When I was new(er) here and I read Merlin's tuning guide; it suggested that 15.5-16 was max fuel econ afr. So I basically changed all the 14.7 cells to 16. Logging showed that my AFRs were still at ~14.7 . I then discovered that Close loop uses the NB O2 as a closed circuit feed-back (hence the reason they call it closed loop) but not the fuel map. You can test it yourself. Put the most extreme AFRs values in and it will still be 14.7 once the engine does not come out of closed loop. If Closed loop used the fuel map in anyway, the more you deviate values from 14.7, the more the WB O2 output should reflect that.

On a side note, I think the 13.3 afr in the fuel map at 500 rpm is just to get the car started when cold and in open loop. I once changed it to 14 and the car idled a little ruff till she warmed up.

 

We're saying that the ECU does NOT reference fuel maps in closed loop. Go put 17's or 18's in your idle cells, the car will still idle at 14.7 as long as it's in closed loop.

 

Good Call! Like they always say, "One test is worth a thousand expert opinions"

The map is set to 13ish, so try putting in 1. If it is anything like any ECU or standalone I've worked on, it shouldn't run, unless something else intervenes.

Attached in an image are the options AEM gives you to control O2 feedback. This is from my DSM's tune. In this case, the feedback "recommended" value can only deviate 10.16% from the open loop IDC. I had my wideband go bad and start reporting lean values regardless of the actual AFR, but the car kept running. However, the O2 feedback correction was pegged at the maximum enrichment (10.16%)

 

yeh the stock ECU has a similar thing, the STFT/LTFT will only drift upto a maximum of 12.5% (or 25% if your using that as your factor)...

thats just to stop mechanical problems from overriding what the ECU has been preprogrammed for (bad NB, exhaust leak, post MAF leak, etc)

regardless of whether the ECU does or doesn't use the fuel map in closed loop its still INVALID to put anything other that 14.7 in that cell..

if your having to put a value other than 14.7 (not that I think it does anything anyways) in there then you havn't scaled your injector/maf/map system correctly...

 

You aren't listening. It does NOT reference the fuel map in closed loop. It uses scaling as the base. This is easy to see for yourself if you like, change the scaling up and down and you can see the trims compensate both ways to compensate for this change.

Then you can also try putting the idle load cells in the fuel map to max lean and see if your wideband fueling changes in closed loop, or if your fuel trims change. Now force the car into open loop and change the idle load cells in the fuel map and see what happens to your wideband value. This might just blow your mind.

it's +25.4 and -25.6 for the evo on the stock ECU. If the 02 fails you can see the car go into a "limp mode" and max out the trims one way or the other. If you want to see this for yourself, unplug your front 02 sensor while the car is running.

See above.

This information is coming from people that has disassembled *this* ECU code. I understand you are smarty pants with the stand alones, this is not a stand alone and shouldn't be compared to one really. Graduate degree in engineering tells the world one thing only, that you completed the courses.

 

In regards to #1, referencing the fuel map doesn't solve #1, the trim max/min feature previously discussed solves #1

In regards to #2, as I previously said the fuel map is just an 'adjustment' of its own.. If you look at the raw values, 14.7 = 0x80. 0x80 is a very commonly used number in the ECU. Its smack bang in the middle of 0 and 0xFF, which is can fit into one byte.

So when at 0x80 then don't do anything. If your either side of 0x80 then do something.. The further you are from 0x80 then the "more" you do...

 

Tephra... any idea what will happen if we lower the .51v specifically set for the ecu to determine stoich in order to lean the equation for stoich out?

 

So, again, I must be missing something. The trim allows the final IDC to vary +25.4 and -25.6 - What is this varying from? I see this as the stoich AFR must achieved at no more than +25.4% or no less than -25.6 the "reference value" - Is this incorrect? If the value in the load cell at idle is 12 and you change it to 1200, it will not idle (at least with my understanding).

What do you mean by "scaling"? I have always understood scaling as a way to convert the raw duty cycles to more intuitive numbers....

The entire point I am making (or at least trying to make) is that only closed loop system needs a starting value!


Not to sure where you're going with this...
(FYI, most engineering graduate degrees require a thesis - I proved to the world *two* things, that I can complete courses, and research! That's twice as good as you had implied! )

 

the injector scaling is the starting value..

in a round about way...

 

But isn't that pretty much the same as the base fuel map? If I were to multiply the entire fuel map (assuming its units are duty cycle) by 1.10, isn't that the same as adding 10% IDC to the injector scaling or is injector scaling something else?

 

like I said, I don't believe the values in the fuel map are used for closed loop...

but if someone wants to go dump 0 into their fuel map and see if the trims change then by all means.. please do...

and anyways like I said, its an invalid/incorrect configuration todo it that way... you are better off dialing in your injectors properly...

 

I was going to reply to all this but realized I would just be repeating myself a couple more times. The answer to every question you have in this post has been answered and explained in previous posts. Go back and soak it in.

 

I am with Bryan and Tephra on this one. You have to force open loop in order to create an "economy tune". It is easy enough to do. I am just a dumb Army guy, and I can economy tune with the best of them. Been doing it for many years.

The values in the fuel tables only work in open loop. You can change the values, but unless you have crossed the open loop threshold, you won't use the falues in the fuel table at all because the ECU will use the predetermined set values with the assistance of the O2 sensors to remain as close to stoich as possible....i.e. closed loop.

+1 on not comparing apples to oranges with ECUs. The stock ECU is nothing like a stand alone.

Paul

 

I think I see what you're saying, 0x80 corresponds to 14.7. Can I assume that 0x00 corresponds to 0 and 0xFF corresponds to 29.4 (i.e. 14.7 x 2), or is it offset by the scaling (where 0x00 could = 10 and 0xFF could = 19.4)?

Either way, we agree that a starting value is required. I am guessing that the values originate in the fuel map in closed loop. However if you are not able to enter a value greater than 125% your original value, it would be impossible to test (and as a result, the origin of the reference value is irrelevant). Maybe we need to move to higher precision numbers so that we can span a greater range, and settle this debate!

But why? You seem to enjoy it so much. (Certainly don't stop on my behalf) Here is what I think we figured out (not exactly by reading or rereading any of your post)
I think we all agree that there is a starting value that can be modified by +/- 25%. Tephra (who is very knowledgable on this) speculates that the value does not come from the fuel map (the origin of the value is still unknown). Based on EFI theory and how an AEM EMS operates, I respectfully disagree, but as I mentioned above, it is impossible to know given the memory configurations of this ECU and therefore, meaningless. (Not mentioned in your post...)

This is not known for sure, as mentioned above.

Getting back to the original recent question posted by silverstream that I was attempting to answer - No, you cannot simply lean out idle by changing your fuel map. You should use a wideband and follow the posted article on how to do so properly. The only other option is to not use closed loop at all (not recommended).

Widebands are pretty cheap now, so I recommend adding it to the upgrade list. Then you can see what your AFR's are really doing.

 

well I am not posting anymore until someone:
1) tells us what their trims are
2) changes the closed loop idle section of their fuel map to 7.4
3) flashes and waits for trims to stabalise
4) tells us the fuel trims

if #1 and #4 are different (by a decent amount) then the fuel map is used in some degree...