That make it clear enough to do some math.
That is what Gm will be targeting for the coils at 14volts.
I see the current tables are way off of that.Volts RPM
0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000
14 3.43 3.43 3.41 3.34 3.24 3.13 3.02 2.91 2.80 2.69 2.56 2.44 2.30 2.17 2.04 1.91 1.82
temp multiplier
Volts Deg. C
-40 -25 -10 5 20 35 50 65 80 95 110 125 140 155 170 185 200
14 0.989 0.989 0.989 0.989 0.977 0.962 0.954 0.958 0.962 0.966 0.981 0.985 0.989 0.989 0.989 0.989 0.989
What I find best as coil response on my test car. Cranking dwell fixed at 7-10ms[based on coil type]. And using stock dwell times. Even slightly off gives some weird response and engine sound.
Now we come to the question. How hard will be to reconfigure the base dwell table to be rpm vs volts. And use the multiplier table for temp vs volts correction.
I can work with the current setup but it will be pain in the ass to dial the volt vs rpm, since it will be multiplier for the base table.
Just for the record. Is the commanded dwell have been compared with actual dwell with some kind of scope.
Now on the interesting part.
How the hell the coils survived.
On the gm chart. at 800 rpm I am looking for 3.42 ms *0.977 at 20*C and *0.962 at 80*C at 14volts
which is 3.32 at 20*C and 3.29ms at 80*C.
I guess these coils are temperature stable and the only correction is made between 20-50 *C with voltage below 12volts.
Looking at the log at 20*C the commanded dwell is 1107*0.004= 4.42 ms
and at 80*C we got almost double 1688*0.004=6.75 with some spikes on higher map upto 2013*0.004=8.052ms.
Now I extracted some data to see what we have at 6000rpm, 95kpa map at 80*C 1688*0.96=6.48192ms at 14volts
while gm targets 2.3*0.962=2.21ms at 6000rpm,14volts
Lets see whats happen at 12volts 2263*0.96*0.004=8.68ms
gm calls for 2.85*0.962=2.74ms.
I think the coils might really get fried on that wot pull, at least the dwell limiter will have suffered, with 20ms crank cycle at 6000rpm it was like almost 50% off-50%on. Getting some off time might help cool them out and recover.
I think we should rework the suggested dwell tables to some stock values and of course keep the max values measured by you for the max output enthusiasts.
Adding some fixed crank target will be awesome too.
I noticed the lookup tables do almost none interpolation between the cells. It takes fixed values from the closest cell. I am not sure if some interpolation is possible and how much burden will put on the processor. Maybe another future idea for improvement. I did notice on my tuning experience that engine don`t like abrupt changes, almost all values have some smoothing and the engine is happy and really smooth on transitions.
Bookmarks