Leaning out - how lean is it still safe?

Discussion of tuning specific to MAZDASPEED3/MAZDASPEED Axela/Mazda 3 MPS vehicles
Post Reply
mituc
VersaTuner guru
Posts: 1323
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Leaning out - how lean is it still safe?

Post by mituc »

Right now I have a solid tune which I've got from Lex@StratifiedAuto and I run it as is with just a few tweaks from me which added somewhere between 5 to 8bhp (0.5-1deg more timing from 5500rpm up top, and just a touch more boost past 6000rpm which I can see and benefit from only in above 10C or warmer weather) . So not too much.

However, when running so much boost (my throttle reduction MAP limit is set to 270kpa up to 5500rpm and 273kpa from 6000 up, and add 5kpa for the entire range for the fuel cut protection) I can't seem to be able to hit my target timing values not even in cold weather. The boosted air temps are in general 15 to 20 degrees Celsius higher than the ambient but due to the extra air entering the intake the engine coolant temperature increases above 90C, sometimes hits 96C in 0-10C ambient if I do several pulls with like 10 seconds delay between each other.

I also noticed that my AFR's dip into the low 10's even if the min desired lambda is set to 0.7 (10.29AFR) and the target AFR is around or above 10.7 (10.5 is the lowest at 7k rpm which I never reach, 10.7 at 6500, and 11 at 6k). And I'm pretty convinced that this happens because of the occasional 0.35KR (sometimes 0.7) I see around 5500-6000rpm (I can rarely see that above 6k rpm, like 6300rpm, but these are pretty rare occurrences).
However, at this point I'm not sure if it's the fuel I'm using and its knock limits, or if it's mechanical noise which I think normally leads to even higher KR values, not just the minimum (the radiator bushes are kind of worn, but there may be some other things loose there as well which I may not be able to see).
However, since I'm running that rich, sometimes 0.3AFR richer than target and since my MAF cal is spot on (commanded is pretty much equal to actual) I thought that maybe I experience that light knock because of the richer AFR's which are commanded so apparently.

A while ago I thought that running just a bit richer it clears out light knock. Now, after reading this: http://www.innovatemotorsports.com/resources/rich.php I came to learn that having light knock while already running pretty rich is a dead end, ECU will try to go even richer which can make it even worse if it's the too rich mixture causing it.
Another thing maybe the SWAS. I'm holding the steering wheel straight during these pulls and even if I have a pretty good feedback from what the wheels are doing on the asphalt I have no idea what the sensors may pick.

Now, I've decided to up the minimum desired AFR with 0.1 (from 10.29 to 10.4) and up the target AFR's with 0.2 everywhere on the OL tables (WOT or PT) above 5000rpm.
I did not flash this map just yet, it's winter and I can't even get the car out from the mountain of snow where it's buried under, but I thought a bit of research will not hurt.
I'm not after too much more increase in performance because I'm on a stock block/head, but I'm after like 0.5 maybe one more degree of timing. I mean actual timing, because I already target pretty much what I'm after. Maybe someone can come up with an idea that would lead me to a solution and fix it myself before I go back to Lex for more serious stuff.

My mods are in the signature and I'm running only on 93-94OCT pump gas.
A sample log file is attached.
Attachments
VTO-tune-fueling.csv
(27.66 KiB) Downloaded 553 times
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
Kevin
Posts: 129
Joined: November 18th, 2011, 10:09 am

Re: Leaning out - how lean is it still safe?

Post by Kevin »

Im not the most clued up but will give it a shot. Regarding the drop in AFR; the car will always default to the min desired lambda value when a series of kr's is detected. Its always a good idea to have ur afr knock present and min desired lambda as the same value.
Tuned with Image by Image
mituc
VersaTuner guru
Posts: 1323
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Leaning out - how lean is it still safe?

Post by mituc »

The knock AFR tables are stock. The values there are a bit lower than the no-knock ones. And min desired lambda is even lower.
Does it make sense or should I just make the min desired lambda table and the knock AFR tables equal to avoid variations? Even though it makes much more sense, at least to me, how they are now.
By example at 6500rpm min desired lambda is 0.7, OL PT/WOT knock AFR is 0.722 and the no knock for the WOT AFR tables is 0.728.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
User avatar
joey
Posts: 51
Joined: November 2nd, 2011, 8:12 am
Location: South Afirca

Re: Leaning out - how lean is it still safe?

Post by joey »

Dude I get the same.. Its like the ECU is not acting on commanded min lambda. My min lambda is set to 0.755, but I see 10's.

I'm dialing in my timing and its taking forever. 95 unleaded fuel is ballz.. We have no other option here in SA.
Versatune,FMIC,3"SRI,3"TIP,HKSSSQv4,3"Turboback,Devil'sOwnWMIStg2kit,AutotechHPFP,JBREGRdelete,JBRBSD&Baffleplate,RacingLogiccoilovers,TRZTMM,AWRPMM,SURERMM,SUREShortcut&Counterweight,SUREg/knob,ApexiTT,HDIStg2Evoboostcontroller
Post Reply