What exactly is "knock?"
kjewer1
12-17-2003, 10:00 AM
While we are having technical discussions, why not talk about another widely misunderstood part of DSM life. Knock. What the hell is it.
People use the term knock to cover several different things. The most common use of it is another way to say "preignition." We all know preignition as the mixture in the chamber getting compressed, and ingiting before the plug fires. This was common on NT V8s using high compression and low octane, and/or advanced ignition timing. Another use of the word is to describe a situation where you have more than one ignition (all after the plug fires) causing a ripple of pressure spikes in the combustion chamber that can damage valves and pistons, bearings, and other internal parts. In the sense of DSMs, the latter description is the one we are concerned with. With the base timing advance we run, computer controlled advance, and low compression ratios, preignition is going to be quite rare. Even under boost. Not to say that it is impossible though ;) So we will discuss this second and more common occurence of what we call "knock."
Just to cover the basics, a quick overview of what is happening on the compression and power strokes. The piston rises, compressing the air fuel mixture. At some point before the piston even reaches Top Dead Center (TDC) the plug fires. By the time the flame front starts to expand rapidly, the piston is (hopefully) just past TDC, and all of this expansion will drive the piston down over its entire downward travel to Bottom Dead Center (BDC). This is what happens when it all works correctly :)
Its also important to understand a few things about ignition/combustion, and how gasses behave when compressed. A given air-fuel mixture will ignite at a certain temperature. The majority of the temperature given to ignite the mixture is provided by an electical arc in the gap of the spark plug. We all know how this works. But what else can add heat to the airfuel mixture? When a gas is compressed, it heats up. I will include an explanation of the Ideal Gas Law at the end of this article, but for now take my word for it, for simplicity's sake :) In all internal combustion engines, the airfuel mixture is compressed. This adds heat to the mixture. In a turbo car, the mixture is also pre-compressed by the turbo. In compressing the air, heat is increased just as in the compression stroke in the motor. The amount of heat added will be the minimum described by the rules of adiabatic compression, and the Ideal Gas Law. In addition to that, more heat is added by the lack of efficiency of the compressor wheel of the turbo. Pulling in hot underhood intake air will also add heat to the mixture. Raising the boost will increase the amount of air that has to be compressed into that same finite space in the combustion chamber, again rasing pressure and therefore heat. Etc... Lots of heat added to the mixture.
A hot mixture is easier to ignite. Its also closer to that point at which a given mixture will ignite on its own, independant of an ignition source like a spark plug. Here is where the problem lies.
Lets take a closer look now, armed with this information, at the combustion process/power stroke. The piston is just before TDC. The plug fires. A flame "kernal" starts at the point of ignition, in the gap of the plug, and grows outward as neighboring fuel/O2 molucules ignite. The "flame front" propagates outward in all directions, just as the piston passes TDC. It hits the roof of the combustion chamber quickly, but still has to spread all the way out to the piston in the other direction. As more fuel/air is burned up the gasses expand greatly. This is after all the whole point, it is what drives the piston down the bore. This expansion however has increased the pressure in the entire cylinder dramatically. And we know that pressure is heat, and heat will cause fuel/air mixture to ingite at some point without an ignition source, like the flame front touching it. The flame front is now halfway to the downward moving piston. In the far reaches of the combustion chamber, we'll say between the piston and the cylinder wall, but above the top ring, in that little crevice, the pressure/heat is great enough to ignite a few molecules of mixture there, before the original flame front gets there. That new flame kernel grows and begins to expand outward. It will increase the pressure further, in addition to the pressure being created by the orignal flame front/burning mixture's expansion. Perhaps another area now gets hot enough and ignites on its own. Now we have 3 different flame fronts/expnasions going on, at once, but out of snyc! They pass each other as they expand outward in all directions, hitting each other and the cylinder wall, head, piston, etc at different times. The result? 3 pressure spikes, in succession, that will "rattle" the piston/rings/valves, and the force is transmitted down through the piston pin, the rod, and rod/crank bearings as well. These multiple pressure spikes at different times and in different directions are what does the damage. This is an example of what a "controlled burn" is NOT. ;) And 3 was just an example. There can be any number of independant ignitions.
This is what we hear as knock, and pick up on our ECUs "microphone" as knock to view on a datalogger. This is the ultimate limit to the amount of power that one can make on a given setup, in most cases. Its what we avoid at all costs to keep out motors running happily.
Now that we are knocking, what has to be done to eliminate it. Well, we all know what has to be done.
-Reduce the temperature of the intake air. Cold air intake. Larger more efficient intercoolers. More efficient turbos. Lower pressure drop of IC and piping, so we dont have to make an extra 2-3 psi just to get X amount of boost at the manifold. We can also dump in more fuel, which acts as a coolant (though water and alcohol are 6-7 times better at absorbing heat, think water injection) to lower the temp of the mixture.
-Reduce the pressure of the intake charge. Lower boost. Lower compression ratio. Fire the spark plug later in the stroke, when the piston is further down and pressure is reduced (retarded timing).
-Increase the fuel-air mixtures tendency to ignite on its own. In other words, raise the temperature required to ignite it. This is done by increasing the octane.
-Eliminate secondary sources of combustion, like sharp edges in the chamber that will glow and act as another spark plug. Carbon deposits, imporperly size head gaskets, protruding sparkplug threads, etc.
The whole key to making more power is to burn more fuel/air mixture, without knocking. If you can push of knock further, you can run leaner, run more boost, run more timing advance, etc. This is how you make power. ;) When there is no knock, there is a controlled burn that can make tons of power with the abuse of mulitple pressure waves rattling and damaging internal engine parts. Just one smooth push down on the piston, turning the crank, and making things happen.
As I mentioned above, there is the Ideal Gas Law. Just when you thought you could sleep through physics class, you missed an important DSM lesson :icon16: Here it is.
PV=nRT
P - Pressure
V - Volume
n - amount of gas
R - Universal Gas Constant
T - Temperature
While the actual number and units used are not important, the way the equation describes the relationship between the variables is. In a fixed volume, raising pressure raises temp. Or raising temp raises pressure. Etc. Like in any math equation, what happens on one side of the = must also happen on the other side, for it to remain balanced. If you hold all variables on one side constant, and change one on the other side, the other variables on that side must do the opposite. For example, if you hold n, R, and T constant, and raise P? V must go down. And it does. For a given amount of gas, at a given temperature, if you raise pressure, volume had to have gone down. See how simple it is? :)
This is an oversimplified overview of what knock is. I dont know everything, but I hope it helps to describe the laws of phsyics at work in a way a typical DSMer can understand it. As always, discussion is encouraged :biggrin:
People use the term knock to cover several different things. The most common use of it is another way to say "preignition." We all know preignition as the mixture in the chamber getting compressed, and ingiting before the plug fires. This was common on NT V8s using high compression and low octane, and/or advanced ignition timing. Another use of the word is to describe a situation where you have more than one ignition (all after the plug fires) causing a ripple of pressure spikes in the combustion chamber that can damage valves and pistons, bearings, and other internal parts. In the sense of DSMs, the latter description is the one we are concerned with. With the base timing advance we run, computer controlled advance, and low compression ratios, preignition is going to be quite rare. Even under boost. Not to say that it is impossible though ;) So we will discuss this second and more common occurence of what we call "knock."
Just to cover the basics, a quick overview of what is happening on the compression and power strokes. The piston rises, compressing the air fuel mixture. At some point before the piston even reaches Top Dead Center (TDC) the plug fires. By the time the flame front starts to expand rapidly, the piston is (hopefully) just past TDC, and all of this expansion will drive the piston down over its entire downward travel to Bottom Dead Center (BDC). This is what happens when it all works correctly :)
Its also important to understand a few things about ignition/combustion, and how gasses behave when compressed. A given air-fuel mixture will ignite at a certain temperature. The majority of the temperature given to ignite the mixture is provided by an electical arc in the gap of the spark plug. We all know how this works. But what else can add heat to the airfuel mixture? When a gas is compressed, it heats up. I will include an explanation of the Ideal Gas Law at the end of this article, but for now take my word for it, for simplicity's sake :) In all internal combustion engines, the airfuel mixture is compressed. This adds heat to the mixture. In a turbo car, the mixture is also pre-compressed by the turbo. In compressing the air, heat is increased just as in the compression stroke in the motor. The amount of heat added will be the minimum described by the rules of adiabatic compression, and the Ideal Gas Law. In addition to that, more heat is added by the lack of efficiency of the compressor wheel of the turbo. Pulling in hot underhood intake air will also add heat to the mixture. Raising the boost will increase the amount of air that has to be compressed into that same finite space in the combustion chamber, again rasing pressure and therefore heat. Etc... Lots of heat added to the mixture.
A hot mixture is easier to ignite. Its also closer to that point at which a given mixture will ignite on its own, independant of an ignition source like a spark plug. Here is where the problem lies.
Lets take a closer look now, armed with this information, at the combustion process/power stroke. The piston is just before TDC. The plug fires. A flame "kernal" starts at the point of ignition, in the gap of the plug, and grows outward as neighboring fuel/O2 molucules ignite. The "flame front" propagates outward in all directions, just as the piston passes TDC. It hits the roof of the combustion chamber quickly, but still has to spread all the way out to the piston in the other direction. As more fuel/air is burned up the gasses expand greatly. This is after all the whole point, it is what drives the piston down the bore. This expansion however has increased the pressure in the entire cylinder dramatically. And we know that pressure is heat, and heat will cause fuel/air mixture to ingite at some point without an ignition source, like the flame front touching it. The flame front is now halfway to the downward moving piston. In the far reaches of the combustion chamber, we'll say between the piston and the cylinder wall, but above the top ring, in that little crevice, the pressure/heat is great enough to ignite a few molecules of mixture there, before the original flame front gets there. That new flame kernel grows and begins to expand outward. It will increase the pressure further, in addition to the pressure being created by the orignal flame front/burning mixture's expansion. Perhaps another area now gets hot enough and ignites on its own. Now we have 3 different flame fronts/expnasions going on, at once, but out of snyc! They pass each other as they expand outward in all directions, hitting each other and the cylinder wall, head, piston, etc at different times. The result? 3 pressure spikes, in succession, that will "rattle" the piston/rings/valves, and the force is transmitted down through the piston pin, the rod, and rod/crank bearings as well. These multiple pressure spikes at different times and in different directions are what does the damage. This is an example of what a "controlled burn" is NOT. ;) And 3 was just an example. There can be any number of independant ignitions.
This is what we hear as knock, and pick up on our ECUs "microphone" as knock to view on a datalogger. This is the ultimate limit to the amount of power that one can make on a given setup, in most cases. Its what we avoid at all costs to keep out motors running happily.
Now that we are knocking, what has to be done to eliminate it. Well, we all know what has to be done.
-Reduce the temperature of the intake air. Cold air intake. Larger more efficient intercoolers. More efficient turbos. Lower pressure drop of IC and piping, so we dont have to make an extra 2-3 psi just to get X amount of boost at the manifold. We can also dump in more fuel, which acts as a coolant (though water and alcohol are 6-7 times better at absorbing heat, think water injection) to lower the temp of the mixture.
-Reduce the pressure of the intake charge. Lower boost. Lower compression ratio. Fire the spark plug later in the stroke, when the piston is further down and pressure is reduced (retarded timing).
-Increase the fuel-air mixtures tendency to ignite on its own. In other words, raise the temperature required to ignite it. This is done by increasing the octane.
-Eliminate secondary sources of combustion, like sharp edges in the chamber that will glow and act as another spark plug. Carbon deposits, imporperly size head gaskets, protruding sparkplug threads, etc.
The whole key to making more power is to burn more fuel/air mixture, without knocking. If you can push of knock further, you can run leaner, run more boost, run more timing advance, etc. This is how you make power. ;) When there is no knock, there is a controlled burn that can make tons of power with the abuse of mulitple pressure waves rattling and damaging internal engine parts. Just one smooth push down on the piston, turning the crank, and making things happen.
As I mentioned above, there is the Ideal Gas Law. Just when you thought you could sleep through physics class, you missed an important DSM lesson :icon16: Here it is.
PV=nRT
P - Pressure
V - Volume
n - amount of gas
R - Universal Gas Constant
T - Temperature
While the actual number and units used are not important, the way the equation describes the relationship between the variables is. In a fixed volume, raising pressure raises temp. Or raising temp raises pressure. Etc. Like in any math equation, what happens on one side of the = must also happen on the other side, for it to remain balanced. If you hold all variables on one side constant, and change one on the other side, the other variables on that side must do the opposite. For example, if you hold n, R, and T constant, and raise P? V must go down. And it does. For a given amount of gas, at a given temperature, if you raise pressure, volume had to have gone down. See how simple it is? :)
This is an oversimplified overview of what knock is. I dont know everything, but I hope it helps to describe the laws of phsyics at work in a way a typical DSMer can understand it. As always, discussion is encouraged :biggrin:
EclipseRST
12-17-2003, 12:27 PM
are you sure you dont know everything, cause it seems like you do!! well atleast about DSM's :iceslolan and that is a good thing cause i think everyone is glad that you do!
goalieman24
12-17-2003, 02:41 PM
i saw the question as the title of the post and thought holy shitt gsxracer doesnt really know everything... i was wrong :thumbsup:
JoeWagon
12-17-2003, 05:21 PM
Some of these are longer than essays i turn in for english, hope you're saving them for next year when 50 new people come and ask the same questions :)
kjewer1
12-17-2003, 05:47 PM
You bet your ass I save these :biggrin: It took me damn near an hour to type all that. I have a folder for all this stuff. I just looked and there are 93 text files there. I never would have thought it was that high. Some are short though. This is by far the longest I have written on knock. It still doesnt capture everything I would like to say about it, but the problem is you start to venture off into other topics worthy of thier own article, like the ideal gas law. To try to put it all in one post makes it confusing if not just too damn long to read :biggrin: Everything is all connected. The further you get into this stuff, the more apparent that becomes. :) I write this stuff hoping it will make sense to most other DSMers, but in reality it actually helps me get my thoughts organized on the subject. Its easy to understand all this stuff, but to put it all into words, in a way that makes sense to even myself, is a pain in the ass :icon16: And I still have a lot to learn. I find myself coming back to some of these text files as the years go by to touch them up or add more to them. Ah well, hope you guys find it useful.
95_GSX
12-17-2003, 10:22 PM
once again thank you kevin for sharing your knowledge, there is a lot of good info in here. and it cleared a lot of the DSM knock question up in my head. i thought we were getting pre-ignition knock from to high of intake-temp(which i understand could still happen under in-proper conditions). but i love reading these tech discussions about our cars. there is no such thing as to much knowledge.:thumbsup:
kjewer1
12-18-2003, 07:55 AM
Some people are asking about the other articles I have written. When I get around to it I plan to add them all to the POSR website. I tried to upload that one about the 20 dollar home depot intake, and the program screwed up and I lost it all. :lol: So thats an empty link still. But thats the section where you guys will find them all eventually. :thumbsup:
Nalfein
12-19-2003, 06:04 AM
hey 95_GSXracer,
I find this stuff fascinating...
Thanks a lot for taking the time out to teach us this stuff.
You say you have a folder with 93 texts like that one?
Can you email me the folder please
mattpark@telus.net
thanks so much
it woould be like an automotive engineering text book and hours of fun reading for me.
Thanx again. Without ppl like you, these foruns would not work.
Matt
I find this stuff fascinating...
Thanks a lot for taking the time out to teach us this stuff.
You say you have a folder with 93 texts like that one?
Can you email me the folder please
mattpark@telus.net
thanks so much
it woould be like an automotive engineering text book and hours of fun reading for me.
Thanx again. Without ppl like you, these foruns would not work.
Matt
Automotive Network, Inc., Copyright ©2025