Jump to content
LegacyGT.com

How to address lean condition in open loop


xt2005bonbon

Recommended Posts

Busy dad and tuning noob here :spin:....

 

 

I noticed lately that in fairly low/mid load (0.70-.85 g/rev) with a 3900-4000 rpm range, my AFR is 15.2-15.7 consistently. Good thing (I think) is I am still in low boost (0-1 psi). But it obviously results in some knock and FKC.

 

 

I checked my 'Primary Open Loop Fueling' table in my current tune and see that the variable shown is an 'estimated AFR' (see attached). As you can see in the aforementioned load/rpm ranges, the estimated AFR is on the rich side. Actual does not achieve that.

 

 

 

attachment.php?attachmentid=276585&d=1560434412

 

 

Simple question: which table do you modify in order to achieve a descent actual AFR while in Open Loop? Do you modify these 'estimated AFR' values or some other MAF related variables? And let's assume I do not have any vacuum leak.

prim_OL_fueling_06.png.88d6e02214dc8e162abf3ddfd3833f40.png

Link to comment
Share on other sites

If it's running closed-loop, what is your AF learning and AF correction values in those conditions?

 

 

If it's running open-loop there, that means somewhere your fueling calibrations are off (MAF calibration, injector scaling, etc.) or your sensors are dirty/on their way out (MAF or front O2 sensor).

Link to comment
Share on other sites

I should have been clear. All comments in my OP assumes open loop scenario. And I also have just cleaned the MAF sensor.

 

 

 

My long term fuel trims A/B swing a bit between -4/+4%. So not too bad. However, my LTFT C/D are closer to 5% now. I do admit in the past that these were closer to 0%.

 

I checked for vacuum leaks and could not find any.

Car has now 164K. I believe MAF and 02 sensors are original.

 

I have been having issues with the MAF wiring harness a bit in the past. So that sucker is ziptied and I cleaned once again the MAF plug.

 

 

 

 

Mods to car:

SF cobb air intake with their air box

stock turbo with Grimspeed EBCS and AVO wastegate actuator

AVO intercooler

catless up and hiflow catted Invidia DP

3" mid pipe with stromung mufflers.

Link to comment
Share on other sites

Okay, so it's likely a calibration issue. Start diagnosing by logging Final Fueling and comparing that to your recorded AFR. That will tell you how far you are off. From there it could be a mixture of MAF calibration, load compensation, or other compensations (IAT compensation, most likely).
Link to comment
Share on other sites

'Final Fueling'? Is that the name of the variable to log? Did a quick google search on this term and I see people referring to 'estimated AFR'. I use btssm and do not see this variable listed. Custom PID?

 

 

I'm fairly certain BtSsm doesn't have it built-in... I also almost never use the built-in BtSsm gauges and use custom PIDs for all of my gauges anyway, since that allows different mins/max and step sizes, along with 2-byte and 4-byte options. It's "Final Fueling Base" in RomRaider lingo.

Link to comment
Share on other sites

mmh. I recall you may have been the one providing a link to some github page or something else where one can find info about what can or cannot be logged for a given ROM. Something like that... will have to dig...
Link to comment
Share on other sites

Okay, so it's likely a calibration issue. Start diagnosing by logging Final Fueling and comparing that to your recorded AFR. That will tell you how far you are off. From there it could be a mixture of MAF calibration, load compensation, or other compensations (IAT compensation, most likely).

 

 

OK. So I've been able to log the 'final fueling' parameter which is essentially what is shown in the table I attached in my OP (I was under the impression it was going to be some other parameter I was going to log).

 

 

So, say there are a few cells from that table that needs to be addressed (where true AFR is way off compared to what the table shows), what do you modify in the tune to address the issue? I've read that one could modify the final fueling values in the table so as to get the true AFR within the range you want (not necessarily the range shown in the table).

OR, you leave this 'target AFR' value alone and modify something else in order for the true AFR to hopefully be in the vicinity of the target value. Am I making any sense :lol: ?

 

 

Edit: now that I think about it, there must be an equation relating MAF and fuel injector related variables to the AFR...

Edited by xt2005bonbon
Link to comment
Share on other sites

OK. So I've been able to log the 'final fueling' parameter which is essentially what is shown in the table I attached in my OP (I was under the impression it was going to be some other parameter I was going to log).

 

 

So, say there are a few cells from that table that needs to be addressed (where true AFR is way off compared to what the table shows), what do you modify in the tune to address the issue? I've read that one could modify the final fueling values in the table so as to get the true AFR within the range you want (not necessarily the range shown in the table).

OR, you leave this 'target AFR' value alone and modify something else in order for the true AFR to hopefully be in the vicinity of the target value. Am I making any sense :lol: ?

 

 

Edit: now that I think about it, there must be an equation relating MAF and fuel injector related variables to the AFR...

 

 

See my earlier post:

 

Okay, so it's likely a calibration issue. Start diagnosing by logging Final Fueling and comparing that to your recorded AFR. That will tell you how far you are off. From there it could be a mixture of MAF calibration, load compensation, or other compensations (IAT compensation, most likely).
Link to comment
Share on other sites

Yeah I saw that but I never touched this stuff before. And obviously, to be safe, I would like just change one parameter at a time. So which one first? And by how much? My guess is first play with MAF calibration. I still don't understand that thing though, especially given the fact that I have an aftermarket intake. So I need to educate myself on that one... Unless you have some pointers :).
Link to comment
Share on other sites

At this point, you're probably best off reading up on tuning at the various online forums and actual books available (RomRaider forums is a good place to start). Tuning isn't a change one thing and magic happens, you really need to understand what you're changing and how the motor responds to it.
Link to comment
Share on other sites

Sure. I want equations though!! That would help a lot to show the relationship between the various parameters. When I find relevant links, I will post them here for others as well.

 

Some dimensional analysis of the parameters you're logging would probably suffice. But Cobb's got a pretty comprehensive guide that has the basic equations for Subaru factory ECU tuning (engine load, mass airflow, fueling, etc.): https://cobbtuning.atlassian.net/wiki/spaces/PRS/pages/26870189/Subaru+Accesstuner+Tuning+Guide

Link to comment
Share on other sites

Here's sort of some basic tuning info.

 

What you're doing is modeling airflow and fuel flow in the ECU. The airflow model is your MAF scaling. The fuel flow model is your injector settings. Injector settings take into account fuel pressure and assumes it is 1:1 rising rate with base fuel pressure of 43.5 psi.

 

If there is a discrepancy between target AFR (open loop table) and actual AFR (wideband) then there is something wrong with the airflow or fuel flow in the ECU not matching reality.

 

Is the injector scaling setup for the injectors you are using? Have you verified base fuel pressure? Does fuel pressure rise with boost? If fuel flow is doing what it is suppose to then there is an issue with the airflow calibration. 100% certain there are no boost leaks? The breather and PCV hoses can draw in unmetered air if they have leaks. These hoses tend to get overlooked and turn into brittle hard plastic when they are old.

 

You say you have an aftermarket intake. Has the car been tuned for it? Any change to the intake that affects how air flows over the MAF sensor requires a complete recalibration of the MAF scaling. This includes replacing the factory air box with a cone filter and/or changing the MAF housing. The turbo inlet itself may affect things, but probably not as much as the other things.

Link to comment
Share on other sites

Ok, so I fail at reading comprehension. I reread your first post and saw the area you're having issues with. For some reason Subarus have weird fuel delivery issues in areas between 3000-4000 RPMs. I had to deal with the same problem with multiple setups.

 

If you have no boost leaks and are using the Cobb map for the intake then it's safe to say your airflow calibration is probably pretty close. This definitely is a fuel delivery issue caused by some type of resonance in the fuel system. This is the problem people try to solve by installing longer fuel hose or changing the fuel dampers, etc.

 

The proper way to fix it is by directly injecting more fuel into the problem areas. The only way to do this is with the per injector pulse width compensation tables. If you look at these tables you can see Subaru already adds fuel between 2800-3200 RPMs (at least in my 2007 ECU.) These tables let you add pulse width directly without making a change to load. Some people use the load compensation table to fix this issue, but raising load to add fuel will decrease timing in that area. You can go into your timing map and increase the timing, but this is a workaround and doesn't solve the issue directly.

 

There is information and spreadsheets at Romraider.com about this issue. You basically have to datalog the RPM and pulsewidth when the problem happens and make changes to those areas. Or you can try the engine load compensation.

Link to comment
Share on other sites

Here we go again. I haven't thought about this stuff in a while and now it's all coming back to me. I'm really not trying to up my post count.

 

I remember now I was having issues with the car hesitating from leaning out when it would switch from closed loop to open loop. When I was using a MAF I used the per injector compensation tables to try and fix this. When I switched to SD I used the VE table to try and smooth things out (workaround similar to using load compensation table.) But I was always fighting fuel issues. If I got things dialed in at part throttle the car would lean out during a WOT pull between 3-4k RPMS. If I dialed in WOT, things would lean out at part throttle in that area.

 

What I ended up doing was using the closed loop and short term fuel trims to always adjust the fuel in those problem areas and then raise the RPM and load that the ECU would switch to open loop.

 

So, how does this work? First you have to understand how and when the ECU decides to switch to open loop. You have to look at two tables for this (assuming your closed loop to open loop delay is zeroed.) First is primary open loop fueling and second is minimum active primary open loop enrichment.

 

Anywhere in the primary open loop fueling that is higher (leaner) than the minimum active will be closed loop. Anytime you move to a cell that is lower (richer) than the minimum active table the ECU will go to open loop.

 

The thing is that the ECU uses a couple other tables to determine what the AFR will be when it is in closed loop. The numbers in open loop table are just used to determine when it is in closed loop.

 

Now look at the CL fueling target compensation tables. When the ECU is in closed loop this is added to the AFR number in whatever cell of the open loop table you are in. So I'm looking at my stock map and at .73 load at 3600 RPMs the open loop table says 14.7. So, I'm in closed loop. Now I look at the closed loop comp table. Same load and RPM is -0.257. The ECU is actually targeting 14.443 AFR.

 

We can use this to our advantage to maintain closed loop short term trims much, much higher up than Subaru does. We know the factory front o2 will read as rich as 11:1. It gets inaccurate once exhaust manifold pressure rises. Knowing this I set closed to stay on at a much higher load and RPM. Pretty much any stand alone worth a damn has complete full time closed loop fuel control, so why not try to use our factory setup to stay in closed loop as long as we can? You definitely need to have the D long term trim disabled for this to work right. You don't want these compensations to be learned and applied later on. Just short term trims, that's all. You also don't want to use fuel trims to cover up a bad calibration. You should try and get everything as close as you can, then let the trims get things perfect.

 

I included some pictures to show how my tables are setup. This resulted in a much smoother transition to open loop since I was always out of the problem areas when it switched. I would also always maintain target AFR during a low RPM WOT spool up or a part throttle low-medium boost trip through the map.

 

All this assumes your ECU has these tables. I've only ever seen the 2007 maps.

 

Time for bed.

1.thumb.jpg.1a769a6c4a5ac1f9601abae7c7e1e824.jpg

2.thumb.jpg.5690717b642499101d8671cf589678b9.jpg

3.thumb.jpg.730b5216911c46d67404be980979b5be.jpg

Edited by BrandonspecB
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...

Important Information

Terms of Use