A few basic questions here, but please respond only if you really know the answer through (a) trying this yourself or (b) you can present at least some logic to back this up.

On a N/A car, does running lean predispose to knock? (if so, why?)

Further, can you create *more* power by advancing timing on a car that runs lean without predisposing to knock?

Thanks!

-----------

Those are the questions; below is my logic which you can ignore if you know the answer already (but if you can identify errors, it would be appreciated).

The basis for these questions is my understanding of the pressure wave speed (burn rate) that builds after the spark in the cylinder. As an example, say around 12.5 AFR, the burn rate is optimal (fastest). Let's say the timing that produces the most power at this mixture at 3000 rpms is ignition at 25 BTDC. In other words by igniting at 25 BTDC will result in a pressure peak that occurs at the optimal point as the cylinder is headed back down (ATDC) to create the most torque.

If you go leaner or richer, the burn rate will be lower and the peak pressure point occur later. So in a car say running a 15 AFR, the peak pressure point will actually be much later ATDC. Advancing the spark (BTDC) should cause the peak pressure point to occur at a more optimal point ATDC.

I understand that under boost, the increased mixture density will burn faster and timing retard is required to prevent the peak pressure from building too soon ATDC (which would lead to knock).

But in a NA car, the mixture density should be low and I wouldn't expect that to affect the point at which the peak pressure builds in the cylinder. So screwing around with timing (within reason) on a NA car should be pretty safe, right????

 

1. The 350z tends to knock a little from the factory. It has knock sensors however, and the knock is not a harmful or devastating event. I believe this info is listed in the owner's manual...

2. The 350z runs generally stoichiometric from the factory, except under full throttle where it gets richer as the RPMS rise. Running stoichiometric is optimal for emissions - that is why they do it. The problem is, even stoichiometric is sort of "lean" in that not all the air molecules will be used up in combustion. Why? Because the combustion is not 100% efficient. So, running at stoich. with more advanced timing can cause knock because that un-combusted air can suddenly make the remaining fuel in the cylinder "pop" ... it's not that bad unless you start to get serious pre-ignition. But that shouldn't happen with your factory ECU/knock sensor.

On that note, the 350z seems to "like" ~13.0 AFR, which is a little "rich" ... but most likely, adding this little bit of fuel compared to stoichiometric helps to make sure all the oxygen (and oxygen is the scarce resource in your engine) gets combusted. Adding fuel will ensure a slower / more even burn, and no "pop" or knock in the cylinder. Adding fuel also cools the cylinder because when fuel goes from liquid -> gas, it causes dramatic cooling. Think about how much energy it takes to boil a pot of water. Well it's the same thing with fuel. The fuel droplets get "boiled" and evaporate, which absorbs tons of heat from the surroundings.


3. If you can compensate and add some fuel, advancing timing a couple of degrees is safe (because of your knock sensors) and makes power. I would not advance timing unless you have the capability to modify fuel also.

Finally, advancing timing more than a couple degrees on these cars is not necessary and (probably) won't make (much) power. For max power, you want peak cylinder pressure at 15 degrees after top dead center. This is done by literally "timing" the spark correctly. Typically advancing a couple of degrees will bring you closer to this magic number, but more than that and you might overshoot and lose power (not to mention it might not be safe for your engine).

In short... don't touch timing on an otherwise stock ECU tune, unless you are running 100+ octane fuel. If you have the capability to modify fuel curves (UTEC for example) then advance timing slowly (ON A DYNO by the way, never on the street) and go for peak power. Light knocking is ok and your ECU will adjust. If you are getting too much knock, you'll need higher octane fuel, or to add fuel, or to pull timing.

 

Wired > great info - thanks a ton for the detailed data.

If I'm reading between the lines correctly, the VQ has a tendency to "knock" from the factory because factory-tuned excess air ignites with unburnt fuel on its own, independent of the controlled burn. A little preignition is the norm??? I presume this would occur because of [standard] high cylinder temperatures???

If that is not correct, I don't understand what is causing the "standard knock". Or put another way, I still don't understand why a marked lean condition would cause knock. The spark should burn *most* of the air and fuel. If a cylinder is running lean, all of the fuel should be used up and the excess air should not normally cause an issue unless it is igniting before the spark, right?

 

Why 15 ATDC and not 7 or 22? Thanks

 

I presume he knows this through empirical (aka dyno) testing of the stock cylinder/piston/crank that at that point peak pressure would generate the most torque. Of course knowing this value is not too helpful because you have to provide the ECU with timing BTDC when to fire the spark and there is no mathematical formula that I know of that accurately provides this as it would be dependent on mixture density, AFR, temperature, and engine speed.

 

Is there a sticker saying peak pressure at 15 ATDC, or a pressure sensor inside a cylinder, or perhaps ECU can figure it out based on other parameters? Thanks

We can only speculate because on this board those that know are not talking

 

i don't think knowing the value really helps cause what can you do with that number?

to my way of thinking it's really the concept of how to calculate the best timing based on measureable data that would be really helpful. most tuners would just use a dyno to take a look at the end result of a timing change so even the experts may not really know the answer to your question...

i was (am still) hoping a professional tuner would provide a little insight...

 

The problem is that each injection and air gulp is minutely different.......the exact count of molecules of oxygen and the exact count of each of the hundreds of chemicals in gasoline - so each firing is different called combustion scatter*. This scatter causes varing temperature inside cylinder which added to above results in a constantly varing flame speed from one firing to the next assumming constant rpm...........in gets worse in acceleration.

Measuring during or after the fact results in a probablity for where to set the next firing [almost always wrong].

The ecu averages a bank so for sure the next firing is wrong.

Most engineers use a range of the peak pressure point say 14-16 degrees after TDC.

*coherent anti-Stokes-Raman scattering (CARS)

http://www.eng.ox.ac.uk/ice/publicat...nginePaper.pdf
http://www1.eere.energy.gov/vehicles...engine_2a1.pdf
http://www1.eere.energy.gov/vehicles..._engine_2c.pdf
http://www1.eere.energy.gov/vehicles...r_adv_cidi.pdf
http://www.eng.ox.ac.uk/ice/publicat...06-01-1198.pdf
http://me.engin.umich.edu/autolab/Pu...e/P2005_11.PDF

The above thousand pages should help your understanding

 

Thanks for the links; interesting data and measurement techniques but I did not find them that helpful (and other than the DOE annual report, the combined page count for the rest is under 200).

They did not answer my question at least.

The answer to Dovla's question I think you basically replied as "no one can really predict exactly", which I think seems reasonable.

So anyone else with thoughts?

 

I would assume any scientist would use the 420+ pages as an introduction to look up all the references at the end of each document.

The first question asked has a simple answer: running lean raises the temperature which increases the propensity to knock [because all fuels have a knock index which varies with temperature] why oem use extra richness to lower combustion temperatures since the unburned gasoline adsorbs heat

 

thanks for the answer, which makes sense to me.

...and looking through references at the end of published documents (particularly government reports) must truly be the most backward, inefficient way to learn ever conceived of! This is the Internet age, hence electronic knowledge interchange, quick searches, etc. Google anyone?

And any *real* scientist knows that the majority of literature out there is little more than skewed data and half truths published for the sake of advancing a career here and there... (sorry for the cynicism; the articles you referenced with the exception of the one on catalytic converter substrate materials looked at least somewhat appropriate)

 

One may pay for and read the millions of pages from the SAE all automotive engineers give their right Nxt to get published.

http://www.sae.org/automag/
http://www.sae.org/automag/techbrief...1-115-1-14.pdf

 

The techbriefs are fascinating - thanks.

This one talks about Porsche's variable geometry turbos:

http://www.sae.org/automag/techbrief...1-115-2-22.pdf