Timing advance for better fuel mileage using Link V3

[turboglenn]

It's little stuff like this that keeps me in favor of full on stand alones, although i do think they've made a grea troduct with DSMlink i would still go crazy not having the full conrol over some parts of the engine's operation..

Anyway.. my question is still this.. How can it be called a true "open loop" mode in the ECU when the fuel table is based around a targeted AFR and not a set pulsewidth at a given load/RPM cell? I'm not complaining i'm just trying to better and fully understand the DSMlink system. With a target AFR the computer has to be looking to something for the actual AFR to compare to and see how far off target it is. And then, i would guess it some how can correct in order to get the AFR"s on with the target value which at that point would still closer meet the defined term of "closed loop" more so than meeting the definition of an "open loop" operation.

I give major kudos to the whole ECMlink crew for the product they've developed and for cracking the ECU's code and putting it into a nice, effective, graphical visual interface. And like i said i'm not knocking the system at all,. Hell, it's proven it's self to be effective even for big power setups and ET's into the high 8's (i believe) but low 9's for sure which is faster than i've ever gone in any car i've driven. BUt either way these are the little things that explain why i still reffer to the system as a piggy back unit compared to calling it a stand alone system. But i'd take it over an SDSEFI setup any day of the week

 

[Calan]

Anyway.. my question is still this.. How can it be called a true "open loop" mode in the ECU when the fuel table is based around a targeted AFR and not a set pulsewidth at a given load/RPM cell?

Well... in open loop, there is no feedback or error checking; the loop is "open"

The ECU just looks up a desired AFR indexed by RPM and load (in the "OpenLoopMaxOct" table), calculates the PW needed to hit that AFR based on what it knows about airflow and fuel delivery, and then fires the injectors....at which point we hit that big opening in the loop and nothing else happens.

In closed loop, there is no AFR lookup; the ECU is always targeting 14.7:1, and uses the front O2 sensor to see if it hit it. The front O2 feedback drives the fuel trims, which are basically just moving correction factors that get tacked on to the PW that the ECU calculated to hit 14.7:1 at the current RPM and load.

Or did I completely misunderstand the question?

EDIT:

Damn it Glen... now you having me thinking about how silly the phrase "open loop" is; there is no such thing as an open loop. If we must use geometric terms, the two modes of operation should be called "loop" and "straight line".

Something like "phase-locked" and "linear" would make more sense.

 

[turboglenn]

Well... in open loop, there is no feedback or error checking; the loop is "open"

The ECU just looks up a desired AFR indexed by RPM and load (in the "OpenLoopMaxOct" table), calculates the PW needed to hit that AFR based on what it knows about airflow and fuel delivery, and then fires the injectors....at which point we hit that big opening in the loop and nothing else happens.

EDIT:

Damn it Glen... now you having me thinking about how silly the phrase "open loop" is; there is no such thing as an open loop. If we must use geometric terms, the two modes of operation should be called "loop" and "straight line".

Something like "phase-locked" and "linear" would make more sense.

Ok,that completely clears everything up that i was asking about. My main thing was getting where the ultimaley final pulse-width is derived based on the fact that you don't ente a pulsewidth but simply an AFR. This also clarifies to me about the "VE" table and the "AFR table", i've always said that a true Sd setup should only have one fuel table and that being pulse width.

based on your edit, I think you're starting to see what sits in the back of my mind bugging me abot these terms when used on DSMlink... So is there no fuel trimming in the wide open areas, as i hough there was based on what i've read about the OEM ecu and it's ability to compensate for dirty injectors and such over years of time, even while at WOT.

Anoither thing that makes me argue the point about closed or open looping in the wide open areas is that there's two fuel tables and two timing tables and it vaires between them under WOT conditions based on at least knock, but im still guessing that there's some o2 feedback input into it's decision also (again this is based on my findings that the OEM ecu in stock configuration does monitor the front o2 sensor to choose ultimatey what fuel and spark curves it's going to utilize, and even if there's not a "loop" for AFR's there's definitely a closed loop effect to the timing, but it's a beneficial onethat I won't argue with

I think a good bit of what im bringin up has probably got more to do with being the only options based on taking over the OEM ecu and especially with converting it to SD from MAF, either way they've done a great job with the ECU and i can't complain, but there's a lot about it i have questions on as compared to all the stand alones i've tuned in my life (more than i can remember to count) but really i think as long as it reliably works and is consistant with it's fueling under the WOT areas there's no reason to say any one way is better than another, i think all of them have their ups and downs

 

[Calan]

My main thing was getting where the ultimaley final pulse-width is derived based on the fact that you don't ente a pulsewidth but simply an AFR.

But wouldn't you rather just enter a desired AFR?

Anoither thing that makes me argue the point about closed or open looping in the wide open areas is that there's two fuel tables and two timing tables and it vaires between them under WOT conditions based on at least knock

It's a very slow process though, and you can simply bypass it altogether in ECMLink and gain a set of auxiliary maps. Even if you keep this functionality, it will reset on every start-up, which means the learned octane function will never have enough time to do anything.

From ECMTuning:

"In practice, though, the ECU almost always uses the max-octane maps because nobody with ECMLink allows the engine to knock very long without correcting the problem.  So the min-octane maps really end up unused."

As set up from the factory, the ECU will just slowly start interpolating between the "normal" maps and the "low octane" maps when it sees extended periods of knock over multiple sessions. I guess it figures you must have gotten hold of a bad tank of gas based on the repeating occurrences of knock, and will start trying to protect your motor. (I think that is how it is supposed to work anyway).

Pretty cool idea, but not really a closed loop type of operation in the true sense.

 

[skizzo81]

ok so i am running 14.5 afr at cruise and i get 27mpg.