IW STi Forum banner

Giving up place in line due to detonation

5646 Views 104 Replies 24 Participants Last post by  darwood
How many of you are contemplating passing on the STI, when your turn comes up, if the detonation "problem" hasn't been adequatley addressed by Subaru?

(I hestiate to say problem, because it is not all STIs, but w/ only a few hundred in circulation, it seems widespread enough to be a legitimate concern.)

See following links if you aren't up on the detonation issue.



(see 2004 STI Tech 101 section, @ bottom left)
1 - 5 of 105 Posts

I am quite concerned with the pinging issue, and as such I've delayed my STi purchase for 1-2 months. I am waiting for other tuners (like Shiv) and experts to weigh in on this one, as I've yet to see any CA cars on the dyno with professional ears at work. One thing I am sure of is that asking owners to determine if they hear pinging is not the best way to see if there is a problem. The only way to be sure is to dyno 5-10 cars in a weekend and get some real statistical data on the subject. Since I am convinced that there is indeed some kind of issue, I will wait to learn how widespread it is and how easy it is to fix.

To turn a point made by another post on its head, what if your STi is one of the ones that pings? Even if it's a minority of the cars, it still means more risk in the future. How long will the motor last? What will it do to the resale value? How does it impact warranty claims? I don't relish the idea of taking the car in with a toasted motor and saying, "well, the car sure pinged alot". They'll claim bad gas, abuse, acts of god, and everything else they can to not fix it. I like gambling, but not with $31k.

I'm waiting to see how this pans out over the next few months. Look at the anguish of first-year E46 M3 owners. I don't want to be in that boat.

See less See more


For the best data we have so far, I'd point to the poll at NASIOC


The way I read this, 25% of the poll say thier car pings. Another 21% aren't sure. The poll isn't scientific, but another way of looking at it is that 35 people have reported ping in their STi's. So it's not just one or two cars.

You shouldn't read too much into intenet polls, but it's pretty clear that there is indeed some kind of issue. BTW, a guy in SoCal has posted a potential workaround--I'll wait for the SOA version.


See less See more
Hmmm, these definitions aren't getting to the point of what detonation/ping/knock is, and why it's a bad thing. This is from memory, but it's 95% correct:

The most basic meaning of the word detonation is an expanding cobustion zone with a supersonic flame front. This is opposed to deflagration which is a relatively slow, controlled, and gentle combustion. The latter is what a spark-ignition internal combustion engine is supposed to do.

When a fuel-air mixture detonates, the combustion process happens much, much faster than normal which produces extremly high combustion chamber pressures. Also, the supersonic flame front creates shock waves which can shatter things like piston ring seals and other fragile components. This is also what produces the audible ping or knock assoicated with detonation.

Detonation can be caused by several factors. These include spontaneous combustion or "pre-ignition" where the fuel-air mixture burns without a spark, and sometimes in several places at once. This is caused by abnormally high intake charge temperatures, very lean mixtures, or high compression ratios.

Another way detonation happens is when the spark fires too soon in the piston cycle (i.e., the spark is too "advanced" in timing). This causes the maximum cylinder pressure to occur as the piston is moving upward in the stroke. This again causes pressures that are too high, and the unburnt mixture will spontaneously ignite. As the flame fronts meet, detonation occurs. (For reference, under normal circumstances it takes about 30-50 degrees of crank angle for combustion to happen. During detonation, this can happen is only 5 degrees. BANG!)

You can think of detonation as bad like this: if a normal combustion process expends it's energy in 1/50th of a second, detonation does the same in 1/500th of a second. Think of it as a 150 man stepping on your toe with tennis shoes. Now try a 150 woman with high-heels. (That's a force/area problem, not exactly the same, but it's not a bad analog.)

Detonation can also be induced by "hot spots" caused by poor head/piston crown design or carbon buildup, although this usually causes issues like dieseling where your car keeps running a few seconds after you shut the key off.

It should be noted here that diesel engines are *designed* to work this way--they are a "compression-ignition" system that actually uses the high compression ratio of the combustion chamber (as high as 24-1, unlike your STi at something like 9.5-1) to ignite the mixture spontaneously. Diesel motors do not use spark plugs to ignite the mixture. Diesel combustion chambers are designed to promote ingition from a single location to avoid harmful detonation, but as anyone who has been beside a diesel motor before, it actually sounds a lot like your car when it's knocking.

I did a few searches for visualizations of combustion chamber detonation and found this site. It's not a bad write up:


This is why "even a little knock" under WOT can lead to cumulative and significant damage. It's also why it's imperative that Subaru fix the problem. Having your Geo Metro knock at 750 RPM is very different than WOT at 4500 RPM. If this problem is not fixed soon, I predict that we will begin to see a large number of complete engine failures.

Just my $0.02.

-I think I will repost this on the top level.
See less See more
basic, "essential" mods for STi...


For what it's worth (and I've been one of the more vocal critic of the detonation problems) I just put a deposit down on an STi today. I am betting you that my car is one that pings, and if it does I plan to do the following:

1) Bring it up every single time I have the car in for service and get a signed statement from the dealer that they could reproduce the problem

2) Write a letter to Subaru

3) Call their customer service hotline and escalate it quickly to a manager who I will force to send me a letter detailing Subaru's response to my concerns

4) Wait for Subaru to fix the problem or my motor to blow up

In the final analysis, Subaru simply can't afford to have their cars blowing up. It's awful PR. Look at what happened to the Mazda RX7, quite possibly the best production sports car for under $75,000: reliability issues killed that car, and Mazda lost a bunch of cash (in the US market, anyway).

I know it sucks, and I am very disappointed in Subaru, but when they fix this issue you'll have a very special car. Believe me, it has taken me long hours of soul-searching to plunk down $32k for this car, but I think in the end it the benefits will outweigh the minuses.

The STi community will be VERY vocal if the ping issue goes unfixed.

Just my thoughts. Good luck with your STi,

See less See more
Both TurboXS and Vishnu have documented on their respective sites the following:

1) The ping is related to spark advance
2) The ECU does a very bad job of controlling the advance; it apparently never gets out of "learning" mode and essentially forgets where the last ping was and keeps plowing into it over and over
3) This does not seem to affect all STi's, but it has affected every STi that TurboXS and Vishnu have tested so far
4) It is unclear if simple tuning maps will fix the problem. This is probably deeper in the code and will require a completely new software load from SOA
5) This problem has been seen across the nation regardless of fuel rating/type, environment, etc and always seems to happen from 4000-4500RPM under WOT, especially under heavy load (e.g. up hills)

1 - 5 of 105 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.