Notices
Drivetrain Gearbox, Diffs & Driveshafts etc

Detonation & knocking ?

Old 20 March 2001 | 08:21 AM
  #1  
NAF's Avatar
NAF
Thread Starter
Moderator
 
Joined: Feb 2000
Posts: 4,523
Likes: 1
From: Malta
Post

Hi
Could someone please explain these to me ?

Thanks
Niall
Old 20 March 2001 | 09:54 AM
  #2  
NITO's Avatar
NITO
Scooby Regular
 
Joined: Nov 1999
Posts: 1,001
Likes: 0
Cool

Hi Niall,

Knock,detonation, pre-detonation,pinging,pinking are all words describing the same effect.

If you imagine a piston moves up and down its bore (in the case of the scoob it moves left & right from the outside towards the centre as the pistons lie flat in the block horizontally opposed boxer layout) anyway, as the piston moves to the bottom of its stroke the inlet valves open and allow fuel and air into the combustion chamber, the piston then travels up to the top of it's stroke compressing this air and fuel mixture together, the spark plug then receives a signal from the coil pack and ecu to produce a spark which is effectively ignition timing (the timing of this spark) the spark causes an explosion which forces the piston back down in turn moving the crank which turns the flywheel etc.

With detonation or pre-detonation, what is happening is that this explosion is happening before the piston has reached the top of it's stroke. Hence the piston could be on it's way back up when the explosion occurs meaning that the inertia etc of the crank is forcing the piston back up against the force of this explosion (which is trying to force it back down) which obviously knackers the piston pretty quickly depending on the severity of the situation.

Effects of knock can mean peppered pistons and scratched bores with loss of compression etc...as weaker pistons will break down and little bits of metal from the crown of the piston will get caught along the walls of the bore causing striations in the cylinder wall causing all sorts of problems, or mild det can break down the oil properties causing the oil to become far less efficient which will eventually cause bearing failure, pre-dominantly on no.3 as it's the last to be fed oil and also the hottest since it sits just under the turbo/last to be fed/ heat transfer etc. According to STi the exhaust gas temp from no.3 can be upto 100 degrees hotter than the other cylinders!

There can be many reasons for det. The lower the octane of fuel, the easier it will self combust, basically imagine that your combustion chamber is very hot, for example if you run higher boost it means higher inlet temps etc and more air than fuel which means a leaner mixture etc which all equals heat, a low octane fuel will ignite when it comes into the chamber even with out the spark as the heat could be enough to ignite it. The higher the octane of fuel the more resistant it is to self igniting as it requires hotter temps or a spark to ignite.

More boost=more heat
Leaner mixture=more heat
Crap fuel=more easily self ignites=more knock
too advanced ignition timing=can equal det
Smaller turbo= more heat
Heat soak=more heat

More fuel(richer mixture)=less heat
Higher octane fuel=less chance of det
Better Intercooler=less heat
Bigger turbo=Less heat
retarded ignition timing=no det
Better cooling=Less heat soak

Tuning is all about balancing these. Less ignition timing will make the car slower, more advanced timing quicker but needs to be accompanied by a decent air/fuel mixture, too rich=safe but less power (also can cause bore wash and other problems if excessive) and lean = more power but the leaner you go the more you are going to run into potential problems.

The correct air/fuel ratio for the scoob as std seems to be about 12:1 - 12.5:1 however this can depend on state of tune etc, STi recommend as rich as 10:1 on the sti5 with exhaust gas temps of no more than 875 degrees max.

I'm not an expert in any way on this subject and the above is a basic explanation of my understandings of det etc.. I'm sure someone more technical can give you a more in depth account since there are probably more profound reasonings and implications.

Hope this helps,
Nito
Old 20 March 2001 | 11:40 AM
  #3  
NAF's Avatar
NAF
Thread Starter
Moderator
 
Joined: Feb 2000
Posts: 4,523
Likes: 1
From: Malta
Post

Nito
Cheers M8
Fantastic explanation ..
Must have taken you a while to type that lot !!
Great stuff
Thanks

Niall
Old 20 March 2001 | 11:40 AM
  #4  
NAF's Avatar
NAF
Thread Starter
Moderator
 
Joined: Feb 2000
Posts: 4,523
Likes: 1
From: Malta
Talking

Just as a matter of interest ...
What are the sympthoms of Detonation ??

Thanks again
Niall



[This message has been edited by NAF (edited 20 March 2001).]
Old 21 March 2001 | 02:34 AM
  #5  
pat's Avatar
pat
Scooby Regular
 
Joined: Aug 1999
Posts: 679
Likes: 0
Post

Just to throw a spanner in the works....

there's more than one kind of detonation. There is pre-ignition or surface det, where the fuelmix is lit up too early due o glowing carbon deposits, then there is compression det where the progressing flamefront overheats the endgas and it lights up causing a collision of flames, and there's also spontaneous det (or dieseling) whereby the whole lot just goes up because the mixture has reached its autoignition point. None of these are very engine friendly and should be avoided.

It is also worth pointing out that the spark is normally ALWAYS timed before the piston reaches top dead centre, typically between 8 adn 45 degrees BTDC in a Subaru engine. This is necessary because the mixture takes a finite amount of time to burn. It is important to get the peak cylinder pressure to occur between 10 and 20 degrees after top dead centre, if it happens any sooner then the piston has to "fight" against the expanding gas, if it occurs later, then less useful work is extracted from the combustion products and more energy is blown out of the exhaust as heat. Now, since it takes a finite amount of time for the charge to burn, it needs to be lit up before the piston reaches this 10 to 20 degrees ATDC region; the amount of time needed depends heavily on combustion space geometry adn compression ratio, but in the case of the Subaru engine, the timing will be about 45 degrees BTDC on high vacuum (the amount of charge drawn in when there is vacuum in the manifold will be low, hence the likelihood of fuel and oxygen "bumping into" eachother is also low, so it takes proportionately longer to burn). As the pressure rises the burn time decreases and so does the required spark advance. At 4000 RPM and atmospheric pressure it is down to about 37-38 degrees. Now as the boost builds the "perfect" timing continues to reduce but the effects of compression det become problematic. At 4000 RPM and 1 bar of boost, the "perfect" timing is probably about 28 degrees BTDC, but it's very difficult to get the spark at 28 degrees BTDC without it detting What is happening is that there is a combination of effects conspiring:

The mixture is being compressed by the piston; anyone who has ever held their hand over the end of a bicylcle pump and compressed the air inside, then let it out while compressed will have noted that it comes out rather hot. Now imagine that it's already partly on fire, this will tend to want to make the remaining lot go up spontaneously. This is exactly what happens in the engine when it dets due to much spark advance. So why does delaying the spark help? Well, it actually reduces peak cylinder pressure and temperature. Remember that the piston will be working against an expanding mass of gas; if we delay this expansion then the pressure and temperature seen before the piston reaches TDC will be reduced. Once the piston has reached TDC, it's pretty plain sailing, because as it moves back down the gas will naturally cool off a bit (just like the air in the bicylcle pump does when the plunger is allowed to return without venting the air).

Today, pretty much all production engines are a compromise. In order to make good thermal efficiency, an engine needs high compression (ideally between 14:1 and 15:1 but that's rarely achieved). But if it's got high compression then its tendency to det will also be high (barring some exotic combstion space geometries developped by EnDyn in the states). So manufacturers choose a compression which will give decent thermal efficiency in normal use, while also allowing close to "perfect" timing at full bore. Perfect, or MBT, timing is that timing which produces the maximum amount of torque for a given combination of RPM and load and must be found by running an engine on a dynamometer while adjusting the spark advance.

The example above of 28 degrees BTDC at 4000 RPM and 1 bar boost is a guesstimate, because I have yet to get that far. Perhaps with very high octane fuel it may be possible, but with 97 RON fuel I'm able to get to about 24 degrees before things get "interesting" but it's possible to tell that "there's more to come", if only it were possible to advance the ignition more. So, in this instance, Subaru have sacrificed top end torque to get better fuel economy. The default Subaru spark advance map runs about 20 degrees BTDC at that point, so they are quite conservative, and 8 degrees away from MBT will be between 6 and 10% reduction in torque, which is a reasonable compromise. In order to get achieve MBT timing, the compression ratio would need to be lowered to about 7.8:1 from 8:1 but this would impact the thermal efficiency on cruise, denting fuel economy somewhat (and wouldn't be doing the emissions any favours either).

Dieseling is perhaps the most severe form of detonation, although detonation is perhaps not the correct term. I doubt that the conditions in the chamber ever get intense enough for a detonation. Under normal operating conditions it is simply controlled combustion; what we call det is probably just explosive deflagration, which has a propagation speed substantially faster than combustion but with detonation it goes supersonic (some conventional detonations, ie non-nuclear, can reach Mach 10!). Dieseling can be provoked by too much boost, what happens is that the whole mixture overheats due to compression related temperature rise and as it reaches autoignition temperature the whole lot goes up in one big bang. This can shatter a piston, conrod or crank! If it were to reach detonation then the crank and rod would survive but the piston would be polverised (literally turned into dust).

Surface det, or pre-ignition is problem because a glowing deposit can light up the mixture too soon. This in itself is not so much of a problem, but it can lead to problems just in the way that too much spark advance does under high load. It can also cause dieseling in the remaining gas, which is a very bad thing.

This is just a very brief overview of what can go wrong, and doesn't even touch on the factors which affect RON requirements and the dependancy of knocking on fuel mix composition and the affect that the mixture composition and swirl and squish (or quench) areas can have. For more info I can thoroughly recommend the book "Internal Combustion Engine Fundamentals" by John B Heywood, but be warned it's a bit "in depth"

Hope this helps,

Pat.
Old 21 March 2001 | 08:53 AM
  #6  
NITO's Avatar
NITO
Scooby Regular
 
Joined: Nov 1999
Posts: 1,001
Likes: 0
Wink

Excellent stuff Pat, learn something new everyday. Thanks for taking the time and effort to clear that all up!


Naf..with regards to symptons of det..again here are just some of my observations; If your car suddenly feels a lot slower but it's boosting the same as before then chances are your ecu has retarded the timing as it has sensed det. This could happen for example if you were running above normal boost and then changed fuel or did some driving that causes a lot of heat soak like a trackday or something. UK cars can retard timing more than STi's so are fairly well protected against det in that they can usually retard it enough to prevent det on uk fuel (97ron). STi's on the other hand can retard timing enough on jap market 100ron but not enough for 97 or 95 ron hence the need for octane boosters etc and the advisability of a remap.

Usually you probably wouldn't be able to hear det, which is why many fit a knocklink or an EGT meter.

I've only ever heard det on my old Aprilia two stroke motorbike a couple of years back which sounded like a stuttering noise, (engine blew up pretty shortly after too, I think it was down to a different type spark plug), almost like hitting the rev limiter in a way. I think it can also sound a bit like a diesel too but a more metallic noise (not to be confused with piston slap).

rgds
Nito
Old 21 March 2001 | 01:36 PM
  #7  
DJB's Avatar
DJB
Scooby Regular
 
Joined: Jan 2000
Posts: 516
Likes: 0
Post

I've not heard pinking on my current MY98 but older cars I had in the past used to do this under certain circumstances. It occurs at low revs (ie in the wrong gear, too high) with the engine under load (going up a hill) and with the accelerator fully depressed. It is a tinkling sound which disappears as the revs rise.

As an aside, since it has not been mentioned, water injection (as fitted to some STi's) is an effective method of preventing det (by cooling the charge) and thus allows more ignition advance or higher boost.

D.

Old 21 March 2001 | 06:48 PM
  #8  
rnr's Avatar
rnr
Scooby Regular
 
Joined: Mar 2001
Posts: 243
Likes: 0
Question

To go from NuL-SuLwould i need to reset my ECU and does it make a great deal of difference? Car MY00 Rich
Old 21 March 2001 | 06:56 PM
  #9  
DJB's Avatar
DJB
Scooby Regular
 
Joined: Jan 2000
Posts: 516
Likes: 0
Post

rnr

The ECU on MY99 onwards is able to adapt to a change in the quality of the fuel much more quickly than older models. So there is no need to reset the ECU in your car if you are changing to SUL.

Does it make a difference?

Difference is slight but fuel consumption seems to better and I think turbo boost comes in slightly earlier.

D.

[This message has been edited by DJB (edited 21 March 2001).]
Old 21 March 2001 | 07:05 PM
  #10  
rnr's Avatar
rnr
Scooby Regular
 
Joined: Mar 2001
Posts: 243
Likes: 0
Thumbs up

Thanks Doc.
Old 22 March 2001 | 08:13 AM
  #11  
NAF's Avatar
NAF
Thread Starter
Moderator
 
Joined: Feb 2000
Posts: 4,523
Likes: 1
From: Malta
Post

Nice 1 Nito & Pat
Loadsa info for me !

Cheers
Niall
Old 27 April 2001 | 05:22 AM
  #12  
Legacy GTB's Avatar
Legacy GTB
Scooby Regular
 
Joined: Nov 2000
Posts: 66
Likes: 0
Unhappy

Detonation damage looks like this....



Old 27 April 2001 | 07:18 AM
  #13  
Sigmund's Avatar
Sigmund
Scooby Regular
 
Joined: Mar 2001
Posts: 128
Likes: 0
From: Norway
Unhappy

Those pics didn't work...
Old 27 April 2001 | 07:35 AM
  #14  
EvilBevel's Avatar
EvilBevel
Scooby Regular
 
Joined: Oct 1999
Posts: 3,491
Likes: 0
Post

Pat, thanks a bunch for that info.

Would you care to elaborate on why you can also have "lift-off" det ? My Knocklink only ever gave me 1 full christmas tree lightup in 5 months, and that was at lift-off after some heavy trashing.
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
Mattybr5@MB Developments
Full Cars Breaking For Spares
28
28 December 2015 11:07 PM
Mattybr5@MB Developments
Full Cars Breaking For Spares
12
18 November 2015 07:03 AM
WrxSti03
Drivetrain
0
30 September 2015 10:24 PM
charlesr
General Technical
9
28 September 2015 09:16 AM
bluebullet29
General Technical
2
27 September 2015 07:52 PM


Thread Tools
Search this Thread
Quick Reply: Detonation & knocking ?



All times are GMT +1. The time now is 12:25 AM.