With all of the questions lately about detonation, and after personally breaking two piston ring lands, I was very curious as to why this phemomenon happes, as i am the type of person that has to know why things do what they do I will discuss what I have learned recently, from many web sites, and people I have talked to. Most of this is fact, although some may have their own opionions, which is excellent for discussion. Lets make this an informative post, and lets all learn what we can do to prevent detonation. I'll start off by a technical definiton of detonation.

Detonation is basically an uncontrolled, uneven burn of the air/fuel mixture. Ideally, the mixture, will be ignited by the spark plug, at the top of the mixture. It will then ignite, and burn outward in an even, controlled manner, to the end of the mixture, closest to the piston, also called the "end gas". Detonation is when the top of the mixture ignites, and then the middle or the end gas ignites suddenly. The two flame fronts collide, and cause an EXTRMELY high cylinder pressure, which makes the cyliner walls and other components go into oscillation, causeing the "knock" or "ping". Factors that can promote detonation:
Increased cylinder pressure( ie. more boost pressure)
Lean air/fuel ratios
Increased intake air temps ( reason why an IC is so important)
Advanced igniton timing
Over heating of the cooling system (insufficient cooling, esp the radiator)
Low octane fuel

-Increased cylinder pressure: This happens when we run more boost.
The more boost we run, the greater the chance for detonation to
occur. Higher compression engines will have a much greater chance
for detonation, because there greater cylinder pressure (pretty
much why most will advise going low compression and turbo)

-Lean Air/fuel ratios: One of the greastes causes of detonation. Lean
A/F caused a very rapid heating of the cylinder and cylinder
pressure, causing detonation.

-Increased intake air temps: Again, this raises the temp of the air
charge, makeing detonation more likely. Thats why an IC is so
important to use no matter what boost you run. The cooler the intak
air is, the less likely detonation is to occur

-Advanced igniton timing: This again causes very hot air temps in the
cylinder, which caused the end gas of the mixture to ignite, that
= detonation

-overheating of the cooling system: Our cooling system IMO, cannot
sufficiently handle 15+ psi of boost effectivly. The more boost you
run, the more heat generated. And the more heat, the greater the
chance of detonation. A fluidyne or better cooling radiator is a
def good choice. When the coolant is too hot, it keeps the
cylinder walls too hot, thus making a greater chance for detonation

-low octane fuel: Dont run it. When i detonated, i was running 93
octane. My A/F was a consistant .93-.94 at WOT. The higher the
octane, the slower it will burn, which is perfect to help deter
detonation. Even with 93, i detonated because of too much timing
advance.

So what can us turbo and N2o guys do to get rid of this engine killer?

From my own personall experience, any boost pressure over 8 psi, you will need to retard igniton timing. How much, well, that is still up in the air. Dyno tuning will be the best if not the only way to dial ignition retard in. Timing retard is THE best step to take, and an imperative step to take when turboed. This is esp true for stock internal turboed people, as stock cast pistons will easily crack ring lands off of them with just enough detonation. Lower the compression ratio of the engine. I have gathered that 8.5:1 is a general rule of thumb, some may digress. A higher compression engine will need more ign retard to prevent detonation. More retard= less power. Upgrade your cooling system. Get a fluidyne or similar radiator to help coolant cooling. Be sure your fans are working properly. Get a 160 degree thermostat, this will help lower the engine temps slightly. Get the drift so far? Lower intake air temps, retard timing, keep the engine from getting to hot, and run the highest octane fuel you can.

Igntion timing: When I get my engine built, this will be the MOST tuned part of the engine. How much retard, and when to dial it in, is THE key factor to getting no detonation, regardless of how much psi of boost, and how much retard with each pound of boost. Many use the Crane unit, and it works, but is limited to 15 psi. To run more boost than that, you will have to pull out too much timing at 15 psi, to run lets say 20 psi. I, as others, have brought up the product J&S safeguard. Technically, this will work wonders for our cars, but could be rather "touchy". It uses an aftermarket knock sensor to detect detonation, and will retard timing on an individual cylinder basis, becuase it will know exactly when and which cylinder just detonated. The J&S will keep ignition timing to the "right before detonation" point in time in the fuel burn, and this has been proven to be the point in which max HP is made, Right before detonation happens. Hopefully now we can discuss detonation and igniton timing in more detail, from those who are boosting using igniton retard units, and what they are doing, how much power they are making, ect.. Hopefully we can all ask questions, and try and learn from eachother here, as no one wants to blow an engine any more than the other guy. The key for all of us should be to retard the timing just enough to not get any detonation. This is where max HP is made. And another side note, detonation does not occur in every cylinder. All the factors listed above can be slightly different, which is why all-cylinder retard can hurt power. Unfortunately, my ignorance on this topic broke my piston Knowledge is power !

13.5 @108 MPH-2.2 60ft(stupid FWD!)
S16G @ 18 PSI/FMIC/Running on MegaSquirt II (Now with sequential
fuel injection)
My webpage: http://eclipsed4evr.home.comcast.net
-1998 Mitsubishi Eclipse RS-T- "Toy"
-1992 Plymouth Laser Turbo AWD(SOLD)
-2000 Honda CR-V(daily)

I agree with you, Matt. Detonation seems to be the big engine killer for us on this forum. It will be in everyone's best interest to find out and experiment with the best method and product to control this. Although I have built internals, I'm always afraid that it could happen again. I believe we've come up with various fuel setups that work great, now let's concentrate more on detonation control. Of course, proper and sufficient fuel setup will always be a major issue.

___________
Kon C

"Sometimes, a man has to do what a man has to do."

I second that one. I just pulled 3 bad pistons from my car. I'm begining to wonder if my cartech FPR is about done for. Either that or I have it hooked up wrong.

Kevin
The old:


The new:

Good idea for a thread Matt. I like that we are starting at the basics here too. So I have a basic idea that I want to hear your comment on as well. One of the things I'm trying to figure out is some kind of base algorithm for spark retard/advance based not only on boost, but rpm. My understanding is that while retarding is good for boost situations, higher rpms require a bit of advance over lower rpms. This is because there is a certain lag time between the spark firing and ignition of the mix in the cylinder. In the case of a lower rpm (and thus slower piston speed) the spark should occur late so that the fuel ignition is in sync with the downstroke. In a higher rpm scenario, you want spark to occur somewhere in the upstroke so that by the time ignition occurs, the piston is starting the downstroke. If timing was kept the same as in low rpms, the piston would be well into the downstroke before the igniton wall hit the piston.

If the timing is not advanced enough, we won't have maximized efficiency and power. Although it's worth sacrificing some hp for safety. Anyways, I was wondering if these ignition systems you guys are running are taking into account rpms at the same time boost. It would be great to just have an equation like f(psi,rpms,intake charge temp)= to begin with. Then I could make an xy coord graph with x-axis as rpms, y-axis as advance/retard, and several different plots on that graph of different psi's. (it'd be a good visual for this kind of stuff) Then we could tune our cars from that starting point, since each engine will be somewhat different.

Oh yea, and who had J&S safeguard on this board?

2gnt: '99 RS-T, killed by a toyota, pending rebuild...
Daily: Volt
Daily #2: '99 EVG ebike- STOLEN by PEDRO

Excellent info there. You are completly right about the timing with low/ high rpms. Lets take an example first of our engines without boost. Our ecu's adjust ign timing based on many factors: rpm, intake air temp, coolant temp, TPS postion, and engine load, and MaP sensor readings. A datalogger will be the most benificial tool ever IMO. Our ecu will advance timing under vacum conditons constantly. When under a minimal engine load, timing will be advanced quite a bit, up to 30+ degrees. When we punch the gas, ecu goes into open loop, and timing is pulled back to around 15 degrees, and may go up slightly (at least this is what i noticed on the datalogger). I strongly believe my episode of my destroying detonation happend in closed loop, part thottle boosting. I hit almost 10 psi in third gear at around 3500+ rpms, with 80% or slightly less throttle. Then I heard the nasty rattling. Then i came to a stop light, and was smoking from the hood out the breather catch can filter i have installed. Why did this happen? In closed loop, the ecu will advance timing to where it gets the most power, and the least amount of emissions (adjusts the A/F ratio to stoich). I think closed loop part throttle boosting is a major concern, as our ecu's don't KNOW we are boosting! Even more improtant for spark retard in this situation. IMO, it will be VERY hard to tell how much timing should be advanced or retarded with rpm alone, as engine load is the key factor to how its done by the ecu. WOT boosting is much easier to come up with, as we can do this with units like the Crane or MSD. Retarding with rpm will most def hurt power over all rpm ranges if used that way. Retarding with boost only is a better way to go IMO, for better power production. Thats what is so nice about factory turbo charged cars. Their ecu's can read boost, and adjust the timing for the load, knock amount, and boost. We need an aftermarket device to simulate this. But now, WE have to figure out how much timing, when to retard, and with how much boost. As rpm increases, the engine load decreases,which leads me to believe most of our engine damage happens at a lower rpm, where timing is the most advanced,with the greatest load, with higher boost. There will be more load in different gears, at diff rpms also. Example, there will be more load at 2000 rpm's in fifth, then at 2000 rpm's in third. Same is true for WOT runs. You will have more engine load at WOT in 5th gear, than at WOT in 4th gear. I noticed this just by reading my o2 voltages on WOT runs. I got .98 v in second gear, and by WOT in 5th gear, i was around .92 v. My fuel pressures were rock steady and at the same psi in all gears. Increased engine load makes the car run hotter, and leaner, because more load needs more fuel. I only wish there was a way to tune our A/F ratios at WOT to engine load, to get a perfect .93v in every gear. At least start out leaner in a lower gear, and then richen it up for higher gears, exactly opposite of what happens now. To me, it just seems that we should not part throttle boost in closed loop, without timing retard, past 5 or 6 psi. WOT boosting, not going over 7 or 8 psi with out minor retard also. How much, well, thats what we need to figure out, to make the max HP we can anyway, without detonating. Some of the dyno numbers posted here, IMO, have been rather low. Esp with the guys running 10 + psi with rpm, and linear boost timing retard units. And also, i am basing my theories and numbers here running stock 9.6:1 comp ratio. Lower comp ratios *should* let us run a little more boost without any retard. Another thing that will make it harder to ignition tune is the A/F ratio. Richer A/F will have a tendancy to detonate less, as leaner A/F can detonate with greater possibility. This leaves a big window as to how much timing needs to be retarded. We need to find the *best* A/F ratio to start with, then tune the igniton retard. They both go hand-in-hand..

I am searching hard for someone with a J&S Safeguard..no luck yet.. Their input and HP numbers with and without it would be perfect. I do plan on getting one of these, and hopefully, if funds permit, i will dyno tune the car at like 9 or 10 psi, with, and without the J&S. I'm sure some interesting HP numbers will result. If i had the 600 bucks, i'd do it right now, on the stock internals.

13.5 @108 MPH-2.2 60ft(stupid FWD!)
S16G @ 18 PSI/FMIC/Running on MegaSquirt II (Now with sequential
fuel injection)
My webpage: http://eclipsed4evr.home.comcast.net
-1998 Mitsubishi Eclipse RS-T- "Toy"
-1992 Plymouth Laser Turbo AWD(SOLD)
-2000 Honda CR-V(daily)

Xtreme, what are the drawbacks for the J&S other than cost? id be willing to try it on the new engine.

1997 Silver RS
Built
T3/T4 BB Turbo
20psi
Most current 1/8th mile time: 8.1@93mph 2.015 60'
1996 Black Supra TT
2004 Silver Snake

>Xtreme, what are the drawbacks for the J&S other than cost?
>id be willing to try it on the new engine.

I have spent some days here researching more about the J&S, looked for testing done on it, ect. It seems it came about for the RX-7 guys. A lot of them use it, and they have relayed info back to the devloper/builder of the J&S. They have come out with the 3rd gen prodcut now, and is pretty versitile. THE most imortant drawback that I can see from this unit, is the sensing of the actual knock. Every engine has a different knock frequency, but they all have a general frequency range of which knock has. Noisy pistons, and other engine noises can confuse the J&S, and cause uneeded timing retard. I read a post or two like this when i was doing my research. This was with the older units also. The idea is great, and for some people who use it, it works excellent for them, has saved their motor many times. Some have probs with it. Its a toss up really. There were more positve comments on it than negative, and most agree that it works at saving their engines from the destructive forces of detonation. I'm still researching the whole topic, and trying to find more info from actual users of the J&S. As of now, I still plan on buying one. Whats another 600 bucks?? hehe...oh well..

13.5 @108 MPH-2.2 60ft(stupid FWD!)
S16G @ 18 PSI/FMIC/Running on MegaSquirt II (Now with sequential
fuel injection)
My webpage: http://eclipsed4evr.home.comcast.net
-1998 Mitsubishi Eclipse RS-T- "Toy"
-1992 Plymouth Laser Turbo AWD(SOLD)
-2000 Honda CR-V(daily)

It seems to me that the absolute best solution to controlling detonation would be a full standalone engine management system, ie Haltech or TEC3. With the Tec3, you can retard timing based on rpm/boost,even adjust individual cylinder timing. Granted, it is pricey at about 2500 fully installed, but when you figure 500 for a boost controller, 400+ for an ignition amplifier, 500 or whatever for a boost retard box,500 for a datalogger, you start to see the value of having all those features in one box. Couple that with the tunabilty of a wideband 02 and the advantage is clear. This is DEFINATELY the way i'm going to go once i go turbo.

**************
Fucking Kangaroos.
**************

I am quite interested in the J&S as well. The Miata guys have been using this product for some time now. We may want to head over to some Miata forums and talk to them.

I got this froma Miata forum.

"Normally, the J&S Yellow wire taps into the airflow signal. The unit is then armed to detect knock when sufficient airflow is achieved.

The mass air signal on 1.8L engines increases with increasing airflow. The J&S becomes armed when the voltage exceeds 2.75 volts.

On a 1.6L, the voltage decreases with increasing flow. The unit is armed when the voltage drops below 0.8 volts.

When you switch to a Link, you need to set the J&S up as a 1.8L, and connect the J&S Yellow wire to the MAP signal. The J&S will arm when the MAP voltage exceeds 2.75 volts, and the RPM exceeds 1250.

Temporarily disconnecting the Yellow wire will allow the unit to self arm, but to test for knock retard, make sure the RPM exceeds 1250."

What voltage dose our map give? If we use a Missing Link will our map be able to activate the J&S?

>To me, it just seems that we
>should not part throttle boost in closed loop, without
>timing retard, past 5 or 6 psi. WOT boosting, not going over
>7 or 8 psi with out minor retard also.

I tend to agree with you on this one. It's a big hurdle - using a non-turbo ECU with a turbo. Every turbo kit already clamps the MAP voltage or modifies it with something like the missing link. The open/closed loop timing curve is not addressd by any kit(like it is with the stock turbo ECUs). I wonder if we could fashon a sensor that would somehow 'clamp' the ECU to open loop after seeing 5 or 6 pounds...much like the device that Corbin designed to fool the S-AFC to run from boost pressure. Perhaps it could intercept/control the TPS voltage. When it sees 5 pounds and TPS<80% then clamp the TPS to 85%...but as soon as the actual TPS hits 85% then it relinquishes voltage control back to the actual TPS sensor. You can't just blatently hold the TPS at 85% at WOT, that would be no good. It would have to take boost pressure AND the TPS voltage BOTH as inputs and output an 'open loop' TPS voltage when nessicary. Would there be any drawbacks with a sensor like this? What are they?

As far as the retard at higher boost, you would still need a Crane or Safegaurd or something like that.


ride__________95 Eclipse RS : 5 speed
power________Gude cams : AFX UDP : ram air : test pipe : Thrush glasspack
suspention___BOMZ F/R upper strut bars : ES inserts
sound________Kenwood KDC 5000 : Audiovox 8" sub toob - no sub amp(damnit!)
shizzle_______AutoCommand remote start : dual air horns
incomming___Hurricane F/R lower strut ties : Jeep TB : Field SFC Hyper-R : GST catback

95 Eclipse RS : 5 speed
14.76 @ 94.72mph
Jeep TB writeup - http://www.dimensia.com:81/jimbo/JeepTBfor2gnt.html