Knock related questions and thinking

[Marsmik]

Hello guys,
this is my first thread on this board.
I'm thinking to the safe knock limit based on real experience.

This is mine:


After you can see a pic of a log of a full 4th gear on a dyno. 93vr4 , Stock boost, kn filter, catback exhaust, stock fuel system. After 5500rpm you can see up to 12 counts of knock and that with 100ron fuel..shell v-power here in europe which is the best pump gas we have I think 93-95 octane of USA fuel.
Anyone as some logs of the stock car in 3rd and 4th gear?..
After some years of experience, spark plug reading on my car I think that up to 10-12 knock counts is tolerated by the factory in the design stage of the 6g72 and that kind of knock mustn't scare no one.

ImageShack® - Online Photo and Video Hosting (http://img43.imageshack.us/i/rullata1knock.jpg/)

And after there is a pic of a 3rd gear on highway with 15 knock counts with this set-up:
13g turbos,stock intercoolers
fuel system: 450s, safc, methanol/water inj, hotwired walbro
Boost: spike at 19psi but not more than 15 to redline (this is the max for that turbos)
Fuel used in the run: shell v-power 100RON , toluene 10%, methanol in tank 5%, water/meth 550cc nozzle with 80 meth, 20 water
DAMN!! HOW it can knock with that fuel and that boost???

ImageShack® - Online Photo and Video Hosting (http://img338.imageshack.us/i/3rdgear.jpg/)

this second run is in 3rd gear because in 4th the speed is too high for my street and because in 4th at 5500+ rpm i have 20 counts of knock.

What do you think??

Watching spark plugs I can see very very little signs of detonation.. some very little spots on the porcelain and on the ground electrode..

[Simace]

Marsmik, If you really want someone to try and help you, you need to post up data logs. The graphs are nice, but are not good enough for what you're asking (by this I mean O2 at 0.91 would look no different than 0.86 on a graph, yet these two indicate different AFRs).

My speculation based off what I see, your knock - because it is continous at high rpm and under wot/load - is likely related to your AFR and possibly high EGTs.

Regards

[KeithMac]

What AFR do you target under boost?, my car with 98 ron + 20% Methanol mixed in the fuel tank does very well on my TD04-16g`s at 22psi, stock IC`s. Remember if you add methanol in tank you have to increase fuel delivery to suit.

I would take an occasional hit of 6 counts as ok, anything more is not. The spots on your spark plugs are pieces of aluminium that have neen removed from the piston tops.

I had a lot of knock problems on my car with straight petrol, could not run 12psi without a lot of knock.

Maybe try hotwining the fuel pump?.

[enigma]

Low down torqe will produce high knock counts, any runs should ideally be done in 3rd, do you have fuel controller fitted? if you have 450 injectors you need something to control them, also a fuel pump hotwire + an aftermarket FPR would eliminate some of the stock setup, one of the best mods is an AEM afr, it will allow you to tune the fuel delivery more exact, also allowing you to monitor the AF ratio on the fly.

[Marsmik]

My narrow band o2 readings are not accurate due to the fuel I use.. I also have the wideband and I'm in the 11.5 - 12.0 range with methanol on.
Looking at the timing is very similar to the stock one.. with 450s I've never heard of timing problems due to bigger injectors.
My fuel pump is already hotwired.
Max egt 820°C in 4th gear..

20% methanol in tank.. very interesting.. any risk of damage fuel system or engine?
I could use 20% methanol + 10% toluene + water meth inj 550cc min nozzle.. the car is not daily driven.
Maybe I could add some ATF to the mix as a lubricant.

[KeithMac]

A few people have said it`s a bad idea and it will destroy the fuel system, Subabru owners have been doing it for years with no ill effects and I`ve run M20 for a year now with no problems. I don`t add lubricant to the fuel/ meth mix.

Try it at your own risk.

[KeithMac]

My knock hot spot was around 4250 rpm which I believe is peak torque for my setup.

[Marsmik]

I use the car for about 1500km in a year..so it's really not a lot and I will try that for sure! The problem can be that the fuel rest in tank for weeks sometimes..

[Marsmik]

and my knock hot spot is at 5500rpm like a switch

[KeithMac]

I have mine sitting for weeks sometimes and it doesn`t seem to go off, would say if you do it in 20l fills or aim to use a mix within a month you should be fine.

[Marsmik]

but...
when you were injecting 2liter of meth per minute you were injecting 20% or more of meth and the results were not good enough..so the same 20% mixed in tank performs way better??
I have done a little bit of quick math on my current fuel set-up and i'm already sprying about 17-18% of meth over the total fuel.. I hope that the 20% in tank has better performance as i think.

[KeithMac]

It was too hard to get a consistant tune with the Meth being injected in high volumes.

Mixing it with the fuel in tank means I can get precise metering of the fuel mix at all times, not just on or off..

[Marsmik]

due to the fact that i'm at 100% idc i could add an afpr or go with 550s.. that is better I think..

[KeithMac]

Think 550`s would be better.

[Marsmik]

what about the lucas from ips at 299$ ?

[KeithMac]

He`s worked it out from the rpm and pulsewidth..

[Marsmik]

i don't know why but i read somewhere (don't remember exactly) that with 450s and tmo datalogger is common to see over 100% idc without fuel cut.

[KeithMac]

Really?, what's that got to do with anything, it's still incorrect. it's a effing computer, it spits out wrong numbers.

Steve

You have got to be fucking joking, words fail me..