, but preignation can perpetuate detonation that's initiated by one of the two other causes.

An important point to understand about detonation is what causes the damage. It it NOT the mechanical impact of the shock waves. It's the heat transfer. As stated above, the rate of heat transfer to the combustion chamber boundaries increases by up to an order of magnitude with detonation. This will rapidly increase the temperature of the piston crown and exhaust valve, then they weaken and fail - a hole in the piston or a chunk of the exhaust vavle breaks off. These hot surfaces can also act as preigntion sources, which is the equivalent of advancing the timing even more, so the detonation becomes worse i.e. a greater percentage of the fresh charge is consumed by detonation rather than normal combustion.

If you're interested in the workings of an internal combustion engine at a low level of detail, it's important to understand both the difference and relationship between preignition and detonation. Chew on this post for a while and ask more questions if this doesn't explain it sufficiently.

Duke

SWCDuke

"Dieseling" otherwise known as "run-on" after the igntion switch is turned off is caused by pre-igntion. It is a particular problem on seventies vintage carbureted engines. Back then there was typically no vacuum advance at idle and very little intial timing. The purpose was to keep the exhaust gas and combustion chamber hot to encourage oxidation of HC and CO, but those hot surfaces can cause preignition, and the preigntion can lead to detonation.

Sometimes with run-on the engine just keeps bumping over, but sometimes this is accompanied by that knocking sound. In this case the pre-igntion is causing detonation. Run-on is not a problem with modern cars because when you turn off the key, the injectors don't open, so there's not combustable mixture to cause run-on.

With sufficient octane that results in normal ie. detonation free combustion, good combustion chamber design, adequate water jacket design, and timing that is not compromised by emission control requirements, the chamber should not develop hot spots that can turn into preigntion sources, however, engines with a lot of deposit buildup can develop "octane appetite" - they need more octane than when new to keep them out of detonation. The deposit buildup can increase CR enough to create detonation under certain operating conditions, and this can led to preigntion, which exascerbates the detonation.

Run-on was more pronounced if the engine was shut down immediately after operating at heavy load. Also, those old engine would often detonate under moderate load, such as half throttle acceleration or climbing a hill after idling in traffic. You don't need a preigntion hot spot to cause detonation. The hotter the combustion chamber walls the more likely the engine is to develop detonation, because of increased heat tranfer to the unburned charge. If it reaches the autoigntion temperature before the flame front arrives, you have detonation. All combustble fuel-air mixtures have an autoigntion temperature, but higher octane fuels have higher autoigntion temperatures, so they are more resistance to detonation. Higher octance fuels are more resistant to detonation regardless of the original cause of the detonation, - overadvanced timing, pre-igntion sources, or increasing the temperature of the fresh charge due to head transfer from hot surfaces. If you have a detonation problem and cure it with higher octane fuel, you don't necessarily know what caused it. Could be too much timing, a pre-igntion source, or heat transfer to the fresh charge. Because higher octane fuel has more detonation resistance, it will eliminate the detonation regardless of the source.

The sharp edges on pistons are not ideal, but they're probably not as bad as most folks think. I would not attempt to round them off with the engine assembled because of the debris contamination. The carbon layer that builds up on pistons actually insulates them a bit and the carbon formation won't follow a sharp edge, so it ends up more rounded.

If you want to extract maximum power, like on a racing engine, you need the highest possible CR so smoothing these corners and the combustion chamber in the heads is a good thing to do, but, remember, racing engines are frequenty torn down, so they don't get much carbon buildup. For the street, you have to come up with a CR and octane rating that has some margin, so it won't have a detonation problem as it builds up deposits with mileage.

Compression ratios have increased about one point for both regular and premium fuel engines in the last 10-15 years, but commercial fuel octane has not changed. Regular fuel engines have gone from 8.0/8.5 to 9.0/9.5 and premium fuel engine from 9.0/9.5 to 10/10.5. The reason for this is knock sensors. The engine will handle lower octane fuel for a given CR because the knock sensor will retard timing if detonation is detected. This has helped fuel economy, but under many operating conditions the engine is on the ragged edge of detonation, which is okay, and best for thermal efficiency. Knock sensors allow this, but with an older engine that doesn't have a knock sensor you have to come up with a more conservative igntion advance map to keeps the engine out of significant detonation.

Duke

MoMo

Thanks Duke. You're very knowledgeable and your posts really helped to clear this up for me.

I have an engine that fits the profiles you've mentioned. It's a 355 engine that was carbureted with iron heads and 10:1 compression. But because I live at 5,800 feet altitude, I never experienced a problem. I knew I was on the ragged edge though, because if the engine warmed up a lot, it would diesel a little on shut-off.
Also, I took this car down to low altitude, and had all kinds of problems. Octane boost with 93 octane pump gas did not help at all, and I had to manually back off the timing until I returned to high altitude.

Now I've pulled this engine out of the Corvette and bolted a tuned port up to it...and a knock sensor on the block...and installed it in an IROC Camaro. The computer keeps things happy. The only time I experienced slight knocking was when the fusible link to the cooling fan burned out and I had to take it home to repair. It didn't overheat, but it did get warm, and I drove it very gently. I ended up getting an adjustable aftermarket thermostatic switch that I could set at about 190 degrees, and that helps a lot. The computer wouldn't turn on the fan until 226 degrees, which is a little warm for a 10:1 iron headed engine.

Thanks very much for your insight. It helps a lot. You're an asset to this forum.