04Green

iTrader: (1)

The Car, Pettit Rebuild.
The Tool, Access Port.
2004 (03 build) 6 speed.

The Process to date:
Did a MAF Scale, and Injector Scale following the Kane process. Seemed to all make sense, not real issues in the scaling department. No big number changes except around 2,350 RPM / 50 MPH. Sorted that to the swap between the mid and high LTFT around 10 to 20 gps. Once I got those two trims about the same, the weird swings have gone away.

The engine is "street ported". I have been updating the VE map to allow for the different characteristics. I am making progress, screwing up a little, but getting better. I think I found a problem in my method though. I have been assuming that if STFT was changing, that the ECU was still in closed loop. Today I started mapping Open Loop areas with partial load and noticed that if I compare commanded AFR to Actual AFR (Equ Ratio) I get some large changes that are DIFFERENT from the fuel trims even when the STFT is moving around. I think that means that I cannot rely on a moving STFT to indicate Closed Loop.

When I went to the Closed Loop Exit Load tables, I found 3. I have no idea the difference between them, but they all pretty much track a similar line marking out a quadrant of the RPM / Load table. And all of them say the car should definitely be out of Closed Loop in areas where I am seeing STFT adjustments. I think this means that the car is closed loop only in the little corner defined by the Exit Loads (below 4,500 RPM and 50% load). And, if it tells me that it is adjusting fuel trims outside this area is is pretty much lying to both me and itself. I think I want to believe the target AFR and Actual AFR number over the Fuel Trim outside the little corner. Right?

Also, as a note, when the car does go OL, the STFT does NOT go to 0, it just stays where it was.

And, for reference, the numbers I see at high load say the car is running about 20% lean.

Thoughts?

logalinipoo

If your STFT is changing you are in closed loop. In open loop you will still see your LTFT for your 20+gps range unless you have it tapered to Zero.

I've watched commad afr verses actual afr quite a bit on my goodbox. For some reason command is way off from What you are actually looking for. Maybe command is based on closed loop and it just ignores it once you are in open loop. It drove me crazy for a while then I started ignoring it, but I set all my open loop maps to 11:1 across the board and my actual was pretty close to my open loop maps, but command was looking for 13-14:1 range with WOT.

Now that I actually think about it a Bit I'd guess that's exactly what was happening. There is actually 5 fuel trim points but the upper points are when the car is in open loop range so you never see them.

Command Is what you car would like to have in closed loop alway's. When you go to open loop it still has a command, but it is not actually adjusting to reach it.



Brett, Kane your thoughts on that?

04Green

iTrader: (1)

Thanks for the reply.

How about this. Does the ECU will still play at adjusting trims, even when open loop, if it is near a closed loop boundary. It would not use the data, but would have it for a transition back to closed loop. At some point it says screw it, I am not going to try to play any more and just freezes trims.

Or, the likely difference, is that the fancy CLOSED LOOP A/F Target Tables are, maybe, used in closed loop, to modify the targets. Re-Running all the numbers..... Multiplying the AFR Targets by the multiplier in the tables.

Brettus

iTrader: (2)

https://www.rx8club.com/series-i-eng...3/#post4713513

If you read the above you will see that the CL switchover point at low rpm is controlled by the map mentioned there . You don't have access to it via the AP unfortunately so can't adjust the switchover point. So what you see happening with STFT is definately telling you when you are in CL vs OL but you are stuck with that ................

The maps you do have access to (AB & C) appear to only affect the switch-over at higher rpm when in cruise mode , not WOT .

TeamRX8

iTrader: (25)

The Cobb AP allows you to adjust the low RPM switchover point from CL to OL.

Brettus

iTrader: (2)

How about you explain exactly what you are talking about . If you are talking about the ABC or throttle maps you are quite wrong .
There is an ability to change it via the RPM exit parameter only.

04Green

iTrader: (1)

The low RPM seems to track very well, trim is pretty much equal to Actual AFR over Target AFR. As Load and RPM go up, and outside the box identified by the ABC tables above 50% load, above 4,000 RPM, the Trim starts to wander a great deal from what is calculated. My thought is that the TRIM were being correctly calculated and applied, the car would hit the target, no matter what. That is not happening.

I need to dig back into my data, but that chart seems to identify some of the area where I am staying in closed loop. However, the ECU royally sucks at hitting target AFR in the bigger zone. For example, I have points like 3,500 RPM, 85% load, a trim of -7% but when I check AFR I am 19% leaner than target. Inside the little box it tracks wonderfully, trim is pretty darn close (2% difference max) to Actual AFR over Target AFR.

But, in the 3,500 and 85% example, that point is outside the box the AP shows you and outside the curve on the Brettus chart from Mazda Edit. But, the STFT is changing merrily along every log sample. It is just wrong, compared to the hand calculations.

What I have found is that doing the math leads to much more stable numbers. There are RPM/Load cells that consistently hit the same calculated number, with the Trim number being all over the planet.

This may be related to the fact I have an 04. I think all three of you have newer cars.

My current plan is to start ignoring Trim outside the small box and see if I can dial out the lean spots using the math approach. But, that is a tomorrow thing.

Thanks everyone for the help.

TeamRX8

iTrader: (25)

That's exactly what it is and it does exactly what you're saying we can't control. I either don't understand what you're trying to say or you're not explaining it clearly

TeamRX8

iTrader: (25)

Oh, I see. You want to control the Exit Load parameter at low rpm. That's not really how it's stated. I don't really understand why it's needed.

Brettus

iTrader: (2)

. It's only s problem on some later models that have that map I posted set to 100%. That can lead to some lean situations on boost transition . For the op it isn't a problem . I was just answering his question.

TeamRX8

iTrader: (25)

Ok, gotcha. I was scratching my head on what it had to do here.

Mike, it's really hard to watch you continue to go about this all the wrong way when it's really so simple. When you decide it's finally time to listen and learn please let me know.

Left4Dead

delete wrong place

Left4Dead

I really need to stop doing this DELETE again