Spark Tuning

[fatguy]

Hello All,
First off, the wealth of knowledge of the members here is AAAWWWSSOME. I can't stop reading the posts and answers to issues. My hat is off to each of you.

Second, I hope to tap into some of this knowledge. I understand MAF tuning, VE tuning, idle tuning, but now I am working on spark tuning and am wondering about the methodology of tuning. Is there a custom PID that can be made, or any tricks to dialing in the advance quickly? I have worked on a few vehicles now and I am getting up to 10 degrees of knock retard on 1 of them when the MAF and VE tables are properly set.

Thank you for any and all help
Corey

[eficalibrator]

The term you want to search on and do more reading about is "spark hook test."

[cptinjak]

The term you want to search on and do more reading about is "spark hook test."

Using this forum's search function and searching "spark hook test" will result in 3 threads; two of which are unrelated to this topic and the third thread being this one.

Here: http://books.google.com/books?id=vmq...page&q&f=false
On page 49, you describe spark hook testing. Short answer? You can't do it without a load bearing dyno. Period.




OP:

In theory, though, you could fudge it on the street. It will not be as accurate as a load bearing dyno, but it might get you pretty close.

The old "advance until you see knock then pull it back" mentality usually works for most "low hp" street cars, but it is a pretty crude way of doing things. The goal is to find MBT, or Minimum (spark advance for) Best Torque. You do not need to find out where it rattles to find out where it stops making power.



Using a software like Virtual Dyno, you could log a WOT run of your current tune for a baseline. Next, pull 2 degrees across the board and watch what the power does per RPM. Add 2 degrees across the board and see what happens. As long as the advances continue to make power, keep going in moderate steps. If adding timing at a given point does not make more power than the previous timing setting, change it back. You don't need timing for the sake of timing. Give yourself the cushion and use the LOWEST POSSIBLE advance that makes the most power.

Finding the knock limit and then retarding backwards will put you on the opposite side of that spark hook, quite possibly making the same power, but being that much closer to detonation.

At WOT, in high MAP pressures, and high RPMs you will find that the spark table will be directly related to your particular engine's VE. There is a reason you see that timing dip in the stock tables near where the engine makes peak torque: The higher the rpm for a given load (VE), the more advance that the engine will desire due to the increased speed of the piston vs the constant speed of the flame front. The higher the load/VE for a given rpm, the less advance that will be required for MBT due to the higher cylinder pressures creating a faster moving flame. You will have increasing engine load on the way up the RPM scale towards peak torque (just as the VE increases along this same path), resulting in a slightly declining timing scale. After peak torque, (highest possible VE) you will see the engine load go down, and more timing can be added back in approaching peak power and eventually redline.

THIS WILL AFFECT FUELING. Many people like to tune fatter at peak torque and lean it out toward peak hp, which is a way to safeguard against detonation that works in conjunction with a properly tuned spark table, but is another discussion entirely. This rich mixture near peak torque will lessen the effect that the VE of your engine has on the MBT. Due to its slower moving flame front, the richer mixture will require more advance than a leaner one would to achieve the same MBT for a given RPM/load condition. Therefore, with a steady lambda across the RPM range, you will see more drastic changes in your spark tables across that same range.

When you are satisfied with your new spark tables, go back and see if it affected the fueling, and how much. Assuming you are running a MAF, and these changes are mostly being made in parts of the spark table you only see in OL, then you will likely only need to check the MAF Hz table, and will not have affected your VE tables. Just always remember when changing fuel and spark, that one affects the other, and you can never assume that fuel stayed the same if you make spark advance changes, and vice versa.



In the end, it will never be as accurate as a dyno. To get yourself close, though, you need to make ABSOLUTELY SURE that your consecutive logs occur with the exact same road, starting location, gear, ambient temps, IATs, Baro, windspeed, etc, etc etc. The same spot on the road is critical, because even the flattest looking roads can be just the opposite.

The key to it, as it is with all tuning, is consistency. If you cannot find a road where you can do back to back to back 3rd or 4th gear pulls in the shortest time possible to keep the weather variables to a minimum while not upsetting the law or endangering others, then you might as well give up and find a load bearing dyno, IMO. Monitor all possible variables, like tire pressure changing and altering rolling resistance, or fuel consumption, as a few hours of this could burn enough fuel to skew the graphs due to a lower vehicle weight, falsely showing you more power. Heat soak, high IATs, high ECTs, etc etc etc.




Hopefully this helps somewhat. It is nothing more than my crude method of trying to do things "the right way" while not wanting to pay the price to actually do it the right way. Hopefully some professionals (cough cough ... Greg ... cough cough) will come in and correct any inaccuracies herein. I'm sure I made a few mistakes considering it is past 3am!



Jack