Official / Injector Scaling and Latency Forum
Nov 11, 2009, 07:34 PM
#17
My actual AFR's are about .5-1.5 leaner than the figures in the map (depending on load). Relatively easy to predict.
Is there another method that gives an exact 1:1 result? If so, fill me in! By the by, I have LS enabled.
Is there another method that gives an exact 1:1 result? If so, fill me in! By the by, I have LS enabled.
Last edited by lan_evo_mr9; Nov 11, 2009 at 07:40 PM.
Nov 11, 2009, 07:53 PM
#18
Evolving Member
Join Date: Apr 2008
Location: Reading, PA
Posts: 265
Likes: 0
Received 0 Likes
on
0 Posts
I think AFR_MAP MUT32 will give you a AFR with Hi and Lo Octane, LeanSpool, and Compensations done to compare your logged WideBand to.
Nov 11, 2009, 08:13 PM
#19
I usually use AFR for Size and the Fuel Trim for Latency, but my method is more trial and error. I'd like to see a method where you are off a %, then you change Size or Latency a certain %.
I think AFR_MAP MUT32 will give you a AFR with Hi and Lo Octane, LeanSpool, and Compensations done to compare your logged WideBand to.
I think AFR_MAP MUT32 will give you a AFR with Hi and Lo Octane, LeanSpool, and Compensations done to compare your logged WideBand to.
that's the stuff that's over my head....at least the way worded. Think of my way as "Injector scaling for dummies"
My theory is, get trims as close to possible, which equals great drive ability.. Then tune the AFR map, if it's different, then it's different.
I suppose it's just like driving styles, different styles can equal same results in the end.
I'm a VW tech, so I think of this stuff very trim like and not equation/formula like. I wish I knew more about the equation stuff....That would make me a perfect "driver".
Edit: Fast Freddie mentioned % while I was doing this "test", the % theory did not work with my injectors, could work with others and it did make absolute sense, but still didn't work with my setup.. Not saying that it couldn't work with others. If there was a definitive way of calculating all this very quick, it would save a lot of time an fuel. I just don't know how to do it. I'm a newb in that sense, and many others.
Last edited by lan_evo_mr9; Nov 11, 2009 at 08:17 PM.
Nov 12, 2009, 01:58 AM
#20
Edit: Fast Freddie mentioned % while I was doing this "test", the % theory did not work with my injectors, could work with others and it did make absolute sense, but still didn't work with my setup.. Not saying that it couldn't work with others. If there was a definitive way of calculating all this very quick, it would save a lot of time an fuel. I just don't know how to do it. I'm a newb in that sense, and many others.
I usually use AFR for Size and the Fuel Trim for Latency, but my method is more trial and error. I'd like to see a method where you are off a %, then you change Size or Latency a certain %.
I think AFR_MAP MUT32 will give you a AFR with Hi and Lo Octane, LeanSpool, and Compensations done to compare your logged WideBand to.
I think AFR_MAP MUT32 will give you a AFR with Hi and Lo Octane, LeanSpool, and Compensations done to compare your logged WideBand to.
So lets say that your WBO2 at WOT is reading 10.9, and the AFR map value is 12, you would increase the Injector size scaling and retest at WOT until it is closer, then afterwards get the fuel trims in check by adjusting latencies. If that is the route you are going it is going to be trial and error because each vehicle set up is different.
More scaling= Leaner
Less scaling= Richer
Nov 12, 2009, 03:15 AM
#21
Nice. great idea.
Dont know where this quote came from exactly, but it might be good to add.
Dont know where this quote came from exactly, but it might be good to add.
If low ~= mid & both > 0 then decrease injector scaling
If low ~= mid & both < 0 then increase injector scaling
If low > mid then increase latency values
If low < mid then decrease latency values"
If low ~= mid & both < 0 then increase injector scaling
If low > mid then increase latency values
If low < mid then decrease latency values"
Last edited by Evo_Jay; Nov 12, 2009 at 03:25 AM.
Nov 12, 2009, 05:40 AM
#22
That's a very cool idea on how to approach a very accurate/no guess work fuel map! Very cool. Is it possible to still have near zero trims with that method?
My AFR map is just a tad bit richer than actual WB readings, so would I lower or raise the scaling? And then reset latencies afterwards.
My AFR map is just a tad bit richer than actual WB readings, so would I lower or raise the scaling? And then reset latencies afterwards.
Nov 12, 2009, 06:21 AM
#23
That's a very cool idea on how to approach a very accurate/no guess work fuel map! Very cool. Is it possible to still have near zero trims with that method?
My AFR map is just a tad bit richer than actual WB readings, so would I lower or raise the scaling? And then reset latencies afterwards.
My AFR map is just a tad bit richer than actual WB readings, so would I lower or raise the scaling? And then reset latencies afterwards.
Nov 12, 2009, 07:15 AM
#24
That sounds right. I'm just not sure if I want to go through the whole "exercise" again and redo my AFR maps. After messing around with my car for so long, trying this, trying that, the one thing I have definitely learned is that to leave well enough alone. My trims are near perfect, great drive abilty, good starts, hitting the afr's I want under WOT, etc.
In other words, I think I'm done with this project. Lol
In other words, I think I'm done with this project. Lol
Nov 15, 2009, 01:01 AM
#25
So lets say that your WBO2 at WOT is reading 10.9, and the AFR map value is 12, you would increase the Injector size scaling and retest at WOT until it is closer, then afterwards get the fuel trims in check by adjusting latencies. If that is the route you are going it is going to be trial and error because each vehicle set up is different.
What I want to know is how to figure out latences myself without having to tune and reflash until I get an idle that just seems ok. There's usually some sort of formula for everything, but I can't figure this one out. From what I can see, the difference between the injector size in ecuflash and the actual size is 8% (560cc injectors show 513cc). I don't know if this was a coincidence, but when I tuned for injector scaling using LTFT with a PTE 680cc injector that it was also off by 8%. At that point this finding led me to believe that as long as the latencies are correct, the difference will always be 8%.
BTW: I'm using AEM's injector latency master sheet to get the injector latencies, but there are a couple of injectors that aren't on there.
Nov 15, 2009, 09:32 AM
#26
From what I can see, the difference between the injector size in ecuflash and the actual size is 8% (560cc injectors show 513cc). I don't know if this was a coincidence, but when I tuned for injector scaling using LTFT with a PTE 680cc injector that it was also off by 8%. At that point this finding led me to believe that as long as the latencies are correct, the difference will always be 8%.
BTW: I'm using AEM's injector latency master sheet to get the injector latencies, but there are a couple of injectors that aren't on there.
BTW: I'm using AEM's injector latency master sheet to get the injector latencies, but there are a couple of injectors that aren't on there.
stock roms for 5/6/7/8 jdm 542cc and different latency 11.8v down
Nov 18, 2009, 03:05 PM
#28
http://injector-rehab.com/kbse/lag.htm
1600s on gasoline are a bear. DO NOT ATTEMPT TO USE O2 FEEDBACK / TRIMS. Tune them in open loop and get used to 12.5-13.0 idle and odd tip in issues. Scaling will be in the mid 1300s.
E85 is marginally better.
1600s on gasoline are a bear. DO NOT ATTEMPT TO USE O2 FEEDBACK / TRIMS. Tune them in open loop and get used to 12.5-13.0 idle and odd tip in issues. Scaling will be in the mid 1300s.
E85 is marginally better.
Dec 28, 2009, 09:09 PM
#29
Evolving Member
iTrader: (3)
Join Date: Jul 2008
Location: Central Florida
Posts: 267
Likes: 0
Received 0 Likes
on
0 Posts
So I have a funky AFR issue and the answer based on my research is to rescale the stock injectors. Long and short of it is that once my LTFT is settled my car gets a whole point richer from 3500-5500 rpms.
Should I rescale my stock injectors from 513 downwards to decrease LTFT or is it better to use latency to decrease LTFT? If latency is the answer how do you calculate what values to enter or do people use trial an error and decrease the latency by the % the LTFT is over?
Should I rescale my stock injectors from 513 downwards to decrease LTFT or is it better to use latency to decrease LTFT? If latency is the answer how do you calculate what values to enter or do people use trial an error and decrease the latency by the % the LTFT is over?
Dec 28, 2009, 09:23 PM
#30
So I have a funky AFR issue and the answer based on my research is to rescale the stock injectors. Long and short of it is that once my LTFT is settled my car gets a whole point richer from 3500-5500 rpms.
Should I rescale my stock injectors from 513 downwards to decrease LTFT or is it better to use latency to decrease LTFT? If latency is the answer how do you calculate what values to enter or do people use trial an error and decrease the latency by the % the LTFT is over?
Should I rescale my stock injectors from 513 downwards to decrease LTFT or is it better to use latency to decrease LTFT? If latency is the answer how do you calculate what values to enter or do people use trial an error and decrease the latency by the % the LTFT is over?
And I already told you in your other thread, that you should COMPLETELY disable lean spool.
Leaving Lean Spool on at all will lead to "funky", inconsistent AFRs, just like your having.
Do yourself a favor and disable lean spool completely.
