RomRaider

It turns out that the spreadsheet and MAF tool in romraider don't do a good job replicating AFR learning because the ecu doesn't seem to "weight" the data equally. Some data will cause AFR learning to adjust and others will rarely. I have always known this but I don't know the logic. The spreadsheet tried to emulate the logic but it was incorrect.

So... what I do is:

1. Change WGDC to zero to prevent blowing up engine.
2. Change CL/OL delay to half the stock value.
3. Change AFR Learning d to 70+ g/s and AFR Learning C to 60-70 g/s.
4. Cruise on the freeway at high speed.
5. Change the MAF scale between 70-80 based upon AFR learning D.
6. Change the MAF scale between 60-70 based upon AFR Learning C.
7. Cruise on the freeway at highway speed.

Keep repeating and dropping AFR Learning range until AFR Learning D is 40+ and AFR Learning C is 30-40. I have found that AFR Learning C between 30-40 is better. It keeps it more stable and allows a better transition into open loop.

There is one last step I do and that is to adjust the MRP compensation table but that's a lot more complicated.

Very curious to see how you adjust the MRP comp table.

FWIW I find that my WB completely agrees with OEM O2 sensor up to about 60-65 g/s, after that the two go their separate ways.

Is your OEM O2 sensor still in the up-pipe?

Do you mean Learning range B, LBGT?

I dont really understand why one would adjust the latency or scaling. The ECU will adjust to hit the proper AFR anyway (as best it can in closed loop). So if the ECU is adding say 3% in latency because the AFR showed a rich condition, then that rich condition is fixed; and vise versa. So to me it just seems like one is negotiating with the ECU to show 0's on the logger or learning view. Maybe im just way off base here, but I dont see a performance increase coming from this; after all the ecu was making the same corrections before to achieve the proper AFR right?

Or is the benefit from this adjustment only really seen in open loop? Which is why the wide band is necessary? So with the wide band one can fine adjust the open loop to better achieve the proper AFR, but only in open loop is when it would make a difference?

Yeah I feel pretty stupid...

You're mostly correct, but there is a little bit more to it.

The D trim (or E trim, if your car has 5 of them) is effective even in open loop. If you do all of your fuel tuning without resetting the ECU, that doesn't matter. But if you reset the ECU when you reflash, then there's a good chance that you're logging and tuning with a 0% D trim. The problem is that if the D trim moves to +5% after a couple days' driving (it can take a while) then you're now running about 0.5 richer than what you saw during your tune/flash/log iterations. If the D trim moves to -5%, then you're 0.5 leaner, which is asking for trouble.

For the other trims, it's mostly just about the warm fuzzy feeling that comes from getting the balance right without the ECU having to make up the difference.

Why run open loop then? Is it possible to get a wide band and run closed loop the whole time having the ECU adjust on the fly, so our AFR are always what we want them to be?

Also are the fuel trims viewable in romraider? i dont see a list of fuel trims on the base map or can i only view what trim im using when I log?

This topic1603.html article explains (if you haven't seen it) the CL / OL stuff.
And there's a bit of info here topic5615.html on STFT.
Dale

^ well thats what I mean. The article reads that because our 02 sensors are inaccurate, (even the 32 bit wide bands) that subaru decided to make the open loop fuel trims... But is it possible to stay closed loop the whole time with a good accurate wide band that is place down wind of the turbo?

Fuel trims are not in the ROM. There are some parameters related to them (airflow ranges, min/max, etc) but the actual trim values are generated by the ECU when it's running.

Full-time closed loop is not out of the question (far as I know) but nobody has done it yet (far as I know). It's something I've been wondering about for a while though, so I'll start a thread about that in the Tuning Discussion forum. Short answer: nobody has yet proven that it can be done. Might be do-able though...

Here's the new thread:
topic5731.html

Quick question -- I am now playing with the MAF scaling (logging, updating MAF, logging) .... so far so good. However, it seems (but maybe I will need to test more***) I am getting more knock in some places ... so, should I:

1. Pull timing in trouble areas or;
2. Scale MAF up to shoot more fuel?

I scaled my MAF at idle (to test) .... updated MAF, and it seemed it idled a bit higher on AFR ... and then I re-scalled MAF, and it changed again in the opposite direction ... tricky

Also, like many, if I have it in 5-6 and floor it, I always over boost ... will MAF help this, or should I play with Turbo fuel cut stuff more? My tune was good (12.29 1/4 mile on E85, stock turbo) .. but, it's too rich. So, trying to play with it a bit ... maybe I will just advance timing a bit more

Guy, any issues with E85? I ran it last summer and running it again -- car is FAST, but seems my tune is strange (or the car just doesn't play well with it) ... any issues that you ran into it going to it? Just curious ...

So scaling the MAF means you are changing the load value the ECU uses as it looks up just about every other parameter it needs to operate your car. So changing the load puts you into different cells in many tables but maybe not by the same percentage in each table as many tables are different sizes and cells cover varying ranges.
I think if you are getting knock then adjusting the timing table is appropriate, or a combination of AFR and timing.

I believe if you are over boosting you need to adjust down the WGDC...? I haven't done any boost tuning so I'm not 100% sure here.

For idle, and any other time the ECU is using 'closed loop' fueling (basically this means less than 50 g/s) the ECU will adjust fuel delivery to get the AFR it wants. The fuel trims displayed in Learning View show you how much fuel the ECU has learned to adjust fuel delivery. If you're getting different idle speeds, that's probably not related to MAF scaling. I'm more inclined to think it's related to how warm the engine is, or perhaps whether or not the O2 sensor is up to operating temperature (takes a minute or so).

In closed loop, the target AFR is based on the tables in the "Fueling - Closed Loop" section. In open loop, the target AFR is based on the "Fueling - Primary Open Loop" tables. In general, you should adjust MAF scaling so that the ECU sprays the amount of fuel that will produce the ECU's target AFR. If the target AFR is not what you want, adjust those tables. If the ECU's target AFR is set to what you want, but the ECU isn't able to achieve the target AFR, then you adjust the MAF scaling.

Does this make sense?

hi, have a question. lets say i have a unknown cc/min injector. but i key in the values as 500, what will happen next? how do i know if i have guessed the correct scaling.