Porsche 911 UK Enthusiasts Online Community Discussion Forum GB

Welcome to the @Porsche911UK website. Register a free account today to become a member! Sign up is quick and easy, then you can view, participate in topics and posts across the site that covers all things Porsche.

Already registered and looking to recovery your account, select 'login in' and then the 'forget your password' option.

Detonation/Knock explained with photos

ragpicker said:
infrasilver said:
pjvenda said:
infrasilver said:
Your knock sensors should of picked it up and retarded the timing to accommodate.

But it might have taken a few milliseconds/ignitions to compensate, hence the bangs?

It will of, yes.

Rag,

Mite be worth talking to Ken to see why it's happening, could be running lean at a certain point in the map. How modified are you ie larger injectors?

I think it was happening due to a seized wastegate causing it to run very lean. It was in the lower revs that it happened though, say 3krpm. I've since freed the wastegate and its not happened since. I saw 1.5bar at the time (normally runs 1.1). The mods so far are in my sig.

Its not got bigger injectors at the moment although it will have in the next month. I've discussed my car recently with Ken (not about this particular issue) and he's been very helpful as usual.

Again, I'm more interested in how catastrophic det can be, and whether a single episode could be enough to ruin an engine, or whether (as demonstrated above by Mike) it is a gradual thing like water eroding the walls of a building. Also, how common is it? I personally have never experienced it in any of my cars before and they have always been modified.

Maybe its always been happening but I've not been listening.... :eek:



Rags, it is cumulative detonation that caused the damage in the photographs. The aluminium gets crystallised by the heat and local, uncontrolled combustion.

Your car pulled 'load' when the wastegate stuck, as demonstrated in your graphs. So, in short, it already knew it was detonating (hence the load pull) but ran out of range. At this point it became audible but the DME still has tricks up it's sleeve to protect the motor.....

Nicholas has done a TON of background digging in these ecus, I'm reasonably confident of the number of degrees of retard available under knock conditions, but Nick would be able to furnish you with the remaining pieces of the safety regime puzzle. Best to ask him on the specifics. And boy does he know his subject!

Bear in mind though..... A turbo engine by definition has more charge in the cylinders at the time of knock, therefore the potential for damage happens a lot quicker than it would in a normally aspirated engine like the Big Healey in the thread. So whilst the ecu did it's best to protect the engine, the loads placed on piston, cylinders, head gaskets, big ends and head studs etc, is huge during the knock event. Modern engines have great ecus on them, protecting from so-called Super-knock, a kind of run away knock which can only be stopped by resting the engine for a few cycles, in order to break the chain of events. This may have been instigated in yours, leading to the fact that you noticed rough running and turned it off.

Glad you found the culprit. Better still you logged the conditions under which it took place, excellent effort that man :grin:
 
Many thanks for taking the time to reply Mike. Very interesting as ever!

My dad - a mechanic back in the day - used to talk to me about engines 'pinking' when they had the wrong/poor fuel in them. He was referring to det back then. Obviously ECU technology has come a long way since then and more modern engines are able to adjust to avoid this.

Is it possible that a certain amount of det occurs in all engines which goes unnoticed by the driver due to it being only minor? Clearly with modified cars the potential is much greater for more significant events. Or are things so precisely controlled now that no det occurs unless something is not functioning perfectly.

There are so many variables which would impact on this which makes me think that perfect or 'ideal' functioning would be very difficult to achieve 100% of the time, which leads me to question the background incidence of det.

Funnily enough, I'm meeting up with Nicholas tomorrow to talk injectors amongst other things... Quite looking forward to picking his brains!

:thumb:
 

New Threads

Forum statistics

Threads
124,548
Messages
1,441,407
Members
48,961
Latest member
gpc
Back
Top