PDA

View Full Version : Ignition Tuning?



AUTiger
08-19-2008, 11:58 PM
Okay, I have searched and studied and watched the k-pro videos but I can't see the upper most screen in the video example so I can't answer my own question. That and unless I am blind(feel free to link me if I am), there really is not much in depth material on tuning ignition like there is for fuel.

I have tuned my a/f from 13.0-13.5 and am working on the imperfections. I wanted to start playing with ignition tuning and today when I added 2 degrees and took away 2 from my normal calibration and then plotted them on thetadyno it showed my normal calibration being the best of the 3. However, when I was trying the ignition advance I had the edit all tables enabled and added 2 in both the hi and low cam profiles in columns 7-10. Now was that okay, or should I really have added 2 degrees in column 7-10 on both the 0 degree tables, datalogged and compared, and then go so on and so forth for the rest of the cam angles. In essence comparing 5 or 6 datalogs?

I know a lot of people would say take it and get it professionally tuned, but even Talon has said that "K-pro is plug and play, not plug and forget" I just want to get to know as much as possible, which is why I am asking before I did something stupid. Plus with just i/rh/e I don't have major mods to really require a professional tune.

talonXracer
08-20-2008, 07:20 AM
That is basically how you I do it, but I go with much larger jumps in timing at first and watch for knock during the run and if it gets bad I stop that run and pull a little timing in that specific area that started to knock(I also try adding a little fuel sometimes to see if that helps) and if there is no knock I add a little fuel and advance the timing and see if any more power is made, if not then I go back to the original timing and fuel(useing the undo function).

Then you blend all the ignition tables. I have seen Jeff Evans get it very close in about a dozen or so pulls, just because he has the experience. It takes me at least 30 or more pulls(too cautious,LOL)

AUTiger
08-20-2008, 11:13 AM
Thanks Doug. Two questions though. How large of jumps in ignition are you talking about, 4-6 or even more? And is it possible for you to explain the blending process a lit more in depth? If it's too time consuming for you to type out then don't worry about it. I would just like as much documentation on the process not just for myself, but for others wanting to understand street tuning as well.

edit: Okay, so after watching the videos on hondata's website several times I think I understand it now a little better. I am however still afraid of trying to add more ignition initially than just increments of 2.

Ba82Ep3
08-30-2008, 08:57 AM
I do it in increments of -/+2 myself. When im removing/adding ignition timing AFTER the initial change, i do it in increments of .5 and i do NOT do it to all tables simultaneously. I do it usually in rows of varying size depending on either a two lobe or a three lobe head... and if im using a VTEC window or just a switchpoint (that will change the type of cam map you have altogether in my opinion). I use the lambda overlay to give me an idea of where i have initial "play room" as far as fuel goes with the KAL. I choose that degree map of ignition and adjust accordingly.

AFTER i have made those changes, i look at the ignition map itself in 3D, and attempt to smooth any spikes or steep transitions, save the KAL separately of your original KAL, and then make a pull. Compare the new KDL to the modified KAL and then use the lambda overlay again. Look for knock, its frequency and its voltage. If your ignition change didnt significantly change knock count or knock voltage, then adjust fuel in the lambda overlay to half of what the cells suggest. (if it suggests -6% then do -3% for that cell and move on to the next) Save the KAL again and relabel it so you know you are progressing forward. Pull again and compare the new KDL to the second modified KAL. If your changes are negative in results... go back to the previous KAL (not the original), and start from there.

I look at it as a simplistic algebraic formula. x+y=z. X=fuel Y=ignition advance Z=cam advance. The three are connected and can be interchanged for your purposes... but just as you have to have three basics of air/fuel/fire for an engine to run... you have to have these three basics to tune... and varying amounts of each affect AND rely on the other. For example. 3000 parts of fuel (+) at 50 degrees of ignition advance (=) will likely NOT work at 5 degrees of cam advance WOT 6krpm in a basic NA setup. I think of things in a weird way though... :mredface:

If the lambda changes DIDNT increase knock or knock voltage with the +2 of ignition advance and the lambda overlay adjustments, then you know you are likely still in a safe zone on that degree map.

Be sure your target lambda is where it needs to be. If it isnt then your overlay will be off accordingly.

If you ever get lost on A/F and just where it is, or where the ECU WOULD have it if it was in control... plot open loop and closed loop A/F ratio in a separate window (two computer screens help here a lot)... the A/F line by the ECU will likely be smooth and consistent all the time. The bumpier your A/F line is... the more work you have to do. Targeting those spikes/dips can shorten your time behind the laptop and make tuning a lot faster. Your line however will likely NOT be as smooth as the ECU line because of cell resolution, regardless of dyno/street time.

If your KAL on the high cam goes as high as 8 or 9k and you never rev there... get rid of those and add lower RPM lines to increase your resolution midband (say 4 to 5k)... this will help your tune in a lot of ways... not only for power, but fuel economy too. (for instance... if you never rev a JDM k20a to 11,000 and your rev limit is at 8600... then have your last RPM line be 9000. The RPM lines you remove can help you find a little midband TQ and fuel economy with the ability to adjust with better resolution in that area)

KPro's cell count isnt as high as the original ECU map. If you look at your fuel maps in 2d, you see theres a bunch of lines connected by dots. The original ECU map would have many more dots between each of KPro's dots for greater adjustability (='s power/economy/smoothness). This applies to all of the maps (ignition/cam/fuel) and as far as A/F goes... its why its so hard to get a completely smooth line. (not to mention IAT and engine cooling variations compounded by TPS variations during part throttle tuning. Your foot cannot hit the same spot on the KAL EVERY single pull)

Kpro's cells are a good AVERAGE for tuning. So keep that in mind when you make large adjustments to either of them... and is partly why Hondata says smoothing the maps out during/after tuning will generally generate better results. There are ALWAYS situations that do not adhere to this standard.

Now get out there and burn up some fuel! :mcool: