I had a friend ask for
advice on using the dyno tune AFR calculations charts I made (with John L's
help), as well as for advice on dyno tuning.
I thought a few people might be interested in this, since I've never really
seen it spelled out anywhere. If people want supercharged advice I'll add that
in if people want it.
This is mirrored at:
http://carprogrammer.com/Z28/PCM/Dyno/Dyno%20Tuning%20Advice.htm
Disclaimer: This is a rant
from my twisted brain, and I was nice enough to spell check it, but I didn't
check for typos.
If it doesn't make sense,
read it again. If it still doesn't make sense try translating it into another
language, then back to english- that might help.
If something is just
downright wrong then correct me, my friends always tell me when I'm smoking
crack.
Hope this helps,
-Christian
I've got them on my website, although without
an explanation no one would know what they're looking at.
I've tested and used them to
nail my AFR from 13.5:1 to average 12.5:1 (probably leaner on the street)
http://carprogrammer.com/Z28/PCM/Dyno/AFR13.5.gif
http://carprogrammer.com/Z28/PCM/Dyno/AFR12.5.gif
Here's the chart to use for
12.5:1
http://carprogrammer.com/Z28/PCM/Dyno/dyno%20tune%2012.5%20to%201%20AFR.htm
and here's 13:1
http://carprogrammer.com/Z28/PCM/Dyno/DynoTune%2013%20to%201%20AFR.htm
It takes a lot of change to
make a difference of just 1/2 a point AFR ( like 6-7 %
PEvsRpm in my experience.)
To use the charts, I set my
PEvsCool to 16 in all operating temps, this takes out the variable there.
Run the car on the dyno with
a wideband and examine each of the data points at the rpm for the PEvsRPM
table.
i.e.. at 3600 rpm you have
a value of 1.0 (wild example) and the
dyno wideband said your real world afr is 14:1, so you're running a bit lean.
Find 1.0 in the leftmost
column and go over to the 14.0 column, the value to use is 18,
this should give you the change needed to have a 13:1 AFR.
If running dual widebands
and you see a disparity between the two sides I recommend looking at the
average of the two banks.
Now the guts:
I saw very little power
difference between 12.5:1 and 13.5:1, so naturally I was disappointed. In higher power cars it makes more of a
difference, and in supercharged cars it naturally makes a huge difference. I
expected to see something, but I only saw 1 ft-lb of torque between the two. I
rather run 12.5:1 on the dyno for two reasons:
1.) Since the dyno is 3000
lbs and my car is 3800 lbs, the car will be leaner on the street, maybe 13.1:1
2.) Since I'm running 11.4:1
compression I prefer to keep the combustion chambers cool, so I don't see
knock.
What does make a big
difference on the dyno is timing, and don't assume more timing is better.
If you're running a higher
compression motor, you in fact may need less timing, since the charge is more dense and the flame spreads quicker.
On the other side, if you're
running a higher overlap cam, with longer duration, then it may take more
timing to get the best performance.
This is again because the
charge is not as dense, so the flame speed spreads slower. The whole idea is to
get the charge to ignite at about 15 degrees After Top Dead Center, it take
different amounts of lead time for the flame to spread throughout the chamber,
so finding the optimum timing is the trick.
For a naturally aspirated
car (non-supercharged):
You'll adjust the timing by
modifying the 95 and 100 Kpa rows in the Load vs. Adv. 400-4000 and Rpm over
4000 tables.
Note:
The F-body stock tables max
at what I like to call "total timing" at 2,800 rpm with 33 degrees
advance.
The Y-body and most tuner
programs for slightly larger cams max "total timing" at 3,200 rpm
with 33 degrees advance.
Below total timing the stock
advance decreases at a rate of 3 degrees per 200 rpm, so you may want to follow
the same format.
Several tuner programs decrease
the advance at 2 degrees per 200, some even 1 degree. Most dyno runs don't have
data below 2,000 rpm because people usually aren't WOT that low.
You may want to run less
timing at low there to retard the WOT so you don't have major traction problems
when you floor it on street tires, then again you may want to leave the
traction control up to your right foot.
Here’s the stock y- body for
example:
400 600 800 1000 1200 1400 1600 1800 2000 2200 2400 2800 3200 3600 4000
90 KPA 7.0 9.0 12.0 15.0 19.0 22.0 24.0 27.0 32.0 34.0 35.0 35.0 35.0 35.0 35.0
95 KPA 5.0 6.0 9.0 14.0 17.0 20.0 22.0 25.0 30.0 32.0 33.0 34.0 35.0 35.0 35.0
100 KPA 2.0 3.0 6.0 12.0 15.0 18.0 21.0 24.0 28.0 30.0 31.0 32.0 33.0 33.0 33.0
How to experiment (after the
Air to Fuel is calibrated):
1. Make two base runs with
just the stock total timing, realizing the second run on a dyno is always
higher, since the carbon gets blown out of the ports and chambers. Keep an eye on knock, you don't want any
detonation, since the knock sensors pull major timing and
gives it back very slowly. You
should be able to make max power without being within 2 degrees of knock,
otherwise the sensor is getting "false knock"
2. Try a run with 1-2
degrees more total timing, see if horsepower increases/ decreases.
Try it also with 1-2 degrees
less timing, see what areas like a certain timing- again - avoid knock, your
power should start to decrease before the knock occurs, if not it may be false
knock.
If false knock, set the
knock vs. rpm table to 0, and tune 1 degree at a time, again watching for where
the torque starts to decrease- and it will decrease (since premature ignition
causes increased pressure in the head of the chamber, which makes it harder for
the piston to rise on the combustion stroke.) Knock sensors do go bad, so take
that into account, but don’t assume all knock retard is false, verify it by
seeing if the torque increases with 1 degree less timing, it should. There is a
window, like Ed said.
3. You may find the stock
timing is perfect for your car, many have. Others have had to add or decrease
timing depending on charge quality. If you end up with stock total timing is
dead on, then do what I do - nod your head and say "Damn those GM engineers
are good."
Hope this helps, always
share your findings/results
-Christian