Originally Posted by docmirror

I din't want to go all engineering on this, just to point out that pressure, not CR is what causes detonation.

Originally Posted by docmirror

I din't want to go all engineering on this, just to point out that pressure, not CR is what causes detonation. And pressure can be calculated carefully. Using the Lotus Turbo example again, it has quite a low CR of 8.2:1, but a very high BMEP. Depending on cylinder conditions,

Originally Posted by James Bailey

If that is true then your ignition timing is very advanced from where it is supposed to be.....Have you had a timing light on it ???

Originally Posted by WallyP

...why not tune the car the way Porsche specified, and use the fuel that Porsche specified.

We do a lot of bench racing on here.

Originally Posted by WallyP

A theoretical discussion, especially for those of us with non-adjustable ignition timing.

- few of us are actually squeezing the last bit of performance from our cars;

why not tune the car the way Porsche specified, and use the fuel that Porsche specified.

.

Originally Posted by WallyP

why not tune the car the way Porsche specified, and use the fuel that Porsche specified.

Originally Posted by SteveG

... What advance would be the recommendation for max HP in the lower RPM range? Also I'm getting the sharktuner. That will read detonation in real time, won't it?

Originally Posted by docmirror

Gosh, I wish I had time to cover this better, but I'll do what I can. Let's start with the audible 'ping'. this is a result of detonation inside the engine. It's also called pinging, and even pre-ignition by some folks. Detonation is destructive. It has a resonant freq in most all IC engines of 35-43Hz which is why it's audible.

The shock wave caused by detonation directly impacts the rising piston, and will damage any number of components. it is not affected by ignition timing(because ignition doesn't cause it), and is most affected not by C/R, but by something called BMEP, or Brake Mean Effective Pressure. I won't go into boring detail, just understand that the BMEP of an IC engine is a good deterministic way of measuring what power output an engine will 'tolerate' given the fuel in use, and the efficiency of the intake, exhaust, swirl characteristics and combustion chamber design. Here is the formula:

BMEP = 150.8 x Torque / Disp(CI)

So if we take the typical S4 engine and run the numbers we get BMEP = 150..8 x 317 / 305 or a value of 156.7 PSI.

Why is BMEP important you ask? Well, I'll tell you. C/R doesn't make an engine detonate, pressure and temperature make an engine detonate. This is true also of grain silos, gunpowder, and almost any type of fuel content. Grain can sit in a silo under hundreds of pounds of pressure(on the bottom) but almost no temperature increase without problems, but when the temp goes up a few degrees, suddenly you may have an issue. Now, back to engines.

We have a BMEP of ~157PSI in our S4. this isn't particularly high compared to other hi perf engines out there today. That's why the S4 can get away with 87 octane in many conditions. Now let's take the engine in the Turbo-chargecooled Lotus Esprit for example: 150.8 x 264 /146 for a BMEP of 267PSI! Yikes, never run low octane in your Lotus turbo.

It's pressure plus temp that causes detonation in a clean running engine. Next let's talk about some other factors. One in particular from the aviation world is mixture. We can get way with higher BMEPs in an engine if we simply richen the mixture a bit past stoich. The reason for this is the added fuel acts as a retardant to detonation. It's why piston plane operate at about 11:1 FA ratio when they take off under 100% load factors. We need that extra margin of detonation protection at critical take off time. Later on, we have a control that leans the mixture drastically, improves fuel economy and engine efficiency, but it reduces detonation margin. We do not have this control in our cars, but they do it automatically with the closed loop O2 sensor which provides a decent approximation of FA ratio, or burnt to unburnt fuel. gummint regs require that we burn as efficiently as possible of course.

Our S4 example may be operating near stoich as it should, and the way we control ping(detonation)is with spark retard. But wait, I said it's not spark related didn't I? Yes, that's correct, detonation is not spark related. That's why you guys with sharktuners are seeing detonation detected even though you don't hear anything. It's because the same type of effect can be caused by pre-ignition.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Pre-ignition is simply the incorrect propagation of the flame front when the spark plug fires. There is a narrow and changing window of time for the spark plug to fire in an IC engine to insure proper flame front. it is affected by engine speed, mixture and MP. We have maps in our S4 for all these things, and we determine the optimum spark timing by monitoring the mixture(O2 sensor), speed(tach pulses) and MP(MAF in our case). All these factor go into a matrix and the injector pulse is calculated in the LH and the spark timing is calculated in the EZK. When it works right, it's a thing of beauty, when it goes bad, it goes bad in increments, usually not detectible by the operator.

You can't do much of anything about pre-ignition, except use the proper fuel for the engine design parameters. Lower octane will sometimes have an affect on pre-ignition, but not always. There are cases where we use the engine management to get us up to optimal advance, by driving the engine into pre-ignition and then backing the timing off slightly to allow the flame front to propagate correctly and not allow pockets of mixture to spontaneously ignite. Pre-ignition is not as detrimental to engines, and is usually solved by monitoring the head(knock sensors) and retarding the timing.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Detonation and pre-ignition are sometimes found together. The same BMEP that causes such trouble with detonation, also causes pre-ignition problems, so we can have them independently, or we can have them coupled. If pre-ignition gets too bad, it can also become detonation, and then you'll hear the ping, which is the case where retarding the timing may help.

How to avoid this? In our older engines, that have been around the block a time or two, we develop carbon deposits on the chamber. these deposits can retain a lot of heat, and even have a glowing ember in the chamber as the compression stroke comes up. this is a very common activator for detonation, all it takes is a slight hot spot of an ember, and the pressure will set it off. Next, is the wrong heat range plugs may allow the ground lug, or even the ignition tip to start glowing, and it acts as a glow plug, just like old time diesels. Note that carbon deposits do another thing. They raise the BMEP of the cylinder by displacing the air space, thus increasing the pressure of the charge. It could be related to one injector that has been firing a bit too rich for a year or more and while it's making rated power, it is also loading up the cylinder with carbon deposits. The injector can also help to cause pre-ignition because it is not stratifying the fuel charge on the intake stroke well. In the 928 the low BMEP is partially caused by the crappy intake geometry all the way from the hot air intake over the radiator, down to the bends and kinks that we take to get to the intake valve. The swirl characteristics in that intake must be abysmal but I have no data to back that up, just empirical evidence having taken our intake system apart a few times.

That's why some 928s will run fine on 87, and some will ping on 91/92. Try a different injector set, or spark plugs, or if you're really ambitious remove the heads, decoke them and the pistons, and valves then put it all back together with new gaskets, plugs and injectors. Leave the battery off the car for a few minutes, and let the LH/EZK go though it's normalization learning curve with 87 octane and it might run fine, at least in winter, but you can try it in cooler climes in summer and see what happens.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Lower octane gas has a slightly higher mass efficiency, and similar thermal efficiency to high octane. That is, you'll get slightly better fuel economy with lower octane, but the difference may be so small as to be undetectable. also, your engine may retard the ignition to avoid pre-ignition thus rendering any potential slight gains moot. I will go out on a limb and say that any US pre-85 engines will likely run ok on 87 octane. The CIS cars must be calibrated for it, but it should work. Post 84 US and any euro is going to be iffy. The S4 seems to get away with it due to it's low BMEP numbers, although not all engines, under all conditions seem to work. YMMV, no guarantees, warrantees expressed or implied, written or verbal, do not try this at home, professionals on closed course, all factors subject to change, no take backs or do overs.

<edit: I just recalled that I used regular in my 86 for a while with no ill effects, and I calculated the BMEP and it's lower because the torque of the 86 is lower than the S4 engine. so, in retrospect, lower octane will prolly work in the US 85/86 cars too. I know it worked for me in TX in the winter.>

Originally Posted by Markmandude

Speaking of altitude, we are at a mile high here in Colorado and very few gas stations sell above 91 octane. Anyone know how much lower octane is safe at a higher altitude? We loose about 1 second in the quarter mile up here for most cars.

Originally Posted by Bill Ball

The stuff above is what I meant to write, but I didn't know most of it. I'll write it next time. Thanks Doc!

Anyway, I'm satisfied 87 octane is OK in my S4 (when in essentially stock configuration). I was before, and now I think I understand why. Maybe not.

Originally Posted by dr bob

Lower air pressure means lower cylinder pressure, so generally there's no benefit to high-octane at altitude. If you fill in denver and fall off the edge of the plateu it can be a different story.


On the original question: Premium has been 20 cents more than regular here for a long time, maybe sine gas was a buck or so per. A 20% 'premium' made the choice a lot tougher than it is now. My car suffers noticeably with regular in the summer. Tried it once early in the ownership period, heardsome low-rpm noise just rolling from a light, and immediately topped with good gas. Detonation is a silent killer, a demon that sits on pistons and taps gently with a ball-peen hammer. Even if nothing appears damaged now, the effects accumulate and somehow come together later catastrophically. The bearings in my car looked new ehen inspected at MM/OPG time a few years ago. Cylinder bores look new per Rob Edwards as he peeked in during the intake refresh party a few months ago. So good gas is a must, IMHO.