Tuning the Timing Table
#1
Tuning the Timing Table
I see a lot of posts recently talking about the virtues of dialing in the timing across the board. Most of the howtos I come across deal with on WOT timing, or maybe idle timing, and pulling any where you might experience KR. What are the steps to tailoring your timing table to your cam for best performance and part throttle response?
#2
Interesting question. From what I've read, most people start at 10* for nitrous and boosted application across the spread. Not sure where to start for a bigger cam...most likely need more fuel.
#3
Originally Posted by NOSjohn
most likely need more fuel.
I figure most of the work gets done to the high octane table... But then do you try an duplicate it in you Base spark tables? What about IAT correction? Would this be the reason I get surging and bucking when I go from sunshine to shade and vice versa (parking garage at work)?
#4
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,604
Likes: 0
Received 6 Likes
on
6 Posts
My method was, to repeatedly drive around logging
and view the spark retard max histogram, and pull
out 1 degree for any observed retard while adding
2 degrees anywhere there wasn't. After enough
iterations and eyeball smoothing this will get you to
where you are right on the edge of ping. Then take
out about 3 degrees across the board and you're
about tweaked up. You want to keep an eye on
the O2s and tip-in vs steady ping so you aren't
using spark to cover transient short fueling and
like that.
Then copy HO to LO and subtract 4 or so, for
bad gas.
Use the IAT/ECT tables to pull out additional timing
that only shows up under bad conditions, based on
what you see for those two PIDs when ping shows
up (if they look responsible).
and view the spark retard max histogram, and pull
out 1 degree for any observed retard while adding
2 degrees anywhere there wasn't. After enough
iterations and eyeball smoothing this will get you to
where you are right on the edge of ping. Then take
out about 3 degrees across the board and you're
about tweaked up. You want to keep an eye on
the O2s and tip-in vs steady ping so you aren't
using spark to cover transient short fueling and
like that.
Then copy HO to LO and subtract 4 or so, for
bad gas.
Use the IAT/ECT tables to pull out additional timing
that only shows up under bad conditions, based on
what you see for those two PIDs when ping shows
up (if they look responsible).
#5
ChrisB's method mirror's what is done by the "Safeguard" system for non-computer cars. Add an amount of advance accross the board by scaling the table (5 degrees is recommended) and then log data. See where the PCM is knock retarding (pulling timing) and then adjust those cells accordingly. Re-log and verify until there is no knock retard.
Trending Topics
#8
In the EFI 101 class, they do a demonstration using a load dyno. He performed a method similiar to ChrisB's. He kept adding timing for a given cell until the car's torque began to fall off. It was interesting to note that he could keep adding timing after peak torque for a given cell without getting knock. The difference was that torque was no longer at it's peak, and the car was that much closer to the "knock zone". For example, at 2000 RPMs and 50KPA, the car made peak torque of 280ftlbs with 28* of timing. The car would take upto 34* at that cell without any knock, but torque was down with anything over 28*. In the class, they used a Honda and were doing real-time tuning, so it is a little harder on our cars. However, I am not for sure that simply backing up 1-2 degress after getting knock is the best answer.
#9
Kleeborp the Moderator™
iTrader: (11)
I was screwing around with the timing on my car, and decided to stop shortly after I started. I bumped up timing in different spots, only to find that I needed to knock it back down (I was using a KR histogram). I ended up changing to values that were less than what I had started at, and was still showing knock...it was definitely beyond me. I put my old table back in, and have been fine since.