Smoothing your tune with graphing
Aug 11, 2008, 08:08 PM
#31
It was nothing...only took an hour before I told my O.C.D. to STFU and stop tweaking it,lol. If anything, fuel maps will need fine tuning for real world results. There are so many fuel corrections that it forces the base map to look screwy sometimes.
Last edited by Jack_of_Trades; Aug 11, 2008 at 08:11 PM.
Aug 11, 2008, 08:41 PM
#32
Damn this OCD!!!! I think I have my timing map where I want it now,lol. Was talking to a car/computer geek friend of mine and he made a great point why this is so important. The ECU looks "ahead" to see what the next value should be. If the value leaves too far of a gap, it needs to do a mathamatical calculation to make the transistion as smooth as it can. At 7000rpm the ECU has a MAXIMUM of around 17ms to figure out the next calculation or it repeats the last known value. This can cause a hiccup if the ECU had too large of a calculation to determine. Add in the fact that the ECU is trying to calculate info from the load AND the RPM cells all at once and its amazimg that the ECU can even do it so well,lol.
By smoothing the maps so no calculations are really needed basically ensures a smoother transition every time. This is why it seems like it takes a while for the ECU to recover from a knock event before it gets back on track. This calculated delay makes MORE knock events occur until it gets back on track. Ever seen a momentary knock count of 4 or so and it took almost 3K RPM until it tapered away? Thats why.
By smoothing the maps so no calculations are really needed basically ensures a smoother transition every time. This is why it seems like it takes a while for the ECU to recover from a knock event before it gets back on track. This calculated delay makes MORE knock events occur until it gets back on track. Ever seen a momentary knock count of 4 or so and it took almost 3K RPM until it tapered away? Thats why.
Last edited by Jack_of_Trades; Aug 11, 2008 at 09:13 PM.
Aug 12, 2008, 07:31 AM
#37
I have just been out logging mine again.
Seems it's happy enough with the smoothed timing in 1st, 2nd, 3rd, 4th and 5th if I ramp up through the gears.
It is fine in 1st, 2nd and 3rd if I load it low down all the way to the rev limiter.
But it isn't happy in 4th
As it comes on boost, it starts to pickup some knock, but then it stops as it gets to full boost then picks it up again 1000 rpm or so later and will hold it to the rev limit
Only 3's and 4's so nothing really serious but more than I want it to do
So for the time being i've gone back to my original timing map, but I have left the smoothed fuel map there as it seems to work pretty well.
Aug 12, 2008, 10:10 AM
#39
Evolving Member
iTrader: (1)
Join Date: Jun 2005
Location: Gulfport MS
Posts: 213
Likes: 0
Received 0 Likes
on
0 Posts
I really like what you guys are doing here... I'm surprised to see the stock map so jaggy.
Do you limit your smoothing to some % of the total range of all the cells in a row or do you just move the values until it's smooth?
It would be cool if you could "lock" known good cells and smooth using those.
Do you limit your smoothing to some % of the total range of all the cells in a row or do you just move the values until it's smooth?
It would be cool if you could "lock" known good cells and smooth using those.
Aug 12, 2008, 10:41 AM
#40
Evolved Member
iTrader: (90)
Join Date: Jul 2005
Location: Roselle, IL
Posts: 1,917
Likes: 0
Received 0 Likes
on
0 Posts
Jamie and I were talking last night and this came up so I'll chime in. The percent used in the AFR smoothing is 0.0572 (for the most precise you can get). He also mentioned using two decimal points rather than one for the fuel figures (ie. 10.87 instead of 10.8) It all helps in making the maps as smooth as possible.
Aug 12, 2008, 11:41 AM
#41
I really like what you guys are doing here... I'm surprised to see the stock map so jaggy.
Do you limit your smoothing to some % of the total range of all the cells in a row or do you just move the values until it's smooth?
It would be cool if you could "lock" known good cells and smooth using those.
Do you limit your smoothing to some % of the total range of all the cells in a row or do you just move the values until it's smooth?
It would be cool if you could "lock" known good cells and smooth using those.
Jamie and I were talking last night and this came up so I'll chime in. The percent used in the AFR smoothing is 0.0572 (for the most precise you can get). He also mentioned using two decimal points rather than one for the fuel figures (ie. 10.87 instead of 10.8) It all helps in making the maps as smooth as possible.
The ignition map can't be altered since the car can only make 1* incremental changes anyhow. I wish our cars could do 0.5* increments like the subies
This is one of those "It can only help" things. Went to go test out my final map but my spark plugs have had it and I'm getting knock from 5K and up. The timing isn't any higher up top so I know its the plugs.
.
Aug 12, 2008, 12:47 PM
#43
Newbie
iTrader: (2)
Join Date: Oct 2007
Location: MIAMI,FLORIDA
Posts: 30
Likes: 0
Received 0 Likes
on
0 Posts
i will like to thank you bro(jack of trades)!!,i started reading this thread since a day ago and i been smoothing my timing and fuel maps and o man what a difference!!!! the pulls just feels alot smoother and just commons 1 knocks here and there,the car feels stronguer know,i even advance the timing 2 degress on peak boost and no knock!!,before i coulnt even advance it once degree
Aug 12, 2008, 12:48 PM
#44
Evolving Member
iTrader: (1)
Join Date: Jun 2005
Location: Gulfport MS
Posts: 213
Likes: 0
Received 0 Likes
on
0 Posts
while we're real technical i had another thought...
is the resolution shown in the 3d view the same as what's really in the ECU? If so then the ECU is going to be interpolating the values when it's between points anyway. Either that or it'll just use the value at the closes point. I'm sure smooth maps are happier than rough ones but as far as the ECU is concerned it doesn't matter wether it's interpolating between 2 similar values or 2 very different values. A flop is a flop if all that changes are the values of the numbers. Obviously changing from say 20% of max value to 80% of max value from one cell to the next would be terrible for the engine but the ECU itself doesn't care at all.
It's just as difficult to find the mid point between 2.2 and 2.3 as it is to find it between 2.2 and 88.8.
I can't wait to get a cable and start playing with the software. With the AEM you are so insulated from all this stuff. Just push + or - to adjust a point on the graph.
is the resolution shown in the 3d view the same as what's really in the ECU? If so then the ECU is going to be interpolating the values when it's between points anyway. Either that or it'll just use the value at the closes point. I'm sure smooth maps are happier than rough ones but as far as the ECU is concerned it doesn't matter wether it's interpolating between 2 similar values or 2 very different values. A flop is a flop if all that changes are the values of the numbers. Obviously changing from say 20% of max value to 80% of max value from one cell to the next would be terrible for the engine but the ECU itself doesn't care at all.
It's just as difficult to find the mid point between 2.2 and 2.3 as it is to find it between 2.2 and 88.8.
I can't wait to get a cable and start playing with the software. With the AEM you are so insulated from all this stuff. Just push + or - to adjust a point on the graph.
Last edited by Rob_GPT; Aug 12, 2008 at 12:50 PM.
Aug 12, 2008, 01:10 PM
#45
Newbie
Join Date: Jun 2006
Location: Issaquah, WA
Posts: 88
Likes: 0
Received 0 Likes
on
0 Posts
I had this same question as well...
I've always smoothed out my maps (OCD style) just so that the transitions between timing to timing or afr to afr was nice and smooth to eliminate potential knock or bad tuning... but I wasn't aware of the ecu calc cost differences.
I've always smoothed out my maps (OCD style) just so that the transitions between timing to timing or afr to afr was nice and smooth to eliminate potential knock or bad tuning... but I wasn't aware of the ecu calc cost differences.