AFR targets at loads higher than 2.50 ATR/ATP fix

[chaser27]

Originally Posted by Realgib3 I always always always had issues until I set my all my load limits to 5 It was just a weird thing where even when I was under the limits of 3.0, I still saw issues. I feel like I had other issues that showed up too, other than afr, but it's been so damn long I can't remember for sure.

Load based? Could it maybe have something to do with the table that tells the ECU how much to correct based on how far off it is from target? (can't remember name right now)

Sent from the lead

[superskaterxes]

the following logs are of the problem im seeing. I have seen this in 2 other cars besides these and they were all gen 2. The earlier revision letter log is the one where it barely kisses 2.5 and doesent have the problem while the later revision is with the AFR drop.

Car 1 is BT MS3
Car 2 is stock turbo factory freak.

i have not seen any correlation to fuel used or targets and all of my fuel tables are set identically. I also dont see the AFR always recover when it comes back down below 2.5.

[atvfreek]

Originally Posted by superskaterxes the following logs are of the problem im seeing. I have seen this in 2 other cars besides these and they were all gen 2. The earlier revision letter log is the one where it barely kisses 2.5 and doesent have the problem while the later revision is with the AFR drop. Car 1 is BT MS3 Car 2 is stock turbo factory freak. i have not seen any correlation to fuel used or targets and all of my fuel tables are set identically. I also dont see the AFR always recover when it comes back down below 2.5.

You know, I had a car do something similar to that and said fuck it and scaled the maf. Which fixed it going rich... Maybe @rfinkle2; hit it on the head, and it's the same scenario going lean. I may try scaling the maf on the car with the lean condition.

I wonder if some ecu's lose their OL fuel correction when over 2.5 load.

Gonna play around a little today once I get some logs from @resilientonel;

[Dimenus]

Originally Posted by atvfreek Cars I've seen it on actually go lean. @rfinkle2; I will dig up a log shortly

Edit: Removed. I'm probably wrong. I threw a system too lean code in the cold when it was ~ -10 degrees here. I've since reflashed the map and my LTFT have shifted almost positive on a 93 map. The code hasn't re-appeared though.

I've definitely seen over 2.5 load on the stock turbo. It's not sustained, but it does happen.

[rfinkle2]

Guys, are you using the load target comp / injector base offset fix?

[Lex]

Originally Posted by rfinkle2 Guys, are you using the load target comp / injector base offset fix?

Been using that for a long time and it's related to WGDC.

The injector phasing multiplier also has similar results if used.
@atvfreek; are you using this to gain some headroom?

Another note. The ECU will scale AFRs at WOT up to +/-12-15% which you will not be able to see since these trims are not logged.

If you go "outside" this range, the ECU may actually give up on using the trims.

Everything has to be more or less in range for the tune to work well. Moving load tables, compensation tables, etc. "out of the way" often has unpredictable effects.

[rfinkle2]

Originally Posted by Lex Been using that for a long time and it's related to WGDC. The injector phasing multiplier also has similar results if used. @atvfreek; are you using this to gain some headroom? Another note. The ECU will scale AFRs at WOT up to +/-12-15% which you will not be able to see since these trims are not logged. If you go "outside" this range, the ECU may actually give up on using the trims. Everything has to be more or less in range for the tune to work well. Moving load tables, compensation tables, etc. "out of the way" often has unpredictable effects.

Yes Lex, I understand that that fix is related to wgdc. Thank you.

I'll be tuning cars with no pride and continue flying by night.

[atvfreek]

Originally Posted by rfinkle2 Guys, are you using the load target comp / injector base offset fix?

Yes I do use those.

Originally Posted by Lex Been using that for a long time and it's related to WGDC. The injector phasing multiplier also has similar results if used. @atvfreek; are you using this to gain some headroom? Another note. The ECU will scale AFRs at WOT up to +/-12-15% which you will not be able to see since these trims are not logged. If you go "outside" this range, the ECU may actually give up on using the trims. Everything has to be more or less in range for the tune to work well. Moving load tables, compensation tables, etc. "out of the way" often has unpredictable effects.

No I do not use the phasing injector multiplier

[chaser27]

Originally Posted by chaser27 Now I just need to get fast enough to hit 2.5+ loads.

So apparently, this became relevant to me quicker than I had expected. I've been playing with my tune, it was cold this morning and bumped a 2.52 Calc Load today. Just like @superskaterxes; was seeing, my AFR goes rich as hell on a Gen 2 and stays there until it corrects itself. I'm attaching my log for anyone that wants to take a look-see and a graph that shows the ECU going WTF and fixing itself.

[rfinkle2]

related thread:

Random Lean @ WOT

[phate]

Originally Posted by chaser27 So apparently, this became relevant to me quicker than I had expected. I've been playing with my tune, it was cold this morning and bumped a 2.52 Calc Load today. Just like @superskaterxes; was seeing, my AFR goes rich as hell on a Gen 2 and stays there until it corrects itself. I'm attaching my log for anyone that wants to take a look-see and a graph that shows the ECU going WTF and fixing itself.

That looks more like the WOT trims running out of room, then suddenly kicking off (or on, not sure what the target vs actual is. Your axis scaling sucks ).

[chaser27]

Originally Posted by phate That looks more like the WOT trims running out of room, then suddenly kicking off (or on, not sure what the target vs actual is. Your axis scaling sucks ).

I'm targeting 11.4 for now while I figure out how things play with each other. And they have been hitting almost dead on until the 2.52 today

[Lex]

Originally Posted by chaser27 I'm targeting 11.4 for now while I figure out how things play with each other. And they have been hitting almost dead on until the 2.52 today

Did you verify that this is not the case when loads are below 2.50 with an A-B-A test? Almost looks like SWAS was involved there.

[chaser27]

Originally Posted by Lex Did you verify that this is not the case when loads are below 2.50 with an A-B-A test? Almost looks like SWAS was involved there.

Yes, I was actually just looking at it. Temp rose almost 20 degrees between logs and I only hit like 2.4 or so. AFR's are rock steady.

I can't wrap my head around it... maybe the phantom OL fuel trims @atvfreek; (I think it was) was talking about?
I don't want to do a MAF adjustment like @ms6mil; is doing, because I don't really hit 2.5 ever

Concerning SWAS, only if my sensor is acting up, dead straight road and slight uphill. Not really getting torque steer

Attaching log.

[Lex]

Start with the basics of setting your fuel tables all the same.
After this remove the injector scalars and phasers if they are not OEM.

[chaser27]

Originally Posted by Lex Start with the basics of setting your fuel tables all the same. After this remove the injector scalars and phasers if they are not OEM.

I actually just did that. I only had the OL/WOT no knock tables adjusted in the above logs. If I hit 2.5+ anytime soon. I'll report back. Thank you.

Sent from the lead

[mghtymouse380]

So to bring this back up. I was wondering if anyone had any insight into this. I have problems dipping way rich once it gets over 2.5 load but the tables that are in question are fixed.

Here are a 3rd gear and a 4th gear back to back (within five mins). And here is the map that I am currently running. Thanks for all your help

[rfinkle2]

Originally Posted by mghtymouse380 So to bring this back up. I was wondering if anyone had any insight into this. I have problems dipping way rich once it gets over 2.5 load but the tables that are in question are fixed. Here are a 3rd gear and a 4th gear back to back (within five mins). And here is the map that I am currently running. Thanks for all your help

I didn't know you were home.

I can continue to work on this for you if you want Cody. If you see the link above, it can be corrected by massaging the maf curve.

[mghtymouse380]

Yeah I have been home for a bit just been busy with other things. Seeing family and things over the holidays and haven't really been doing much with the car.

[chaser27]

I forgot to report back in this thread, but @Lex; after adjusting my MAF curve as if the area where it went rich were the fuel trims as if all OL fuel trimming went away, my AFRs were still spot on (I dropped my MAF ~13%). Another thing I noticed after the change was that I no longer went rich on tip in, but rather when I let off. I adjusted back the other way and it seems to be better on both ends now. However, my calc loads are now lower (where they should be, cause I'm slow) so I won't be able to see what happens over 2.5

Sent from the lead

[Lex]

Originally Posted by chaser27 I forgot to report back in this thread, but @Lex; after adjusting my MAF curve as if the area where it went rich were the fuel trims as if all OL fuel trimming went away, my AFRs were still spot on (I dropped my MAF ~13%). Another thing I noticed after the change was that I no longer went rich on tip in, but rather when I let off. I adjusted back the other way and it seems to be better on both ends now. However, my calc loads are now lower (where they should be, cause I'm slow) so I won't be able to see what happens over 2.5 Sent from the lead

The ECU has somewhere around 12-15% authority over fueling under WOT. So yes, it will adjust to meet the AFR targets even if your MAF cal is off based on the wideband feedback.

You can use this to drop yourself lower in load. if issues persist.

I think there are more than a single issue here and it might differ from one generation car to another.

[Jansi]

Been talking with @ms6mil; I've seen the same trend with my AFR's. Here's a screen shot of of a log right at the verge of hitting 2.5. I know I have a lot more logs around but this is what I found so far.

[ms6mil]

Originally Posted by Jansi Been talking with @ms6mil; I've seen the same trend with my AFR's. Here's a screen shot of of a log right at the verge of hitting 2.5. I know I have a lot more logs around but this is what I found so far.

Yeah Jean those are pretty negligible compared to mine, but you see what I was talkin about

[Jansi]

Originally Posted by ms6mil Yeah Jean those are pretty negligible compared to mine, but you see what I was talkin about

Yea I'm not too worried about them as my AFR's deviate at a minuscule amount compared to yours. At first I really thought maybe my calc'd % of E was jacked up but I'm glad you mentioned this thread and makes perfect sense now!

[Lex]

Originally Posted by Jansi Been talking with @ms6mil; I've seen the same trend with my AFR's. Here's a screen shot of of a log right at the verge of hitting 2.5. I know I have a lot more logs around but this is what I found so far.

It is common to see some AFR deviation during spool and I don't think this is the same issue. You can tell your injector duty continues to climb. In the cases where the cars go rich or lean there is a step change in injector duty where it either drops or increases.

[Enki]

I'm still using OTS values in those tables myself (if memory serves, I'll check tomorrow), and my fueling was almost 100% dead on balls accurate; now though, it starts to go rich above a certain RPM (note that I'm also using a reverse-taper boost curve, so airflow increases as RPM increases); I'm almost certain, though, that this is a result of me going from VTA to pulling inert gasses from the crankcase back into the inlet.

[ledfootrob]

Im getting the same problem on a Gen 2. As soon as I hit 2.5 load my AFR goes rich to the low 10's and stays there for the rest of the pull. 3rd Gear is fine as it only hits 2.4~ max but 4th gear is a different story.

All my fueling tables are set the same. And SWAS disabled. Im gonna try a E85 tune without meth on and see if I still get it to rule out meth enrichment.

[Enki]

Just confirmed that all my tables except for OL/WOT Commanded EQ (Knocking) are set to match (targeting 12:1 AFR). This table is set to 11.5:1 AFR, but the AFR doesn't start to go rich until after I drop back below 2.5 load:

http://www.mazdaspeedforum.org/forum/foru...ordeal-4th.csv

[SarcasticOne]

Mine starts doing it when i hit 2.4 (goes a bit leaner, just to 12.05 or 12.2 from 11.8 )

am targeting 2.35... was much cooler the other night than it has been recently... so i noticed it more...

will implement the fuel table changes and report back

[atvfreek]

Originally Posted by SarcasticOne Mine starts doing it when i hit 2.4 (goes a bit leaner, just to 12.05 or 12.2 from 11.8 ) am targeting 2.35... was much cooler the other night than it has been recently... so i noticed it more... will implement the fuel table changes and report back

The fuel table change isn't going to fix it. What is happening is there is OL afr correction and once it gets above a certain load target, the ECU gives up on it and the afr's either go lean or rich depending which way the maf cal is slightly off. Basically the maf cal has to be spot on when these higher loads are reached.

In your log, I see a +5.3 LTFT from 30g/s up. I would add 5% fuel to your maf curve starting at 30g/s all the way to the end of the maf scale. That is what I ended up doing on the couple of cars I've seen it on and it has worked great.

[Lex]

Originally Posted by atvfreek The fuel table change isn't going to fix it. What is happening is there is OL afr correction and once it gets above a certain load target, the ECU gives up on it and the afr's either go lean or rich depending which way the maf cal is slightly off. Basically the maf cal has to be spot on when these higher loads are reached. In your log, I see a +5.3 LTFT from 30g/s up. I would add 5% fuel to your maf curve starting at 30g/s all the way to the end of the maf scale. That is what I ended up doing on the couple of cars I've seen it on and it has worked great.

In certain cars this has fixed it so I think the ECU IS looking at different tables.

But I think you're correct as well regarding the MAF calibration and the car stopping to use the OL fuel corrections past 2.50 load in a lot of cases.

[snailD]

good discussion guys, sorry i have nothing to contribute. Utah loads are significantly less. I remember having issues tuning 801ms3's car however with fueling (pretty sure it went over 2.5 load) and just did maf scaling to get it dialed in, this was over a year ago now.

[Lex]

Also remember that tuners use different strategies to tune and this does affect how the logic plays out especially since we don't get visibility to the logic and don't see all the tables so there's a lot of interpretation that happens. The only time I have seen cars with consistent step changes in AFRs it has been fixed through these tables and more commonly through the use of MAF scaling instead of using the injector scalars.

The gen2 car I posted a log earlier from exhibited these issues and they were fixed by not using the scalars and loads actually went UP. However the MAF calibrations I am also quite careful with when scaling so the issue may not present itself because of this.

It is likely that the problem has several potential sources.

[Lex]

Just had a car exhibit this problem. It is a gen1.

The solution in this case was scaling the MAF to account for both E85 and then further scaling it to make AFRs be on target for above 2.50 load.

The procedure was simple since in this case the MAF curve was not piecewise scaled. So I had to scale the entire curve by ~8% and the AFRs are consistent above and below 2.50 load.

Attached a before scaling and after scaling log for reference. I have also updated the OP with this as a fix as well. Thanks for hints and insight on this everyone.

[superdave2335]

Shouldn't the OP be corrected since what you stated about correcting the issue was false?

[Lex]

Originally Posted by superdave2335 Shouldn't the OP be corrected since what you stated about correcting the issue was false?

All the solutions are listed in the OP sir.

[superdave2335]

Originally Posted by Lex All the solutions are listed in the OP sir.

If you say so. :-)

sent from the igloo.

[ledfootrob]

Thread revival time. I just noticed in my latest 25% E mix tune when my car hit 2.5 load and over my AFR's remained on target. This tune uses the the same maf cal from my other tunes which at 2.5 load will drop into the mid 9's AFR. Has anybody changed there Fuel Scalar on a pump gas tune minimally to see the effect.

[chaser27]

Logs?

[ledfootrob]

Damn I don't have a log from a pump tune which hits 2.5 to show it going rich but here is my E tune where it doesn't.