Detonation - what does it sound like, for the uninitiated?

[elnevio]

Detonation, knocking, det, pinking, pinging - all various names for the process where the air-fuel mixture is compressed in the cylinder to sufficiently high pressure that it combusts prior to the spark plug discharging, thus exploding too early in the combustion cycle and potentially damaging the engine.

So I know the process, but how does it sound?

I found a post by Robotnik a while back that described it thus:

There are people on here who know a lot more about this than me, but I'd have thought that at 0.8 Bar or ~12psi you could get away with a tank or two of 95 without any real longterm damage. When you're driving, have a listen and see if you hear detonation - it sounds a bit like a stone rattling around in a can - diesels make the same noise. If it's making a lot of noises like that, then stop driving the car.

I'm hoping that by posting this up now, people will be helped in the future with diagnosing some noises they may hear from their cars from time to time.



As it is, I am the one that needs clarification at the moment, but here's the blurb:

Saturday morning, Tesco had no 99 in, so I had to get some other variety of super unleaded. I stopped at Esso and filled up with Supreme 97. Never used it before.

I then drove 350 miles, not using WOT at all. No problems. I refilled with Esso 97 due to circumstances later on too.

While driving last night with the window down - gave it the beans and I felt what seemed like a very slight misfire, accompanied by a slight reduction in acceleration, and what I now think may be detonation. I was wondering at first whether it could have been a dodgy spark, plug, lead or coil pack. But it's only at 4-6,000 rpm or so. Also accompanied by a faint intermittent bearing rattle-like noise, but only sounds/feels as if it may be happening in just one or two cylinders.

I've used 97 before, albeit in Shell V-Power form (only once though since I've been running 15 psi), without issue, but I can imagine the Shell being much more superior - backed up by the number of users of it on here.

So - I only get it at WOT, as far as I can tell.

My guess is that this is due to the octane not being high enough to sufficiently defend against detonation. Weird thing seems to be that the timing doesn't feel like it's really being retarded though?

Anyway, my plan of action is:

1) Hook up EvoScan to confirm what knock events are occurring - though I still haven't done the pin 1-4 modification on my lead. Due to impending holiday (i.e. lack of time!), I can't see this happening; so
2) Be gentle, and run this tank of fuel down;
3) Fill up with 99 once my fuel level has dropped below 1/4;
4) Then see if I get any more knocking.

If it still does, then it may be time to investigate the various ignition components (looks like my tools are coming on holiday with us!) for dodgy connections.

If it doesn't, then problem solved! It seems too coincidental not to be the fuel.


Alternatively, could I add some octane booster to the remainder of this tank, just to see if it holds it off? Not a fan of additives, but any recommendations, one way or the other? Any comments on my symptoms are also invited!

[AlanDITD]

I people have already pulled logs using fuel less than 99 ron and seen knock count of 100+

Sometimes using less is una voidable. If you have to i turn the boost right down and try not got to WOT unless i really have to. Plus i put just enough in there to get me where i need to go. i.e home and back or to the next fillng station.

[orionn2o]

I can't say i've ever heard any knocking on my VR4, I have on previous cars though and its a pretty distinctive sound.

I would expect you should generally be fine if you're not boosting everywhere to use the lower octane fuels if there is no other choice. But certainly I wouldnt try to race an M3 after putting it in!

Octane boosters are not that great in my opinion. I suppose they are good that can stop det. however they wont increase the calorific value of the fuel so they wont help you with more power.

[elnevio]

No interest in more power (well, by the additive method anyway! ), just more to resist the knocking while I drain this tank!

I did wonder if the Leggie was just getting me back for using a whole tank of fuel pretty much entirely driving in the vacuum! "You must use me for what I'm meant for! Grrr - I'll show you..."

[orionn2o]

Or you could just not drive at WOT?

[elnevio]

And where's the fun in that! I'm off out in the Peak District this evening!

TBH - that's the current plan, as unfun as it is... let's see what later this week brings.

Actually - I've just had an idea:

How easy is it to siphon fuel from the VR-4 tank? It doesn't have anything in the filler neck to prevent siphoning does it? Thinking I might treat the Swift to some Esso 97!

[Eurospec]

Okay here we go........

Knock is a post spark event, although it is often easier thought and explained as pre spark. (Sometimes i explain it as pre spark as its easier to understand, but since ive started this journey......)

Pre ignition is a pre spark event. It melts stuff. It will likely only happen once and you will only know after it happens when your engine is in little melted bits on a trail back down the road that ends at your exhaust!

Pre ignition is caused by excess heat build up (and i am talking about LOTS of heat). Here the process is exactly as Nev describes. If you imagine when you compress something it gets hot. For example the entire need for intercooling. Imagine how hot it gets when you are compressing the mixture, and it spontaneously explodes, and you try to compress it some more! It builds loads of heat and melts stuff instantly.

Knock is a post spark event. Its caused when two flame fronts collide in the bore. The first is the nice burn from the spark plug which moves about 300 feet a second. The second is a spontaneous ignition caused by the heat generated during compression and trigered by the heat increase of the spark flame front. This explosive flame front (often multiple) moves at 3000 feet a second. The two (or more) flame fronts smash into each other with huge force and cause the damage that we asociate with knock. (Ie broken things!)

As to what knock sounds like, here is a relly good example.

http://www.phormula.co.uk/Player.asp...Analyser%20Pro

its a little advert for phormula products, but it demonstrates the sound that you hear when an engine is knocking. All engines sound slightly different, but the type of sound is typical.

Cheers,

Ben.

[orionn2o]

Nice info there Ben

[Nick VR4]

You might will also loose power and under engine strain it will get louder

[Davezj]

Okay here we go........

Knock is a post spark event, although it is often easier thought and explained as pre spark. (Sometimes i explain it as pre spark as its easier to understand, but since ive started this journey......)

Pre ignition is a pre spark event. It melts stuff. It will likely only happen once and you will only know after it happens when your engine is in little melted bits on a trail back down the road that ends at your exhaust!

Pre ignition is caused by excess heat build up (and i am talking about LOTS of heat). Here the process is exactly as Nev describes. If you imagine when you compress something it gets hot. For example the entire need for intercooling. Imagine how hot it gets when you are compressing the mixture, and it spontaneously explodes, and you try to compress it some more! It builds loads of heat and melts stuff instantly.

Knock is a post spark event. Its caused when two flame fronts collide in the bore. The first is the nice burn from the spark plug which moves about 300 feet a second. The second is a spontaneous ignition caused by the heat generated during compression and trigered by the heat increase of the spark flame front. This explosive flame front (often multiple) moves at 3000 feet a second. The two (or more) flame fronts smash into each other with huge force and cause the damage that we asociate with knock. (Ie broken things!)

As to what knock sounds like, here is a relly good example.

http://www.phormula.co.uk/Player.asp...Analyser%20Pro

its a little advert for phormula products, but it demonstrates the sound that you hear when an engine is knocking. All engines sound slightly different, but the type of sound is typical.

Cheers,

Ben.

so in simplest terms is a knock sensor just a small microphone strapped to the engine! i thought it was something complicated.

great clip and good info Ben.

[Nutter_John]

yep that is correct Dave , it is just a piezo microphone with some filtering ( around 6-7 khz )

[Davezj]

well i am pleased to say i learned something new today!

great info.

[Eurospec]

Yeah with the Phormula stuff you can set the frequency to filter out noise which isnt knock.

Knock frequency (ie the sound frequency) is determined by loads of stuff, but one of the main determinators is bore size. Once you know that you can filter pretty effectively.

Cheers,

Ben.

[Davezj]

is that cylinder bore or manifold bore or what?

the Knock sensor on the vr4 engine is located just to the side of the inlet manifold at the bottom on top of the block isn't it.
i suppose you could say it was in line with the top of the piston and in the middle of the "V" of the pistons, is that the best place for it.

[White Lightning]

I've used 97 before, albeit in Shell V-Power form (only once though since I've been running 15 psi), without issue, but I can imagine the Shell being much more superior - backed up by the number of users of it on here.

I am not sure where I heard it, but I was told Shell Optimax (the predecessor of V-Power) had an Octane rating of 98.3 RON so I would assume that V-Power was the same.

[Nutter_John]

bore of the block not manifold , goto the phormula site and you can work out the nominal frequency that knock would occur at

as Long as the mic is connected to the block then it should pick up knock , when connecting to the head it can and will pick valve train noise ( this is more so at higher revs )

[orionn2o]

I am not sure where I heard it, but I was told Shell Optimax (the predecessor of V-Power) had an Octane rating of 98.3 RON so I would assume that V-Power was the same.

I think Wayne due to the general instability of V power, the Octane rating will vary as the fuel gets older perhaps more than the lower value stuff.

They don't actually specify a value as it could be anywhere between say 99 and 97.

Thats probably why somedays your car feels like its gonna accelerate through the sound barrier, and other days not so much!

[Eurospec]

Hows this for a noodle baker then......

Generally, the stock knock sensors are excellently positioned and very very good. The filtering mechanisms are generally excellent etc etc etc. Sometimes though the stock ecu will pick up on knock that isnt really knock. If it does you will get ecu behaviour that is aimed at stopping knock, but in reality everything is fine.

A good example of this is on an evo. If you put GSC1 cams in it, the valve train noise increases quite a bit and because of the materials involved the frequency of the noise is smack on where knock is! As a result the stock ecu goes OMG, theres loads of knock, and pulls timing (thus losing power) when actually everything is fine.

Sooooooo, you can consider that in an engine that has been modified, but is running a stock ecu, you can get false knock. This is particularly so in forged engines because of the differences in materials. (Go bang your pen on the desk, then bang it on a metal filing cabinet. Same pen, different noise)

Cheers,

Ben.

[HPRULZ]

I had no choice on holiday because on the Isle of wight they dont even do 97 they just do unleaded or diesel so I had to run mine for a week on 95 and had no real problems, but then again I am not running raised boost either.

Kev

[White Lightning]

(Go bang your pen on the desk, then bang it on a metal filing cabinet. Same pen, different noise)

What if you have a metal desk?

Sorry ... I'll get me