My $7.50 AFC

[douglas vincent]

First off, Thanks vrg3!


Some wire, some bullet connectors and a 50K potentiometer.

As some of might know, and for those who don't, I installed 370 cc injectors in place of my 270 cc injectors because i was running horribly lean under boost. However, this now gives me the opposite result of running incredibly rich so that my milage sucks and when under boost, the engine bogs horribly as soon as I get about 5250 rpm and higher.

So I had to trim fuel or trick the computer to trim fuel. And I also had to figure out what signal my MAF sensor was sending to the ECU.

So I hooked up the MultiMeter to the MAF signal wire. Idle started at .8ish and at WOT I would get max 3.4 volts. Under boost I would max out at about 4.8 volts. Important is that anything over 4.4 volts or so would induce the dreaded bog.


Oh yeah.

This is done on a 1990 2.2 NA Legacy. I cut the #4 signal wire near the MAF and ran two wires into the car. I marked the line connected to the MAF as the "IN" line. I took the 50k potentiometer and soldered two wires to it, one of the wires is also soldered to the middle prong, ie wire 1 is soldered to both prong 1 and prong 2. wire 2 is wired to just prong 3. Then the "IN" line is connected to wire 1. The "OUT" line is connected to wire 2.

I connnected a multimeter to the #3 wire (ground) and #4 wire AFTER the potentiometer. This gives me the true reading going to the ECU. Then I can monitor the reading.
So I installed a 50k potentiometer from Radio Shack. This allowed me to trim the signal about 10% or so. At a 3500 rpm cruise level, I can turn the knob back and forth and "feel" the break point from too lean to just right. I can also do this under boost by having it read too "rich" at anything over 5000 rpm and crank the knob toward lean and instantly have the bog dissappear.

I have heard this might advance timing some, but since I run premium all the time, I am not afraid of knock.

And as it is, big difference. Not alot more power overall, but now I can get that extra power from 5000 to 6000 rpm which, as we know, is very important to the ego.

[vrg3]

Just to make it clear, I don't recommend that anybody do this.

The way I'd say to do it, if you're really stuck on using just one component, would be to connect one end of the pot to the signal line, and the other to ground through another resistor. Then connect the ECU's input to the wiper of the pot. Better would be very simple voltage follower before and after (which can be done with a single chip and no external components).

That pot cost $7.50?! Stupid Radio Shack.

I don't think running premium should make you feel unafraid of detonation given that you're running a Roots-type supercharger.

[douglas vincent]

I know you didnt reccomend this, but without your help, I wouldnt have gotten this far. Thanks


And it was $2.89 for the pot, the rest was for a wire pack and bullet connectors.

[vrg3]

You still might consider the latter setup I suggest there. It's likely to give you more predictable control for only a dollar or so more.

Oh, okay, that price is a little closer to reasonable.

[-K-]

So you are not going with the MAF bypass? Do you have the scan tool set up for your car? I'd really like to know how much timing it advances...

I'm all about just doing something, the same people on here that tell me my ecu is smart enough to advance timing because I'm running an intercooler say the ecu isn't smart enough to retard itself if it is advanced by messing with the MAF signal.

[douglas vincent]

I have no idea how much the timing advances.

However, I can tune it by ear and feel somewhat. Turned all the way left, it gets the entire signal. Turned all the way right cuts the signal by about 9% or so.

If you rev the engine and hold it steady at 2500 rpm, or even better can do this in third while cruising (so you are under load), you can turn the knob down from "rich" and perceptively feel the spot at which the ECU decides there is less air than needed and it cuts out some fuel. Same way in reverse from the "lean" position, the car will suddenly "accellerate" a tad.

So I keep it at right above the sweet spot so I know the engine is getting enough air, but not enough so the MAF signal doesnt go above 4.3 on boost.

[sammydafish]

so the output of the MAF is a linier voltage signal? not a waveform?

[Legacy777]

so the output of the MAF is a linier voltage signal? not a waveform?

The MAF's function is not linear, it's quadratic. Vikash would be the one with the most experience as to the function and such.

[sammydafish]

so the output of the MAF is a linier voltage signal? not a waveform?

The MAF's function is not linear, it's quadratic. Vikash would be the one with the most experience as to the function and such.

how can a voltage signal be quadratic? if it was, the same voltage could mean two different things.

it is common for MAFs to be logarithmic or tangential, but not quadratic

[vrg3]

It's quadratic. It's common for MAFs to be quadratic. Search for "quadratic" with me as the author for more info.