Hi Jorge T,

I've seen again and again that people throw fuel into the engine to combat knock. Some with A/Fs in the low to mid 9's. Nobody ever seems to ask why OEMs are'nt doing that based on the knock sensor input. OEMs retard timing a few degrees to combat knock. Instead they could as well just richen up a little, right?
But no OEM is doing that.
But on the forums I read, everybody seems to have a 'better' method to combat knock than the OEMs, namely "run it richer".

There is such a thing as rich knock. It has to do with knock chemistry and how it works to create knock. When the air fuel mixture is exposed for a certain time to high pressures/temperatures, the gasoline HC chains break down and form aldehydes, cetanes and so on. Some of these resulting chemicals have much lower activation energies (ignition temps) than the gasoline they originate from. Running too rich, with consequently slow burn, and consequently higher required ign. advance, exposes the yet unburned gas in the chamber for a longer time to high pressures/ temperatures before the flame front reaches them. That means there is more time for the knock precursor chemicals to form and you get knock. Even though you are running "safely" rich.
The perfect example is an engine that gets "lugged". Most engines knock then even though there is no boost as of yet. But in this case there is plenty of time for the knock chemicals to form.

The "lean spool" people in this thread are talking about is another example. Leaner A/Fs (more toward the max power AFR of mid 12 to 13) extract more energy from the fuel with consequently higher exhaust temps and pressures, which helps the turbo to spool up faster. It spools even faster if ign. is retarded a little at that time, because then less energy is extracted by the piston from the burning gas and therefore more is available to spool up the turbo. It's a balancing act where you loose some here but possibly gain more somewhere else.
Some modern OEM Turbo ECUs for example will run at A/Fs in the mid 12s for a few seconds at WOT before they richen up. The reason is to get high initial acceleration, but if WOT is sustained, prevent buildup of heat-soak on the plugs, head and cat.

- Klaus

 

I almost always see knock in areas of the AFR that has become rich for some reason or another (adjustments to MAF smoothing, AFR map, methanol conc, ect). I never had this before when my timing was way too advanced which required me to run 11-11.5 AFR furthermore 100% methanol. Back then adding fuel actually reduced knock activity.

I have also found that when timing is more optiamal like (12.0 - 12.5) this I am able to run less concentrations of methanol which allow me take advantage of the superior benefits of H2O over the alcohol and I usually dont see knock on the lean side until 12.8:1 (but I do on the rich side of 11.5:1).

This is also with BPR9ES (three steps colder than stock), which I have had no problems with thus far (6000 miles).

Just sharing what I have seen with my setup if it means anything (I usually get bashed for pointing some facts I see, one guy suggested my wideband was off for being too lean )

 

i just did the 2 byte mod to the 0008 map and under misc. with it set to developer mode on the left hand side under misc. it doesn't show the "00" and "01". I can load ecuflash and my saved flash ok with no problems but i dont see the "00" and the "01" im supposed to see. um Help?

 

I did the mod on my ecuflash and when I tried to enter the new values in 00 and 01 they would not take. then I'd get a error. And I'd get the report error message and ecuflash would shut off. code is copied right.

 

/\/\/\ update to the newest version of ECUflash

 

I'm running version 1.32.1399 of ECUFlash and it's also not showing the new maps.

 

It works just fine with the new version. Make sure once you install the new version that you RE-MOD your XML file as the older XMLs are NO longer compatible.

 

A418t81, thanks for the response.

I actually found my problem was something completely different. I had created an "XML Backup" folder inside the Evo folder and made a copy the original file there keeping the original name. Once I changed the name of the xml file in my backup folder my modified xml file (in the Evo folder) with the new maps loaded correctly.

Thanks for posting the working file, however. It was comparing yours to mine (and having yours fail when I tried it) that lead me into looking for alternative causes for my problem.

 

help me out here guys,

so you are saying whatever has been added to the ecuflash file will not work with the new 1.3x?
Do we re-copy and paste what we added before or does it have to be written different?

but everything from the original that has been altered via tuning remains viable.

 

Alright, I have a problem. Everything went smoothly but when I go to ECUflash and click on the 01 2 byte load box to enter the 35164 it only dispays a box with <.. inside and won't let me edit it?

 

Nevermind, Im an idiot. I entered unit at the end of the code instead of uint

Just out of curiosity what boxes need to be checked or unchecked to run the 2 byte load?

 

Which boxes are you talking about? In ECUflash or your logging program?

 

Bringing it back from the dead...
I'm trying this on my 94170015 rom, and after entering the updated changes to the .xml file (since I'm running vs. 1.33) it shows up with the values "26870" in the 00 2 byte load box, and "25078" in the 01 box. Does this seem right? I guess if worse comes to worse I can at least for now use the calc load method or something similar in LogWorks to get a rough idea.

 

did you do step 5?

...add these to logworks protocols.xml
<datachannel id="#00" name="loadmsb" selected="1" units="load" min="0" max="255" expr="x" type="0"/>
<datachannel id="#01" name="loadlsb" selected="1" units="load" min="0" max="255" expr="x" type="0"/>

Enter this formula in tools>formula calculator to get MUT2byte load after you datalog the 2 above.
MC(MUT2byte_load;%;0;300) = ((loadmsb * 256) + loadlsb)*10/32

 

Yeah, I have those added in to the LogWorks .xml file and the calculator. I'm going to try the logging tonight to see if it works, so I'm not 100% yet.

On a side note, I tried adding the calc load formulas that were posted earlier in this thread, and I ended up with this...something has to be way off, there's no way I'm hitting a 350% load...