Detonation and stock L28 NA

[Leon]

Gollum,

the color I've chosen for detonation has nothing to do with its severity, I've done it this way because it looks nicer Color transition brings confusion, I did not think about that. My bad.

The chart is true at one only specific condition. When I said: "Since the chart above is showing only one state of the engine (given RPM, Temperature, engine load, etc.)" I meant that one particular chart exists for each unique condition (i.e. every cell in a timing map for instance with defined AFR + all the intermediate conditions in between cells).

 

Do you suggest also that torque or detonation value would not be so repeatable? it will hardly be repeatable since other inputs might affect the reaction. That's why, I believe, modern car such as Evos (the only modern EFI car I've played with) for instance have a complex formula for engine loads.

 

Looking again at my chart, I still think detonation threshold should be horizontal, here's how I see the physics behind:

1- going from low timing to higher one before detonation, pressure will increase with ignition timing

2- ...until it reaches too high comp to avoid detonation

3- too high timing, past max potential torque, ignition occurs too early; pressure from ignition increases + additional pressure from piston motion is enough to create compression & detonation. Still, energy from combustion will try to slow down the piston. As a result, you loose work & gain calories for the engine.

4- too high timing, past detonation threshold, ignition occurs way too early, pressure from ignition will increase but additional pressure from piston motion is not enough to create enough compression. Energy from combustion will try to slow down the piston even more. As a result, you loose more work & gain more calories for the engine.

 

This is how I see the chart. If what I'm saying is wrong, I'll be glad to hear a different version (even if Tony is the one doing it )

I'm not following #4 here.

 

You're not going to have "too high timing, past detonation threshold" unless you're firing the plug seriously early, something like 120deg BTDC, in which case it's like having your plugs in the wrong spot and the engine won't run at all. Otherwise, spark knock is a function of spark advance and the earlier the knock (whether by spark or hot spots) occurs the worse it will be.

 

I'd make the threshold vertical.

[s30red240z]

Well then consider 500rwhp the foreseeable limit until you guys prove otherwise!

 

Thanks, i think is a good number for break transmissions and R200 non LSD XD

[helghast7]

Wow, I really must say I love this thread. The way you guys really broke down detonation to a scientific level is truly enlightening and honestly up until this point, besides the odd mention of L series characteristics, its really universal knowledge which is great to read no matter the engine. I really feel like I've learned a great deal from this about proper tuning and I hope to read more useful information in time.

 

Thanks guys.

[Gollum]

Lazeum ~ (not quoting since Leon already did on this page)

 

The reason the line should be vertical instead of horizontal is because detonation in and of itself, has nothing to do with torque output, or even timing in all reality. But you can't say there's a given torque limit for detonation for any given setup, or at any specific map point. And also, we don't adjust torque output on an EFI map, so it doesn't even make much sense from a tuning standpoint.

 

What we do adjust, that can bring on detonation, is timing. The more aggressive the timing the more likely detonation will occur. You can have a motor that produces 60 ft-lbs of torque per liter (which is very low, run some numbers), and cause it to detonate, even on premium gasoline. You can then modify that same motor to get over 75 ft-lbs of torque per liter (more significant change than it might sound, again, run some numbers), and the motor can still be made to detonate, and will probably happen at similar timing points. Though not identical points they'll have a much closer relation than the torque and power output of said engine.

 

The confusing thing about arguing about this trivial point (and it IS trivial) is that both torque and timing cause heat, heat causes detonation.

 

I think much more importantly than arguing about graph semantics is bringing to the table the fact that many engines can be brought to a condition in which at peak HP RPM ranges that have a detonation limit beyond max power timing. For example:

 

Take an engine in which detonation won't hurt it. Let's say a 250HP SBC. This I feel is a good example because there's thousands of similar motors out there, some even in Z cars. This motor doesn't need to worry about mild or even some moderate detonation, as it simply won't blow parts up. This isn't because detonation forces are a product of torque, but the reality of the fact that the more power you're making for a given displacement has a proportionate amount of air and fuel and you'll see similar increases in forces exerted to the internals, thus detonation causes more damage the high you go in power per liter. Poing being, this lowly 250HP motor can detonate mildly all day long and not care much at all. Now, as anyone who's driven an old 60's pickup can tell you, detonation will most likely happen at lower RPM. In fact, it's not too uncommon to be driving said truck, start going up hill and hear the engine start to ping, so you downshift and tada noise gone.

 

This isn't because the timing curve for said engine is a bad curve. It's more to blame for the fact that like the burning of fuel, detonation needs time in order for it to happen. Forumla 1 engines don't fret themselves with something as trivial as detonation because they literally out run it. Thus you should see that detonation is much more time specific than power specific. I've even experienced this phenomena in the honda world. There's a big shift just in moving peak power from 7,000 to 9,000. All of the sudden not only do you find that you're running more timing, but you find that you can keep a heck of a lot more of it as you tune down from your race gas. In fact, one engine that I helped a friend out with we didn't lose ANY timing at peak HP going from 110 octane to 91. I attribute this merely to the fact that the higher you get in RPM, the less detonation becomes a concern.

 

Now, obviously detonation breaks a lot of motors, so it's definitely a concern for most of us. I'm just bringing to the table that often it's because of aggressive tuning UNDER THE CURVE that's the issue. It might sound backwards to some tuners, but it's a reason that when finding the limits of timing I start from the top down, not bottom up. I also believe your tune needs to be fairly stable before reaching for limits, but I think when really trying to find those limits you're safer doing it at peak RPM than you are the very bottom off idle.

[Tony D]

1352163912[/url]' post='1027225']

Thanks, i think is a good number for break transmissions and R200 non LSD XD

 

I've said for about 23 years that sine I did it at age...uh...21 that getting 350Hp out of an L28 is child's play. They were EVERYWHERE mid-80's inJapan.

As I gained knowledge, and pushed it, I eventually made more, and 400 I would say with today's technology (and not a blow through triple Mikuni) should be about today's "easy equivalent".

That last 100HP, you BETTER know what you're doing in terms of tuning, or you will be going through ITM Pistons regularly.

Thing is, blowing five slugs out of your stock engine cots you $300... And you can do that QUITE A FEW TIMES tuning before you get to the cost of one forged, balanced, hyper built bottom ends.

My point was, the longer you "learn" on those Junkyard bottom ends, the better position you are to tune it when you put your expensive bottom end in there and tune that area between 7,000 and ????.

Some people realise when tuning to 7,000 and 17 psi that 500HP may just be enough for them... And then putting that forged bottom end in becomes longevity/optional... And you will have a 100% tuned map if you keep it at that.

Like I said all along, don't get greedy, don't do anything stupid...

[KillerBjt]

Gollum i think your forgetting one very important factor in detonation, cylinder pressure and the relation of ignition timing to the amount of cylinder pressure generated. If i remember from my High performance engine prep class correctly the earlier (to a point, i believe it was the intake valve opening/closing) the ignition the higher the cylinder pressure since the fuel has more time to burn before the exhaust valve opens which relieves the cylinder pressure. Then there is the other side where you wait too long to ignite the mixture which allows the cylinder pressure to ignite the mixture before the spark plug fires, resulting in pre-ignition and the fuel either having two flame fronts that collide with each other, dealing damage to the cylinder or burning as it passes by the exhaust valve causing damage to the valve stem and valve guide. Also if your running a turbo it could be burning in the exhaust housing causing damage to your expensive turbo. Its late, but tomorrow i will find my notes from the class and double check this post.

Edited November 7, 2012 by KillerBjt

[Lazeum]

Gollum, Leon,

I'm ok with what you're saying about detonation & timing, we're on the same page. Detonation is definitely related with time. During my studies, we were reviewing detonation & physics behind, detonation measurement is expressed with second as a unit.

 

It seems my chart is confusing things up. My attempt is not to tell that detonation has to be considered as torque related (even, if I admit, this is what the chart is showing) but it is to show that at one specific motor state, the more timing you put, the closer you could get from detonation- if we can even create some. It shows as well that detonation can limit torque output since we might have too much detonation before reaching top torque output (it would be called "knock limited" engine, wouldn't it?).

The detonation threshold can also be above max torque, this is what will happen when RPM are very high for instance. In this case, max torque output can be reached with no chance of detonation. This is what you've explained with Honda engine & I agree.

 

Another way of seeing the physics:

If you put the engine on a dyno, maintain its state constant (load, rpm, IAT, coolant T°, etc.) & you increase advance from a conservative starting point, we will see the torque output going up fast first & flat out the closer we get from max output. This is if the motor at this state is not knock limited.

If at the same state the engine would have been knock limited, we would have noticed some detonation before reaching max torque output. The potential of the motor at this particular state would not be optimum, we would have to leave some torque on the table to be safe. It could also very be not a problem if for instance this state would not be important.

 

We could also debate about the efficiency of the combustion with detonation, it will not be as afficient as clean burn, it would also impact the shape of the curve I've drawn...

 

 

As an example, on my particular car with L28 with centrifugal dizzy (Mallory Unilite) without vaccum advance & max timing at 3000rpm, I can clearly see that detonation would occur at max torque range / high load - around 3500rpm. After 4000rpm, there's no sign of detonation (I might not hear it also but I do at 3500). With what I have to avoid detonation, I could either retard total timing or push max timing above 3500rpm.

 

 

At the end, I put a threshold on torque output but I should have put a point on the curve, it would have avoided such debate

Writting on a board is more difficult than debatting around a beer, it brings confusion if my point is not crystal clear

[Gollum]

Well that's certainly an aspect to talk about, and will lead us getting into cam dynamics. But you also touched on an aspect that my posts should lead one to think about. Let me explain.

 

You you yourself (killerbjt, just so we're clear everyone) just mentioned dynamics of timing effecting BMEP, which of course will be a factor of air/fuel charge heat, which will directly effect detonation. Then you throw camshaft dynamics and that throws a whole different wrench into the machine. But let's talk about that in the real world for just a second.

 

It's been recommend by quite a few people on this very forum, that if you're going to run a high compression setup, especially one like the MN47 with flat tops, that you should be running a more aggressive cam to actually HELP with running pump gas.We could make this very complicated, but I really don't think it needs to be. Let's start at what the more aggressive cam will do.

 

1. Let's start with torque range shift. This is the most basic aspect that nearly anyone who's ever bought a cam should understand. They're even rated in RPM ranges usually. More aggressive duration cams are generally designed for higher RPM usage. By simple logic if you want MORE power at 6,500-7,500, then you're probably going to lose some torque off the line.

 

2. With the added duration and overlap you'll have different dynamic compression effects, which is basically a fancy aspect of what I said in #1 just now. Basically cylinder filling becomes less complete at lower RPM leading to lower BMEP, while it will have a benefit at higher RPM leading to higher BMEP.

 

So what did we really just do? I made our peak HP ideal timing much more attainable. We've lowered how much air and fuel we're trying to control at the lower RPMs were detonation can happen easily, and we've moved where we want the most aggressive timing/power output to a range that detonation is relatively less likely to happen.

 

Again though, it's hard to explain to people that haven't had experience with it, but I HIGHLY recommend experimenting with any form of alcohol as an exclusive fuel. Not methanol on top of gasoline, but straight meth or E85. With these fuels you can be running as high as 200hp per liter and never find detonation at peak torque points bellow 3500RPM (in my experience). It's actually quite amazing. Throw enough fuel in and you can play with timing so far that you'll drop off in power because your peak moment of energy is too early, yet there's no detonation. Once you see where the limit is, you start to see the real requirements of the engine very differently. You'll also stop fretting about getting every last ounce of power from a place the engine never lives at WOT.

 

In short, with my albeit limited experience, if I was experiencing detonation close enough to peak HP that I saw detonation during a drag racing row through the gears then I'd be looking for a PROBLEM. I'd assume either there's something terribly WRONG with my tune, maybe an inconsistent sensor. Maybe I had a hot cylinder. Maybe I've got a cylinder that's not filling nearly close to the others. Maybe I'm not running a close enough deck distance to my quench pads. Or maybe I'm just plain running the wrong fuel for my given engine setup.

 

Now, if I was 100%, or heck, even 80% happy with my engine setup, tune, and power and I experienced detonation outside of peak HP + or - 1,500RPM then I'd just tune it out. If I felt it was strangely out of character I'd start looking for the source of excess heat. Maybe I've got some heat soak issues in my engine bay. Who knows. But it'd have to be a strange onset of detonation for me to be concerned.

 

And like Tony said. These are things to learn on stock bottom ends at lower power levels. You don't just build a forged bottom end and shoot for the moon. Start small, work your way up. You'll learn where you need timing and where you don't where it's safe. You'll learn what to expect each time you add 2psi of boost and you'll learn how to watch your intake temps and tune for them. By the time you've got the hang of it hopefully you won't break too many bottom ends on your way to your end goal. I know for my dream build I want to shoot for 400-450 rwhp which I personally think is a bit conservative and should last a good long while. I also plan to do it with as many stock parts as possible regarding the bottom end and head. I also think I'll have very mild looking timing curves because hopefully I'll have done my work right and have a head that doesn't require a ton of timing....

 

 

Oh, and I almost forgot. This article is worth the read and I feel pertains to this thread, though indirectly to the author's subject.

http://www.davidandjemma.com/mazda/FAQ/quench.htm

 

Also, killerbjt, hopefully you saw what I was doing and didn't mean to say you were wrong in any way. Just that the end point of what you said still correlates to the points I've been trying to get to people that I think have the willingness to listen.

Edited November 7, 2012 by Gollum

[Gollum]

Gollum, Leon,

I'm ok with what you're saying about detonation & timing, we're on the same page. Detonation is definitely related with time. During my studies, we were reviewing detonation & physics behind, detonation measurement is expressed with second as a unit.

 

It seems my chart is confusing things up. My attempt is not to tell that detonation has to be considered as torque related (even, if I admit, this is what the chart is showing) but it is to show that at one specific motor state, the more timing you put, the closer you could get from detonation- if we can even create some. It shows as well that detonation can limit torque output since we might have too much detonation before reaching top torque output (it would be called "knock limited" engine, wouldn't it?).

The detonation threshold can also be above max torque, this is what will happen when RPM are very high for instance. In this case, max torque output can be reached with no chance of detonation. This is what you've explained with Honda engine & I agree.

 

Another way of seeing the physics:

If you put the engine on a dyno, maintain its state constant (load, rpm, IAT, coolant T°, etc.) & you increase advance from a conservative starting point, we will see the torque output going up fast first & flat out the closer we get from max output. This is if the motor at this state is not knock limited.

If at the same state the engine would have been knock limited, we would have noticed some detonation before reaching max torque output. The potential of the motor at this particular state would not be optimum, we would have to leave some torque on the table to be safe. It could also very be not a problem if for instance this state would not be important.

 

We could also debate about the efficiency of the combustion with detonation, it will not be as afficient as clean burn, it would also impact the shape of the curve I've drawn...

 

 

As an example, on my particular car with L28 with centrifugal dizzy (Mallory Unilite) without vaccum advance & max timing at 3000rpm, I can clearly see that detonation would occur at max torque range / high load - around 3500rpm. After 4000rpm, there's no sign of detonation (I might not hear it also but I do at 3500). With what I have to avoid detonation, I could either retard total timing or push max timing above 3500rpm.

 

 

At the end, I put a threshold on torque output but I should have put a point on the curve, it would have avoided such debate

Writting on a board is more difficult than debatting around a beer, it brings confusion if my point is not crystal clear

 

Well I don't think I ever once really disagreed with YOU, just semantics which makes for good talking points.

 

What you say is true. If you have a fixed curve and you have to pull back timing because of pinging at peak torque you're probably losing some power up top. Not ideal, but that's the world with points and simple vacuum/mechanical advance. I'd argue that non computer controlled ignition can still be VERY effective, but I'll be the first to admit it's really not fun when it comes to force induction.

 

And I've also been wanting to point out (since I ran out of time earlier a few days ago) that a motor's combustion properties should really be designed around a specific fuel range. Back in the leaded gasoline at the pump days, muscle cars saw VERY high compression ratios, even by today's standards. These motors were designed around a specific fuel's abilities and when leaded fuel became banned and/or unavailable, manufactures had to design for a completely different fuel. They had to lower compression ratios and change distributor weights/advance curves. Let me stress this point, I'm not angry this is what you do in engineering, you design parts to work as a SYSTEM for a given APPLICATION. This doesn't make either motor setup right or wrong. If you want to run race gas, design accordingly. If you want to run pump gas, design accordingly. Don't design a motor that should run on race gas and then complain that you're detonation limited on 87 octane.

 

That said, I run my L28ET on 87 octane. Yes I'm under 10psi, but I'm also un-intercooled.

 

I personally feel this is the big mistake with flex-fuel vehicles. Any E85 running vehicle in NA form should have AT LEAST 13:1 compression as a MINIMUM. If you had someone like Honda design a E85 only S2000 they should easily be reaching 15:1 if not higher compression. But why haven't they? They don't want to design all new cylinder heads for a marketing brochure. Nobody's sold on E85 or it's availability on a large scale. I get that. But selling flex fuel vehicles in their current state only hurts the case for E85. People see it as "expensive, poor efficiency, and not generally worth the environmental improvement", when ALL THREE of those factors would be vastly improved when designed around that fuel from the get go.

 

So nobody should think that it's just as easy as changing from one fuel to the next and making small adjustments in timing. That's called "making due" with what you have. It's not the ideal solution. The B16A isn't timing limited and the JDM motor I had in my personal car ran damn well on 87 octane when with it's stock JDM tune I really should have been putting 91 in, but I was too cheap to do so. If honda can design engines that make 120+ hp per liter and not run into timing limitation, I think we should be able to get darn close to 100+ with our motors. Just learn from Honda, the devil is in the details.

[Leon]

Well I don't think I ever once really disagreed with YOU, just semantics which makes for good talking points.

 

What you say is true. If you have a fixed curve and you have to pull back timing because of pinging at peak torque you're probably losing some power up top. Not ideal, but that's the world with points and simple vacuum/mechanical advance. I'd argue that non computer controlled ignition can still be VERY effective, but I'll be the first to admit it's really not fun when it comes to force induction.

 

And I've also been wanting to point out (since I ran out of time earlier a few days ago) that a motor's combustion properties should really be designed around a specific fuel range. Back in the leaded gasoline at the pump days, muscle cars saw VERY high compression ratios, even by today's standards. These motors were designed around a specific fuel's abilities and when leaded fuel became banned and/or unavailable, manufactures had to design for a completely different fuel. They had to lower compression ratios and change distributor weights/advance curves. Let me stress this point, I'm not angry this is what you do in engineering, you design parts to work as a SYSTEM for a given APPLICATION. This doesn't make either motor setup right or wrong. If you want to run race gas, design accordingly. If you want to run pump gas, design accordingly. Don't design a motor that should run on race gas and then complain that you're detonation limited on 87 octane.

 

That said, I run my L28ET on 87 octane. Yes I'm under 10psi, but I'm also un-intercooled.

 

I personally feel this is the big mistake with flex-fuel vehicles. Any E85 running vehicle in NA form should have AT LEAST 13:1 compression as a MINIMUM. If you had someone like Honda design a E85 only S2000 they should easily be reaching 15:1 if not higher compression. But why haven't they? They don't want to design all new cylinder heads for a marketing brochure. Nobody's sold on E85 or it's availability on a large scale. I get that. But selling flex fuel vehicles in their current state only hurts the case for E85. People see it as "expensive, poor efficiency, and not generally worth the environmental improvement", when ALL THREE of those factors would be vastly improved when designed around that fuel from the get go.

 

So nobody should think that it's just as easy as changing from one fuel to the next and making small adjustments in timing. That's called "making due" with what you have. It's not the ideal solution. The B16A isn't timing limited and the JDM motor I had in my personal car ran damn well on 87 octane when with it's stock JDM tune I really should have been putting 91 in, but I was too cheap to do so. If honda can design engines that make 120+ hp per liter and not run into timing limitation, I think we should be able to get darn close to 100+ with our motors. Just learn from Honda, the devil is in the details.

Or... just buy an S2000. Muahahaha!