Super AFC won't work?!
 

I had this kinda floating in another thread and no one seemed to notice it so I made a new thread...here goes:

Super AFC won't work because of closed loop mode? I thought that's the only place a Super AFC would work. At wide open throttle all cars function in closed loop, right?! Not just Mazda. Its a predetermined fuel map for power as opposed to efficiency/pollution. The AFC is a very simple device. All it does is intercept the signal from the MAF and fool the ECU into believing there is either more, or less air moving into the engine than there actually is, resulting in more or less fuel accordingly. Where as in open loop mode, the ECU is looking at all the sensors and adjusting accordingly to keep emissions low and keep fuel economy up, so that if you adjust with an AFC, the ECU would just fix your corrections to meet its goals? Am I just mixed up here or what? Could someone explain to me why it won't work?

Are you talking about a "throttle Position sensor"?

Something like this http://www.partsforyourcar.com/Merch...egory_Code=300

No not at all. On another thread I read that the Apexi Super AFC won't work on our Mazda's because they run in closed loop mode at WOT. Its my understanding that most/all cars run in closed loop mode at WOT, so I don't understand why it won't work on our Mazda's.

I think it's the other way around. At part throttle, most OBD2 equipped engines run in closed loop mode, meaning it's locked in an input/output cycle between the sensors and the ECU. At a certain load condition, the ECU switches to open loop mode and works off of predetermined maps in ECU memory. The difficultly with piggyback computers tuning an engine in closed loop mode is, depending on how flexible the ECU is, it can compensate for the changes made by the piggyback computer affectively cancelling them out.

All right guys, this whole AFC won't work thing has really caught a bug in my brain, it just doesn't make sense that it won't work, so I did a lot of research today and I found a few things out. Open and closed loop are commonly mixed up. Closed loop is when all the sensors are taken into account and Open loop is when the ECU uses fixed look up tables and fuel maps for a given amount of air, normally at WOT. From what I could find out, and using Travelers post as back up since he obviously asked the techs about this The Mazda definatley runs in open loop mode at WOT just like any other car. Which means that the only sensor with any input to the ECU is the MAF, which is the one an AFC modifies. I even spent some time on Miata.Net doing some research and there are a lot of guys over there using AFC's. Now you guys have been doing this mazda thing a whole lot longer than I have so what is the reason that an Apexi Super AFC won't work with a mazda ECU, cause I really can't find one?

Here is some interesting info from some of the sights I did my research on I highly suggest reading this whole first page that I linked, I didn't put any highlights cause the whole page is worth reading.

http://ackthud.com/shawnfogg/mixture.htm

http://www.siemensauto.com/glossarie..._glossary.html

Open-Loop Fuel Control: A mode where engine input air/fuel ratio is controlled by measuring the mass of input air and adding the proper mass of fuel to obtain a desired ratio.

Closed-Loop Fuel Control: In an engine, a mode where input air/fuel ratio is controlled by using an exhaust oxygen sensor as the input reference.

http://home.clear.net.nz/pages/phil....0Operation.htm

The exhaust gas sensor basically reacts with the oxygen in the exhaust gases and generates a voltage that is sensed by the ECU. The voltage changes dramatically with any movement from stoichiometric and so the ECU alters both ignition timing and fuel to maintain the correct relationship. With the Lambda sensor in circuit the ECU will operate in a closed loop, however, if the sensor is defective or if the throttle is wide open the ECU operates open loop and enriches the mixture slightly to avoid lean out and possible engine damage. It is for this reason that certainly most if not all Toyotas function quite well without the exhaust sensor connected. Running leaded fuel coats the surfaces of the sensor and stops the voltage being generated. For this reason an unconnected sensor is better than a faulty one in many systems.

http://www.interstice.com/~kevinh/soobfaq.html

Fortunately the computer is sufficiently smart to detect a connected or disconnected O2 sensor, but not a lead fouled sensor. According to Subaru, the O2 sensor is not present to meet the pollution requirements, it is there to meet the fuel efficiency requirements. Apparently if the sensor is disconnected, the ECU reverts to an open loop mode which is slightly richer than the closed loop condition but by law it has to meet the pollution criteria in this condition. Subaru told me that I might even detect a power increase due to the richer mixture but I shouldn't bank on it. This all sounds too good to be true so what is on the down side:

http://www.asog.net/wheatking/Message.html

The computer also works in 2 modes, open loop, and closed loop. Open loop is idle and any point up to approx 75-80% of the maximum value of the TPS (throttle position sensor). In this mode it uses the o2 sensor to help calculate fuel maps and requirements. In closed loop, the computer works off the internal ignition and fuel maps. it ignores the data from the upstream o2 sensor.

http://www.howstuffworks.com/fuel-injection1.htm

In order to meet stricter emissions requirements, catalytic converters were introduced. Very careful control of the air-to-fuel ratio was required for the catalytic converter to be effective. Oxygen sensors monitor the amount of oxygen in the exhaust, and the engine control unit (ECU) uses this information to adjust the air-to-fuel ratio in real-time. This is called closed loop control -- it was not feasible to achieve this control with carburetors. There was a brief period of electrically controlled carburetors before fuel injection systems took over, but these electrical carbs were even more complicated than the purely mechanical ones.

http://www.epa.gov/etv/12/12_vs_miratech.pdf

... will operate in open-loop mode until the pre-programmed target air/fuel ratio is surpassed, at which point the
Controller will go into closed-loop mode of operation. Once in closed-loop mode, the Controller uses input
signals for engine speed and air pressure (the MAG-pickup and MAT sensors) to look up the target valve
positions from the pre-programmed valve table, and set the valve
http://ackthud.com/shawnfogg/fuelupgrade.htm

... those products that did have enough flexibility to adjust fueling based on TPS and RPM adjusted fueling by altering the HFM signal to the DME. The DME bases fueling on the amount of air flowing into the engine. If you change that signal you in effect change the fueling.

Anyway guys, there is some food for thought....

Damn. You like to type a lot evidently. Per Apexi the S-AFC changes the MAP/MAF sensors signals. If the Protege goes into an open loop at WOT and is still utilizing the MAP/MAF sensors signal then I don't see why this shouldn't work. But that is assuming that the Mazda ECU works like you think it does. I asked a friend about this too. He thought it should work also. Maybe somebody has tried one? I don't really have a use for one but they do look cool. Oh. And I have one of those "electric carbs". A 1983 Ford F150 "Variable Venturi". It automatically adjusts the mix (and does a crappy job of it) and has an automatic choke (which sucks too). I took the SOB out and put in a 4-barrel with a manual choke.

Typing?! That's old school stuff...

Copy and paste!! ;)

This is essentially the same thing I said in my last post, and it's true. However, a couple of things are wrong.

During open loop operation, the throttle position sensor (TPS) and the MAF inputs are monitored by the ECU.

The last paragragh below is a good example of mixing up open and closed loop operation. ;)

The super AFC and other piggyback computers will work with our cars and others. The major concern is there tuning effectiveness during closed loop operation. If all you are concerned about is open loop tuning, it will work fine.

Also, the transition from closed loop to open loop is different from manufacturer to manufacturer, so typical piggyback computers may be more effective on some makes than others if they go into open loop sooner.

Right Eric, I didn't include the TPS because its my understanding that the only reason it has an input is to differentiate between when open loop and closed loop are necessary. During open loop the only input for A/F ratio is the MAF, the TPS is just there to say "yeah, still use open loop".

Damnit, and yes, there is a mix up right there in my research. ARGHHH!!

Anyway, I have yet to find any research to validate that a S-AFC won't work with our ECU.

i've 'heard' the s-afc is not useful at WOT. i think corksport 'may' have actually tried to tune a 3rd gen with a s-afc with little success. maybe you could call them and ask about their results(503-252-5700).

for a $100 or so dollars more, you can get the UNICHIP piggyback, which will cover fuel and timing. just a thought.

The Unichip has the same issues as the S-AFC. They should both work on the Protege during open loop mode operation. How effective they are during closed loop is a big question mark. The last question is when does the Protege go into open loop operation. If it's at 75% of max throttle opening, the piggybacks are probably useful enough to be worth the price. If it's at 90% of wide open throttle, than probably not.

I think the AFC is pretty much useless in closed loop. If you read the first page I linked to, there is a link on there for fuel management. He does a really good job of explaining why a piggy back won't adjust in closed loop. Basically the car wants to be at 14.7:1. No matter what adjustements you make, the ecu will make appropriate fuel changes to level out at 14.7:1. Not only that but it will place those changes into memory for short term fuel trim and long term fuel trim. Short term effects only closed loop mode while long term effects both closed and open. Anyway, if you make adjustments to remove %5 fuel in closed loop the ECU will make adjustments to get back to 14.7. Over time this value will eventually make its way to long term fuel trim making it so that you are running 5% lean across the board. If you have 5% for open loop then you really end up running 10% lean.EEK! Luckily I knew about the fact that the AFC won't effect closed loop, so I neve messed with mine. If you guys didn't read the page on that first link above, I highly recommend it.

S-afc
 

I have three questions:

1) do you or anyone else have to AFC hooked up?

2) do you think that it is worth 300 bucks??

3) where did you get the wiring diagram from since the AFC site doesn't list it? Did you get it out of a shop manual? if so then when can i get the wiring diagram from??

Re: S-afc
 

I don't think you are going to find any 3rd Gen 2.0L owners trying this out (for reasons given above ^). Anyway, they are pretty cool but apparently (as this thread seemed to figure out) kind of useless on the Protege. However, you can get a lot of useful information off of one and I believe you can disable the tuning functionality. I know my friend has used his on his Eclipse and he told me that he didn't really have it doing anything. If you can leave it in a read state where it doesn't change signals its a pretty cool addition to your car that you can show off to your friends. You decide whether its worth 300 bucks for no power but a cool gauge.

does the miata do the same open/closed loop operations as our 3rd gen (2.0L, specifically)? Is so then the link engine management systems would work well with our cars. They have been used on many turbo miatas, and flyin' miata uses them with their turbo kits. I think they may be a little pricey compared to the AFC. They are about $800. Here is what I'm talking about: http://flyinmiata.com/store/prodinfo...tem=4&mitem=19

I think you guys are missing a main point here. All modern OBDII equipped cars (as far as I know) are just like the Mazda ECU. Ours are no different than anyone elses OBDII cars that are already using an AFC. It will net you jack squat in closed loop mode. Who cares? That's not where you want to make your adjustments anyway. When I had my IS300 I talked to several people who had intake and exhaust and saw gains, then they put on a S-AFC and saw 10-15 WHP!! at WOT. Now I am not saying that its worth the money, or that you would see those kind of gains on a 2L 4 banger, but the fact remains that you can use it, and probably could use it to score some extra power. As soon as I can scrape together the funds for a CAI and exhaust, I will put a SAFC on to see what I can get, just cause I already have one.

I just started this whole thread cause someone said it wouldn't work with the "Mazda" ECU which just isn't true. Now the validity of it being a worth while mod is a whole seperate argument.

so you're saying the link engine management system will work just fine as well?

Yeah, I should think so. I can't imagine why the Links wouldn't work. Again, it would only affect open loop mode.

Did you guys know that a lot of ecu's do not stay stay in closed loop past 3000-4000 rpm because a regular o2 sensor does not react fast enough to supply proper info for the ecu at higher rpms. Unless you're ecu relies on, maybe, a wide band o2 than it might stay in closed loop longer.

Not sure if this is relevent to this thread... but I thought I'd throw it in anyways since nobody's mentioned it.

That would be nice to know for sure. Does anybody know if the Miata's ECU does open/close loop like the 3rd gen protege (2.0L specifically). I am still looking into the LINK for a turbo system. I basically want what the flyin miata kits have and they include the LINK system so I would like a similar kit....

Well another question is:

Has anyone tried this on the MP5 yet? What was your results i understand that it only works in open loop and open loop is at WOT. So if you are racing then you are at WOT the whole time right? So you would want to have the AFC working then right?i mean what do you care what it is doing during part throttle.

If i am looking at this in the wrong way please correct me?
The way i am reading this trend is that the AFC only works in open loop and open loop POSSIBLY happens at about 75-90 % throttle right? If not then then WOT throttle. So if the AFC is used in a racing sense and you are at WOT doing that, then you will be in open loop where the AFC works at all the time and you can control it through the RPM band right? At all other times i could care less about open loop and closed loop since i might just be cruising. It is just when i am trying to merge or when i am trying to go down the track or what ever it may be that will make me be at about 75-90 throttle or WOT which is open loop.

Help me out fellas!!

There is a guy with a turbo MP3 who uses Link. Here is the link:

http://forum.protegemp3.com/vbb225/s...&threadid=1845

He got:

5psi=161 whp
8psi=188whp
and 230 WHP with the boost up, and timing retarded (he told me on the phone)
Stock whp is about 115-118 if I am not mistaken

charles

In my opinion you are absolutely correct!! Hence the whole point of this thread.

After having my car on the dyno I can tell you that the stock fuel curve is painfully rich. The timing isn't near aggressive enough either, but is damn near tailor made for boosted applications. Since we want to control the fuel first to see what the car will do, I'm going to try something a little different. Instead of running an electronic piggyback and fooling the stock computer, I'm going to try to get an adjustable pressure regulator and take some of the fuel pressure out. The stock ECU will compensate in closed loop, but the WOT settings will be leaner since the comp doesn't change those. I'm trying to source an adjustable regulator and will try it on the dyno to see what happens.

Also, stock HP on the 2.0L is approx 96 for the ES, and approx 98-99 for the MP3. Yeah, I know it's not 10 extra HP but that info comes from several different people that have dynoed the cars. The big difference is that the timing is a lot more aggressive on the MP3 and it has a lot better throttle response.

Accually a stock MP3 was dynoed a little closer to 114 WHP.

When Tripoint did theirs they only got a bit over 98. I've talked to a couple other places that have tested them and they've said the highest they'd seen on a dynojet was 103.

Talked to a rep at Apexi today. They said the only product that would be benificial to us is their Super-AFC. They do not have a wiring diagram for it however and I was told that we'd have to get a shop manual and identify which wires to connect to ourselves.

Also spoke with a rep at Jet and they have an ECU upgrade package for our cars now. The item code is #65002. It's $399 from them, but I called around and found that part # from Landspeed for $279. Basically you pay for it over the phone or online, they send you a box that's ready to be shipped to Jet along with instructions for removing your ECU and packaging it. You're responsible for shipping it to JET. The turnaround is usually 24 hours at JET from the time they receive it. They send it back 2 day service and the cost of return shipping is covered in the cost of the Upgrade package. They are claiming 9-11 WHP with this ECU upgrade. They will ask you what mods you have done via the questionaire that's included in the shipping package. Lot's cheaper than a Unichip and you can install it yourself without having to tune it on a dyno. Drawback is, you can't tune it.

Also, I've been unable to locate a fuel pressure regulator that will work on our cars. All the universal ones use a return line. There is supposed to be one out for the new Honda system without a return line, but that's the only one (figures).

AEM is still saying that there may be a Plug and Play stand alone for our cars coming. They won't give out any more information than that. Got the impression they were being tight lipped and are doing something that they don't want to talk about or broadcast.

I was told the Link system wouldn't work on our cars due to it being attached internally to the Miata computer. I'm curious to know how it was done on that MP3.

There is a wiring diagram on the protegemp3 website. I will have to hunt it down...here it is:

ECU Diagram

I also have the 2001 Wiring Diagram for all Proteges. I bought it from the dealer for $45.00 USD. I will look into it more later.

Tripoint didn't use a dynojet, that is why their numbers are different...They used a Clayton Chassis Dyno.

Grassroots Motorsports dyno'ed a stock MP3 and got 113 at the wheels. They tested the MP3 and a number of other "letter cars" to see what ones offer the best back for the buck in stock trim.

Well i will go ahead and try it and see what i get since it looks like no one has tried this mod. Once i find one and have the money for it then i will install it and let you guys know what i find out.

If it doesn't work then i will sell it like i did when i bought it for my 1990 probe gt and found out that it didn't work due to the voltage in the VAF was to high. The AFC works between 3~5V and the VAF on the 1990 probe gt idles at just under 5V so when you hit the gas the voltage will go up and the car will die!!