j-rat Thats what I was thinking.

I'm only going to be boosting around 10 psi so even with the fcd I should be ok to run it after the fcd.

You could set 8 psi (since thats all the pressure sensor will see with fcd) to be 100% throttle by changing the throttle hi/low points, correct.

Anyone tapped the pressure sensor wire at the ecu?

I tapped into the 5v wire on the pressure sensor before the FCD, but if you're going to have it switch to the hi-thrtle map at 8psi then sure you could tap into that line anywhere.

Thanks.
I was actually going to have it switch over to high throttle map during any boost.
I have 4x720cc so I have to carefully map the lo as well as high or it'll run crazy rich.

I'm thinking about having anything < 0psi on the low map and everything > 0psi high map.
Basically off boost fuel map and on boost fuel map.

That way while cruising on freeway or around town I will have one fuel map and when I boost at all I'll have another.
Right now when I'm on the freeway the dumb s4 tps jumps from high to low and my fuel jumps around accordingly. i'll be off boost cruising and be in the high throttle map dumping fuel for no reason other than the tps sucks.

With the pressure < 0 I can have a leaner af ratio then when I boost I will have a 11:1 ratio.

How is it working?

i don't know how many of you people got thsi working. BUt if you tapped the safc to the pressure sensor BEFORE THE FCD it won't work.

Because the safc throttle % self calibrate or self learns. It can not learn on the pressure sensor voltage. Unless u drive at full boost then full vac for an hour maybe.

What happens is that at full boost (stock 8 psi or 5.5 for S4 cars) the safc thinks 8 psi which is like 3.7 V from pressure sensor is like 100% throttle.

Then it realizes that around 1.7 V is like Zero throttle % (which is voltage at like -300 mmHG vac). So it then calibtrates and centres the safc throttle % accordingly

For my stock S5 TII i get full boost at 8 psi. So after driving for a week the safc calibrated itself to 100% throttle at like 4 psi and 0 throttle at like -200 mmHG.

OK what to do now. You need the safc to learn on a wider voltage signal. I have tried using a hand pump to pressureize the pressure sensor sitting on the driving with the key on the "ON" position. Ie engine not running. THe safc will not learn this way. It requires RPM and Airflow sig to begin calibration. Without both of these it won't calibrate itself after reseting it.

Next move it so remove the safc TPS sig wire back to the TPS wire for your car. Then have it learn on that for a week .The TPS ranges from 0 V to around 4.98 V or something like that (i have an AVCR too).

After one week. Remove the wire from the TPS and place back to pressure sensor before FCD. Now you are somehwere close to the range of the pressure sensor. (see above posts for equation)

So now 5V corresponds to 14 psi or close to it. I hope you are not going to run 14 psi right away. You can do simple math to figure out where you should be. The stock ECU is only good for around 9 psi of fuel dilevery. So say 14 psi = 100% therefore 9 psi = x. The relationship is linear. So that means 64% = 9 psig.

So set your low throttle point to be 65% (this will let the mazda ecu control boost up to 9psig), then set hi throttle pt to 100%.

This means the safc will correct fuel for boost pressure in the range of 9 psi to 14 psi. DO NOT EXCEED 14 psi, i'm not sure what will happen since the mazda pressure sensor is maxed at 14 psi i think.

Anyhow the safc will then correct between 9 psi to 14 psi in RPM increments. GET A WIDEBAND! DO BE DUMB AND BLOW YOUR CAR UP!

good luck.

Sorry about hte horrible english and the long post... i didn't bother to proof read it before posting.

nite

Did anyone read my idea?
I can't do what you said (Cheers!) I have 4x720cc injectors.
I have to map the entire fuel curve from idle to full boost.

I have been using the s-afc on the tps for about a month now so I'm going to switch it to the ps now.

I made up a visual of my idea. Here it is.

Low Throttle Map

< 0 psi Boost

RPM

1000 2000 3000 4000 5000 6000 7000



High Throttle Map

> 0 psi Boost

RPM

1000 2000 3000 4000 5000 6000 7000

What do you think?
One map for boost and for no boost with the boost map fuel increasing with the rpms (along with boost increase over rpm)

How can this be possible? The SAFC is supposedly set for a 1v signal being closed throttle, and 5v being open throttle. It doesnt have any "learn" function that I am aware of.

Thats what I thought mine has run the same since I hooked it up, no learning.

Anyway any thoughts on my ideas?

Exactly. The S-AFC has no learning ability. 1V for 0%, 5V for 100%.

If you have a vac-pump around, you can attach it to the pressure sensor vac line with the ignition on. Then pump until you see positive pressure while watching the AFC throttle percentage. Write down at each PSI the throttle percentage that is being registered on the AFC.

When you have it graphed out, you can set the crossover point on the AFC to switch at any PSI. :D

Edomund: Your idea is good, but you may have to use 1-2 PSI instead of 0 as the cross over point.

edomund,
I have my safc set up this way too. I reset the airflow sensor type a couple of times to try and get the safc to relearn. Then you start the car and let it run under vacuum for a while. Then go out and do a few full throttle runs ( with the safc set plenty rich just in case) to set the high point and do a few big lifts to get max vacuum for the low point. Then turn the car off and watch the display to find 0psi. This is where I set it to switch maps, approx 60% low, 99%high. Another thing I did just to be safe was use the high map settings in the low map above 500 rpm. You won't be cruising that high anyway(I hope) and this gives you a fail safe if you lose the pressure signal under full boost.

Almost forgot, my safc is tied in before the fcd, actually on the fcd wires. This is any easy place to get it, and you don't have to cut another factory wire.

edomund-

I was reading through the manual for the safcII and it sounds like when the sensor (tps or pressure) is between the low throttle point and the high throttle point, the correction is set linearly between the two, depending on where in between you are.

I think it'd be ideal to set the low throttle point at 0psi (or slightly above 0psi) and the high throttle point at 9 or 10 psi, so the map gets richer gradually, and its not excessivly rich at low boost.

Its an interesting idea, but seems to be alot of guess work right now.

Thanks for all the feedback!
I think we are on to something.

I'm going to hook it up today and set the crossover points.
I think your idea of setting the lo at around 0 to 1 psi and the high at around 9 psi is good.
The idea to set the lo map the same as high map above 5000 rpms in case of no boost signal is good as well.

I'll dyno next tuesday to let you all know how well this is going to work.

Anyone else have it hook up like this have anything to add?

This is true, but very soon, it will be backed up by factual dyno graphs. Courtesy of me!

Jarrett

OT, has anyone petitioned the admin for an SAFC section?

Not guess work anymore!
I got it to work.
I hooked it up via pressure sensor (brown wire with red stripe).

I tested it and it works much better than the tps. No more cruising with 100% throttle.

Here's the numbers:

20 inches of vaccum = 0% Throttle

0 PSI (ambient pressure) = 40% Throttle

14 PSI = 100% Throttle

I set my:

lo at 40%

Hi at 71%

I put the hi at 70% because I am only going to run around 10 PSI, since 14psi = 100.
10/14 = .71

*Now when I run no boost (vaccuum) the s-afc uses a percentage of the lo map.
*Once I hit 0 PSI (40%) the s-afc starts using a percentage of the high fuel map.
*Once I hit 10PSI (71% throttle) the s-afc is using the exact hi fuel numbers I entered.

Man this is so much better!

using a map sensor instead of tps for the safc is nothing new, the dsm and honda guys have been doing it for a while.

Cheers, I think you're right. When I tapped in the safc I would see 100 percent on the TPS around 3-4 pounds maybe a little over 3 volts if I used the sensor check. All I did then was add in a series of resistors inline between the pressure sensor and the safc. That dropped the voltage the safc saw until 100 percent was reached around 8 pounds, and that's where I have the cross over set right now. I'll have it on the dyno with some 720's in the secondaries pretty soon. If it turns out that 8 psi is too much for the primaries alone, I'll lower the cross over to 6-7 pounds; that way if I hit 6-7 pounds before the secondaries come online I'll have the primaries spitting out as much fuel as I need. I'm just going for 12 pounds, I might have to get some 680 or 720's for the primaries, we'll see what the dyno says. My previous dyno at 10-11 pounds looked ok before the secondaries came online.


IMO pressure sensor is > TPS for the safc. Of course any stand alone is going to be much better than the safc hack, but for a cheap price it does ok.

just wondering, what is the overall result of this mod? more percise a/f adjustment with the safc? you still retain the airflow meter correct?

Better drivability with larger injectors on the stock ecu.

Yes AFM is definitely still on.

I dont know why your getting 100% at 4 psi boost.
I'm only seeing 60% at 4 psi.
And it's not hard to setup like cheers said.

I think everyone with a S4 TII should hook it up this way. There is no downside to using the pressure sensor as asoped to the tps.

Especially if you are running 720 primaries.

Gen2
The result of the mod is no more 100% throttle when cruising. Hence you actually use the lo throttle map when at lo throttle and the high throttle map when at high throttle (boost).

Normally the S4 TPS sees anything over 30% of your gas pedal at 100% on the s-afc. So it will be using the hi throttle map all the time unless at idle or very very little throttle.

I couldn't figure it out either, but like I said I just added resistors inline which in turn dropped the voltage linearly until I was seeing 100 percent at 8 pounds which is where I want the cross over to be.

I've been running this set up for months.

I definitely agree that the pressure sensor is better than the TPS