How important is knowing knock when tuning?

[jonnygsx]

Ok, I'm going to be going into fuel tuning once I get my 16g back and installed. I am trying to decide what kind of logger/fuel tunner to get. I have noticed that only DSMLINK will read knock on my car. I don't really want DSMLINK because of A)price B)the need to have a laptop. So my question is in regards of how necessary it is to read knock when tuning, or is an EGT guage sufficient enough? Is there a way to read knock on a seperate guage?

P.S. this is on a 1998 GSX

Jon

 

[DrZiplok]

It's important, although you can infer knock activity from timing.

In your case, as long as you avoid running too lean, you can assume that the ECU will manage knock. Just be aware that you can run it into a corner; all it can do is retard timing and dump in more fuel. If you get to the point where it can't do either of those anymore, you're in serious danger.

If you're serious about pushing your setup hard, then DSMLink is really the smart move. Contrast the cost vs. an engine rebuild...

 

[FastRthenU]

Buy the dsm link. If they had it for first gens i would buy it in a heart beat. Count yourself lucky. 2 things cause knock. Lean air fuel ratio, and to much timming. Both can be controled with the dsmlink, plus you can data log with it.

When the ECU see's knock it retards timming. Depending on the amount of knock, it will pull the proper amount of timming. Timming is everything when making power. So i would say its pretty important when tunning.

 

[jonnygsx]

After reviewing the cost of POCKETLOGGER+S-AFC+EGT GUAGE vs. DSMLINK. I have decided to buy DSMLINK. Thanks for you input guys.

Jon

 

[LandoAWD]

Originally posted by FastRthenU
Buy the dsm link. If they had it for first gens i would buy it in a heart beat. Count yourself lucky. 2 things cause knock. Lean air fuel ratio, and to much timming. Both can be controled with the dsmlink, plus you can data log with it.

When the ECU see's knock it retards timming. Depending on the amount of knock, it will pull the proper amount of timming. Timming is everything when making power. So i would say its pretty important when tunning.

Ummm..alot more than 2 things contribute to knock count. And, its TIMING and TUNING...getting a lil happy with our 'n' key today?

 

[FastRthenU]

Im sorry but those two things are the cause of knock. Please tell me what else cause's it then. Instead of just saying there are more things please tell me what they are.

 

[LandoAWD]

Originally posted by FastRthenU
Im sorry but those two things are the cause of knock. Please tell me what else cause's it then. Instead of just saying there are more things please tell me what they are.

Knock can be caused by being too rich, too lean, mechanical noise (i.e. lifter tick, belt slap), and of course there is phantom knock. Don't give me the " " smiley just because I called you out....I was just making a point.

 

[larryd]

Originally posted by jonnygsx
After reviewing the cost of POCKETLOGGER+S-AFC+EGT GUAGE vs. DSMLINK. I have decided to buy DSMLINK. Thanks for you input guys.

Jon

I have mine and the Eprom ECU up for sale in the classifieds if your interested.

 

[FastRthenU]

Ok first off all that you pointed out other then to much fuel(which i have never heard of) is not true knock. Many things can trip the knock sensor but not enough fuel, or to much timing causes the condition for getting true knock.

hell i can trip the knock sensor by taping a screw driver handle against the block. I guess im now a cause of knock now to

 

[LandoAWD]

Originally posted by FastRthenU
Ok first off all that you pointed out other then to much fuel(which i have never heard of) is not true knock. Many things can trip the knock sensor but not enough fuel, or to much timing causes the condition for getting true knock.

hell i can trip the knock sensor by taping a screw driver handle against the block. I guess im now a cause of knock now to

So how am I wrong then? Exactly....I'm not. Somebody is a little nitpicky today.

 

[FastRthenU]

Im sorry but im not nitpicking at all. Im telling the main conditions that cause knock. Real knock, not ####in valve lifters or shit like that. You cant tune out shit like phantom knock, and even though those things due trigger the knock sensor, and hurt performance because of it. They well not hurt your engine.

There are a few other things that cause knock. But the main two are lean air/fuel ratio, and two much timing. When I said that you said there are alot more, but did not tell me what. So who is nitpickin . then you spouted some crap about phantom knock, and shit like that. As of two much fuel i have heard ppl ask if it does cause it but have never heard, or seen proof. So why dont you show me some proof of this, or any other conditions that do cause ping, or detonation (Real Knock) I can think of 1 or 2. lets see how you do

 

[LandoAWD]

Is it true that if you over tighten the knock sensor, it becomes over sensitive? I have problems with timming being pulled big time. I checked the knock sensor and it was toast. So i replaced it from one off a cyclone motor. I changed it and it still does the same thing. the thing is that i over tightend it for sure. Probley by a fair bit. would this be the problem?

This was a post of yours from a 2 weeks ago. I'm glad to see you have progressed so far. I'm done bickering. Sh!t like this makes this site look bad.

 

[FastRthenU]

Why because you dont know what your talkinc about. Thats what makes this site look bad. People that speak before they know what they are talkin about. I wouldnt be surprised if you have to go look up what knock is.

Well here i will help you before you have to. Knock is basically a uncontroled burn. Its when the cylinder presure and temp are higher then the air fuel mixture's autoburn point. Like a diesle engine basically. instead of using spark, it use's the cylinder presure and temp to ignite the mixture. well there you go. atleast know you might know what knock is.

as of the cause's to it.
Pre-ignition is one cause. an other is a lean air/fuel ratio. One more is low octane, one more is to much timing. Things that also make it easyier to knock are higher compression ratio's and cylinder head design.

You happy know. Now you can accually be able to help ppl if someone asked this question again. instead of there a more things then that. Now you know what they are. As of me asking that question on another post. yes i did and what the #### is the problem with that. I have heard yes and no on that same question many times. maybe you would like to tell em the answer

but before you try to act like you know it all. try given some more examples, and maybe just maybe a bit of proof.

 

[Doug99RS]

Ouch...
Ok... so here's some factual information about knock/detonation/preignition or whatever you'd like to call it. First off... I'm not a turbo tuner. I don't have one and don't claim to know how to tune them.

But I do know a thing or two and hopefully, without getting in to the middle of this sqwabble, can help to clarify and enlighten a few.

The theory of operation of the knock sensor is as follows:
The knock sensor is mounted on the cylinder block below the intake manifold. The knock sensor is comprimised of thin piezo-electric ceramic discs bonded to a metal diaphragm. When engine knock occurs, the diaphragm vibrates along with the piezo-electric discs, producing an electrical signal. The signal is used by the PCM (ECU for you turbo boys) to adjust ignition timing to eliminate the knock condition.

Knock is a vibration. A vibration is caused by something in the engine not working in harmony with the other rotating components. By adjusting the timing... the theory is that the vibration is caused by one or more cylinders are igniting at the improper point during the rotation of the crank. When the fuel charge ignites prematurely then it tries to send the piston and rod back the way it just came thus one cylinder is trying to fight the rotation of the crank while the other three are going with the flow.

There are at least two things that can cause this. One is the intentional igniting of the spark plug before the cylinder is at top dead center which could be corrected by adjusting timing. The second is the ignition of the fuel charge before the piston is at top dead center regardless of the spark plugs participation is the combustion sequence. There is normally only one thing that causes this. Too high of temperatures in the combustion chamber.

High temperatures can be caused by numerous things though. Carbon build up on the piston and cylinder head surface inside the cylinder can retain heat, this heat is a hot spot and a hot spot will then ignite the fuel charge too soon. Too lean of a fuel mixture will also increase temperatures resulting in the fuel charge being ignited. Also any time you raise pressure you also raise temperatures. Thus... too much boost can raise pressures too high and without the proper fuel mixture you can't do anything to drop the temperature. Fuel can work wonders. Not only are we talking about something that is designed to blow up. But it can also be used to cool cylinder head runners, intake valves, pistons and cylinder temperatures down to where they are supposed to be.

That's all dealing with the air fuel charge and temperatures that affect your precious knock/pre-ignition/detonation. But there's also the question of a physical condition that causes a vibration. Remember the knock sensor picks up vibration. So... removal of balance shafts which are designed to counter-act engine vibration, installation of an underdrive pulley which does away with the vibration dampener/harmonic balancer, improperly balanced pistons, rods, cranks, or flywheels, uneven carbon build up on pistons, oil caked on to the crank, hell... even a clutch that is worn unevenly theoretically can cause uneven weight distribution which will add to vibrations which can be interpretted as "knock".

The list is really quite large as to what can cause these things ya'll call knock.
So why don't ya'll two play nice now huh?
Doug

 

[FastRthenU]

I was going to go more indepth with the cause's but i guess you did it for me. Thanks for accually putting down somthing instead of just talkin out your ass.

but im done. Ive made my point and I will play nice now.

 

[LandoAWD]

Originally posted by FastRthenU
I was going to go more indepth with the cause's but i guess you did it for me. Thanks for accually putting down somthing instead of just talkin out your ass.

but im done. Ive made my point and I will play nice now.

Sure you were...but I'll play nice too. I am not, nor hav I ever sair "you're wrong", and both of us are correct, but you find it necessary to argue. You restate points I made in different words and call me ignorant (not in so many words). Thanks, Doug, for a more informative post than either of us had made...I bow down to you

 

[Doug99RS]

It's ok. You don't have to bow down. You'll show me up one day about something and return the favor.
D

 

[webbdog]

Knock is a vibration. A vibration is caused by something in the engine not working in harmony with the other rotating components.

I thought the knock sensor "listened" for knock. How would vibrations be a good indicator of knock when the whole engine it vibrating? And, how would lifter "tick" cause false knock readings from vibrating? You're information is correct because sound is transmitted through vibration; I just think it's important for them to understand that "sound" is what the sensor measures.

Here's something I found on another board:

"Knock sensors are microphones. They produce an electrical frequency based on sounds/vibration from the block. The ECU only cares about a range and amplitude of frequencies and because of this there will be a bandpass filter. Bandpass filters consist of a resistor and capacitor of certain values and a certain configuration to filter all but an exact range of frequencies."

 

[FastRthenU]

well more indepth me go then.. (very long)

You have to know how it things are supposed to work before you can understand why they aren't working. The basic "ideal" (meaning what engineers pretend happens when they are lazy and want their calculations to be really simple) spark-ignition cycle for one cylinder consists of four parts. First, the Intake stroke: the fuel and air mixture is drawn into the cylinder through the intake valve as the piston goes down. The pressure and temperature in the cylinder remain near outside conditions. Second, the Compression stroke: the fuel and air mixture is compressed as the piston goes up. The pressure in the cylinder goes up, and so does the temperature. At the end of the Compression stroke, the spark plug ignites the fuel mixture and it (in the "ideal" version) combusts instantly. Third, the Power stroke: the cylinder is now full of hot gasses that are at much higher pressure and temperature than the unburned compressed mixture was. This hot burned gas mixture pushes the cylinder down, which pushes the crankshaft around, which pushes your car forward at 100 MPH past the nice highway patrol officer.... Fourth, the Exhaust stroke: these burned gas mixture is pushed out the exhaust valve into your exhaust manifold. The cycle then repeats.

As I said, this is a very simplified version of what happens in a real engine. The only difference that is really important as far as knock is concerned is that the fuel does not burn anywhere near instantaneously at TDC (top dead center). It takes about 0.5 ms (around 7.5 degrees of crank angle at 2500 RPM) after the spark occurs for the fuel to start burning beyond a small ball around the spark plug. This region of burning then spreads throughout the cylinder, and the burning is mostly complete somewhere around 30 to 50 degrees of crank angle after the spark. Now, clearly, if we ignite the fuel mixture directly at top dead center, the piston will have travelled down quite a distance before a considerable part of the fuel burns. This fuel, then, is unable to help push the piston down as much, and thus the engine is less powerful. Because of this, the spark is usually set to occur before the piston hits TDC, so more fuel will be burned when the piston begins moving downward. If the spark is advanced too far, though, the burning fuel starts pushing against the piston during the compression stroke and hindering things instead of helping. The optimum spark timing is obviously a trade-off between these two effects. Small changes don't make a particularly big difference in power, so usually the minimum advance needed to get the maximum torque is used (this is denoted MBT).

As you recall, the fuel-air mixture gets to a fairly high pressure during the Compression stroke. This, along with the heat of the engine, causes the mixture to become very hot. In a diesel engine, this temperature is high enough to cause the fuel to spontaneously ignite. In a spark-ignition engine, however, the mixture is kept somewhat below the ignition threshold.

Now, consider what happens when the fuel starts to burn. there is a fairly large amount of mixture that does not normally begin burning immediately. As the mixture to the left burns, it compresses and heats up the mixture to the right even more than it is already heated and compressed. If the pressure and temperature of this unburned gas gets high enough, it can spontaneously ignite without waiting for the flame to spread to it.

PRE-Ignition
Pre-ignition is term used for ignition that occurs before the spark plug is fired. I mention it here because it can lead to knock; I will discuss why momentarily. Usually preignition occurs because of some hot spot in the engine that causes the fuel-air mixture to be heated past its autoignition point. There are many possible hot spots in the engine. If deposits are built up on the spark plug, they will be heated to a very high temperature when the spark fires. However, they do not lose their heat quickly, and can remain hot enough to cause pre-ignition in the next compression stroke. Exhaust valves are generally quite hot under normal operation, and if something happens to over heat them or keep them from being properly cooled, they can cause pre-ignition. Finally, carbon deposits on the cylinder walls can stay hot like burning coals and cause pre-ignition. Even if pre-ignition does not cause knocking, it causes higher temperatures and pressures in the cylinder that can cause rings to fail and even cause aluminum-alloy pistons to melt. Not pretty.

What to do About Knock
Knock occurs because the pressure and temperature in the cylinder are high enough when combustion is occurring that the unburned fuel/air mixture spontaneously ignites. Thus, these temperatures and pressures must be kept below a certain level to prevent knock from occurring. I will discuss several factors that affect this temperature and pressure, and thus affect knock.

Engine compression ratio. If the mixture is compressed more, then the pressure gets higher, and knock is more likely to occur. Thus, a higher engine compression will cause knock to be more likely.

Spark Advance. The maximum pressure and temperature in the cylinder occur some time after the spark occurs. If the spark is delayed so that this maximum pressure occurs after TDC, the downward motion of the piston will somewhat counteract the pressure rise from combustion, causing the maximum pressure to be lower. Thus, retarding the spark causes knock to be less likely.

Pre-Ignition. Pre-ignition has the same effect as increasing the spark advance. It causes knock to be more likely.

Combustion Chamber Design. This has a fairly complicated relationship with knock. First, if the distance from the spark plug to the far corners of the combustion chamber is reduced, combustion occurs more quickly. This increases the power, and also reduces the amount of time that knock has to occur. (Minimizing this distance led to the design of near-hemispherical heads in the Ford 429 and Chrysler Hemi, among others.) Second, if the exhaust valve is placed close to the spark plug it will not heat up the later unburned gas as much, and thus will reduce the likelihood of knock. Third, cooler areas such as the intake valve and areas close to both the piston and cylinder head can be placed in the parts of the cylinder that burn last, thus cooling this unburned gas and reducing knock. Finally, increasing the turbulence in the engine (up to a point) increases the burn rate and thus reduces knock. Turbulence can be increased by valve design or by creating areas of mixture that are squeezed between the piston and head, and are forced at high velocity into the middle of the combustion chamber (as emphasized in the Chrysler Wedge engines).

There is once significant factor other than the cylinder pressure and temperature that affects knock. That is the temperature at which the fuel/air mixture will spontaneously ignite. Since the ratios of fuel and air are fairly well determined by other considerations, the main variable here is the fuel. If the fuel is able to withstand higher temperatures before self-igniting, it will clearly be less likely to knock. The resistance of fuel to knocking is quantified as the Octane rating. The Octane rating is defined by the percentage of iso-octane in an iso-octane/n-heptane mixture that is required to match the knock behavior of the given fuel in two different tests. Both of these tests are done on a standardized variable-compression single-cylinder engine, at two different speeds and intake temperatures, and the results are averaged. Octane ratings above 100 are defined by extrapolation from the iso-octane/n-heptane mixture.

 

[Doug99RS]

Well, they pretty much go hand in hand. A sound is produced by a vibration. That vibration happens at a certain frequency which is then interpretted by our ears, this knock sensor, or whatever the listening apperatus is.

So you may call it either one you like but they are one in the same.

"The knock sensor is mounted on the cylinder block below the intake manifold. The knock sensor is comprimised of thin piezo-electric ceramic discs bonded to a metal diaphragm. When engine knock occurs, the diaphragm vibrates along with the piezo-electric discs, producing an electrical signal. The signal is used by the PCM (ECU for you turbo boys) to adjust ignition timing to eliminate the knock condition. "

That part came straight out of the Chrysler Powertrain Diagnostics manual. If you were quoting something there too then hey, glad we both can read and copy. But I think both are saying the same thing just in a different manor.

About the only thing I didn't put in there was the theory of two different flame fronts colliding causing problems as well. I will find the link that GTM posted in because he filled in some blanks I created:
http://dsmtuners.com/forums/showthread.php?s=&threadid=37156

Doug

 

[Doug99RS]

p.s. why is it you can't type worth a darn when you're called out. But when you have time to do some research you don't mispell anything?

 

[FastRthenU]

Because yes it was a summary of a differnt page. I read it along time ago, and looked it up when I wrote that, because i thought it would give a good discription of what was being said. Either way i could have put it into my words, or i could just put it down for you do read. To be honest i did not feel like write all that in the first play. Im sorry if you thought they were my words. Not my intentions. But either way you got the info.

 

[Kevin Jewer]

Just to provide another way to look at it...

First of al, knock and preignition are two entirely different things, and are not to be confused as the same or even similar.

Knock is cause by pressure and heat, which in effect are the same thing (see: Ideal Gas Law). There are countless ways to raise pressure and heat. Those are the causes of knock.

Pressure:

Compression ratio/compression
boost
carbon buildups
multiple ingition points/flame fronts

Heat:

Turbo eficiency
IC efficiency
AF mixture
Intake temp/ambient temp
engine temps


Unless I am missing something everything else can be categorizd into the above. For example, a boost leak affects turbo efficiency, intake temps, af ratio... etc. Also anything that affects pressure affects heat, and vice versa. Like I said, they are the same.

Also of interest, I have NOT been able to make my car rich knock. As far as I can tell, it doesnt exist. I hav been as rich as 8 or 9 to 1, with no knock. Being ever so slightly lean knocks like hell, in contrast.

Just a few more humble ideas...

 

[FastRthenU]

yes knock and pre-ignition are different. but pre ignition cause's knock.

 

[atblacktalon92]

ok knock it off guys