CJ Posted June 13, 2005 Share Posted June 13, 2005 Thanks Tony... So, the knock sensor is sensing when something is not right. Maybe det or pre ignition. Therefore, in a perfect world, setting up your engine to stop the above would obviate the need for a knock sensor - or is that being too simplistic? Link to comment Share on other sites More sharing options...
Guest Usmann A Posted June 13, 2005 Share Posted June 13, 2005 a short graph of AFR relation to motors, its very general, but its a starting point. Link to comment Share on other sites More sharing options...
JohnA Posted June 13, 2005 Share Posted June 13, 2005 So, the knock sensor is sensing when something is not right. Maybe det or pre ignition. Everything Tony posted is correct. However, note that the knock sensor is (usually) just a microphone. It listens to whatever noise is there to listen. The hard bit is the 'interpretation' of that output. Different engines have different 'ring' characteristics, and at different states of tune they have different frequency bands where one should even begin to look for signs of abnormal combustion. It's all too easy to misinterpret high-rev noise for 'knock' and take measures that are not really warranted. It's equally easy to think that all that noise floor is just normal, and miss detonation spikes. Quite a difficult thing to get right with just a 'microphone' type of sensor, bearing in mind that nobody really knows exactly the nature of 'knock' or 'detonation' There are various theories yes, but they are just like electricity: they try to explain the behaviour without anybody really knowing what the undrlying concept is. I've got a lot more on the 'advanced ignition' page of my site if anyone's interested. I've even got code in C for knock-sensor signal 'filtering' if anyone is hardcore enough. Link to comment Share on other sites More sharing options...
TLicense Posted June 13, 2005 Share Posted June 13, 2005 Thanks Tony... So, the knock sensor is sensing when something is not right. Maybe det or pre ignition. Therefore, in a perfect world, setting up your engine to stop the above would obviate the need for a knock sensor - or is that being too simplistic? Yeah, your right setting it up not to knock would mean you could do away with a knock sensor as long as the conditions that don't cause knock never arose Having saying that, and I think I know where you're coming from, it's probably not that much of a pain in the arse. It just means you'll have to take a little more care of you P+J. ie make sure you only use good fuel, if you hear det then back off, maybe set-up the fuel and timing a bit more conservative (= less performance) than you would normally. If you have an EGT probe and a WB02 sensor, then it may be possible to set up the fuel map to read these so that if the by-products of knock are produced (ie high EGT, and I think usually a lean AFR???) then the ECU would do something about it. (ie retard ignition and add fuel) Whether or not these by-products are a)produced quickly enough, or b) are detectable by the relevant sensors quickly enough, to do something about it, I don't know. Link to comment Share on other sites More sharing options...
TLicense Posted June 13, 2005 Share Posted June 13, 2005 Agreed John, that is the difficult thing, being able to filter out the normal noise from the knock. Isn't there a HKS bolt on knock detector?? I imagine it's a fairly generic item in terms of usage. I wonder how that filters out the normal noise of different engines? Link to comment Share on other sites More sharing options...
JohnA Posted June 13, 2005 Share Posted June 13, 2005 I've experimented on this subject. There's more to it than meets the eye. Some people think that you just bolt the sensor, and set the ECU to do this and that when it 'senses' knock. If only.... We're not talking about something as straightforward as direct temperature or pressure sensing here, not by a long shot. As I'm sure you know on standalones like the AEM there is a set procedure for 'marking' the engine noise floor, and that is the method that produces the best results, as far as I know (with microphonic sensing) Link to comment Share on other sites More sharing options...
Guest Usmann A Posted June 13, 2005 Share Posted June 13, 2005 Yep tel, it does, It sees 0-5V feedack to the ECU, thing is though as you need to tell it(define) to the system what actual knock is. This the proper way, so you push the timing until you see voltage spikes on the sensor logs, and then you know your rough knock limit. You can add fuel too, with knock%rich/volt table and howmuch ign timing retard per rev. If its setup properly,it could save ur ass ... __________________ Like I stated above, if you ovbiously set it up on your trusty readily available road fuel, log knock voltage, and then work ur mapping up , in terms of boost,and timing, so start very conservative, and then step by step progess, in terms of upping the timing until you A) make no torque increace by winding timing up higher,B) you see spikes in the knock logs which you didnt when you were at lower ign values. Keep the boost and fuel the same, for the time being. Its a tedious way to calibrate the sensor for the motor, but if you know them inside out, and know whats works, a safe map on AEM can be done with a hour or so. this is all just a rough method, Thats a rough way of doing it, Link to comment Share on other sites More sharing options...
Guest Usmann A Posted June 13, 2005 Share Posted June 13, 2005 You have to remeber as the car ages the motor sound changes too, so knock refrenceing will need to be adapted if its used alot. The bore, and materials used all affect the sound of the motor. I dont know if anyone here has used Ion sensing? But, im looking at getting a sparkplug pressure transduer for the 2J, cylinder pressure testing needs to be done, and the 2 washers come tested to a certain Psi. Link to comment Share on other sites More sharing options...
JohnA Posted June 13, 2005 Share Posted June 13, 2005 ..I dont know if anyone here has used Ion sensing?... Volvo R&D use it, but they don't post here, do they? That's apparently superior to microphonic sensors with filtering algorythms. Link to comment Share on other sites More sharing options...
Ian C Posted June 13, 2005 Author Share Posted June 13, 2005 Detonation is the spontaneous combustion of the left over bits of air/fuel after the normal combustion. Pre-ignition is when the air-fuel mixture errmmm pre-ignites before the spark ignites the mixture properly. My take on it is that pre ignition is when the mixture is ignition a la spark plug by a hot spot in the combustion chamber - be it the spark plug tip as it's too hot, or a valve edge or piston chamfer, or carbon buildup acting like a glow plug. This is bad because it's like massively retarding the timing - the mix is burning as the piston comes up so large amounts of power is lost and the EGTs go up as the burning mix is in the chamber for x% longer than it should be. higher EGTs is a bad thing as well, because then you get more in-cylinder hotspots to cause preignition... Detonation on the other hand is when the mix spontaneously explodes - yup, normally it's a controlled burn, det is an uncontrolled bang. This can be triggered by the same things as preignition but can also be caused by too weak a mix - the mixture becomes volatile when it's leaned out so it'll explode instead of burn so even the normal spark event can cause det when there isn't enough fuelling going in. Det is the bad one of the two, although preignition isn't great in itself, the explosive factor of detonation causes huge pressure spikes in the combustion chamber and it's usually the piston that takes the brunt of this. I'm not entirely sure why detonation occurs as a factor of octane and boost pressure when all other factors are good. I know the human ear is good at picking out det from normal engine noise, it makes a very distinct "sping" sound. I don't know how good these bolt-on knock detectors are - has anyone got any experience of them...? -Ian Link to comment Share on other sites More sharing options...
JohnA Posted June 13, 2005 Share Posted June 13, 2005 In the light of the recent 'bad faith' posts aimed at myself, I prefer to unsubscribe from this thread before people come back and start abusing their 'right' to free speech. PMs or emails are less hassle. I've got a KnockLink for sale, if anyone's interested, by the way. Link to comment Share on other sites More sharing options...
Digsy Posted June 13, 2005 Share Posted June 13, 2005 Detonation on the other hand is when the mix spontaneously explodes - yup, normally it's a controlled burn, det is an uncontrolled bang. This can be triggered by the same things as preignition but can also be caused by too weak a mix - the mixture becomes volatile when it's leaned out so it'll explode instead of burn so even the normal spark event can cause det when there isn't enough fuelling going in. Det is the bad one of the two, although preignition isn't great in itself, the explosive factor of detonation causes huge pressure spikes in the combustion chamber and it's usually the piston that takes the brunt of this. Det is uncontrolled ignition from multiple sources. The charcteristic noise is the multiple flame fronts colliding in the combustion chamber. Link to comment Share on other sites More sharing options...
CJ Posted June 13, 2005 Share Posted June 13, 2005 Det is uncontrolled ignition from multiple sources. The charcteristic noise is the multiple flame fronts colliding in the combustion chamber. And am I right in saying that Det is normally caused by running too high a boost coupled with too lean a mixture? Or have I just proved my ignorance once again? Link to comment Share on other sites More sharing options...
Guest Usmann A Posted June 13, 2005 Share Posted June 13, 2005 Its like a thunderclap in the chamber. Link to comment Share on other sites More sharing options...
TLicense Posted June 14, 2005 Share Posted June 14, 2005 And am I right in saying that Det is normally caused by running too high a boost coupled with too lean a mixture? Or have I just proved my ignorance once again? Not at all mate, running too lean can lead to det. However other things can also cause det, like too low an octane number, and I believe extremely high intake temps? to name probably just a few. Link to comment Share on other sites More sharing options...
Digsy Posted June 14, 2005 Share Posted June 14, 2005 And am I right in saying that Det is normally caused by running too high a boost coupled with too lean a mixture? Or have I just proved my ignorance once again? That would be the main cause for home tuning. Other causes are too much spark advance or poor cooling. Link to comment Share on other sites More sharing options...
Ian C Posted June 14, 2005 Author Share Posted June 14, 2005 How does the spark advance cause det, I've always wondered that... -Ian Link to comment Share on other sites More sharing options...
eyefi Posted June 14, 2005 Share Posted June 14, 2005 Det is the bad one of the two, although preignition isn't great in itself, the explosive factor of detonation causes huge pressure spikes in the combustion chamber and it's usually the piston that takes the brunt of this. -Ian i think pre-ignition is the worst of the two because one event of pre-ignition can finish the engine and det is a more cumulative event. pre-ignition causes phenomenal pressures at exactly the wrong time in the stroke. Link to comment Share on other sites More sharing options...
eyefi Posted June 14, 2005 Share Posted June 14, 2005 How does the spark advance cause det, I've always wondered that... -Ian as the piston comes up on the compression stroke and the mixture of air and fuel get forced into a smaller area pressure and temp increase. the mixture is ignited by the plug before the piston reaches TDC because the mixture does not just spontaneously burn. a flame front heads out from the source of ignition and takes time to reach the entire volume of the cylinder. this is all timed so that the peak cylinder pressures occur just after TDC and the most possible work is got from the expansion of the burn. increase the ignition too far and the peak pressure occurs at the wrong time, too soon. under these conditions the mixture will start to spontaneously ignite at other parts of the cylinder as well as the spark plug point, voila det, multiple flame fronts meeting and an uncontrolled burn. Link to comment Share on other sites More sharing options...
Ian C Posted June 14, 2005 Author Share Posted June 14, 2005 Well, det is shock damage due to it being an explosion, and preignition is strain damage due to being an increase in cylinder pressures at the wrong time. I guess they are both as bad as each other Re det due to ignition advance, does the massive increase in pressure cause a more volatile mix that is more susceptible to detonation, much like a higher compression ratio or a weak mix does? -Ian Link to comment Share on other sites More sharing options...
eyefi Posted June 14, 2005 Share Posted June 14, 2005 yeah, pretty much. under so much pressure it's very hot and just capable of self igniting. yep, definately avoid both. i think det usually precurs pre-ignition, but not always. Link to comment Share on other sites More sharing options...
Guest Usmann A Posted June 14, 2005 Share Posted June 14, 2005 anywhere from 2-4000 psi increace. Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now