F/IC and narrowband manipulation
 

Anyone here with an F/IC - me and a friend of mine are both having no luck whatsoever with the F/IC's o2 table. I have a 96 miata, he has a 95. Both of us are reading 0v in the F/IC software for our o2 signal but according to every wiring diagram, we are tapped into the correct wire.

Personally, I've tried just about every setting imagineable and not getting anywhere. It's so bad I've just decided to drive around with the o2 unhooked. Car drives 100% better but it's not a permanent solution.

I've got my o2 tapped at 3C which appears to be the right wire. Guess tonite I'll idle the car and hook up a voltmeter. Any ideas would be appreciated.

3C looks correct to me :dunno:

maybe try tapping into the rear o2 signal which is the Red/Blue at 3D.

the 95 would be 2N - Red/Blu

I hear ya on that but the ECU won't trim fuel from the rear o2 will it? Guess it's something i could try tho.

i dunno, it was just a thought. I figured the F/IC was just looking for a signal to base it's corrections off.

No, the rear sensor is just there to verify that the catalyst is working. It expects to see a lean signal all the time when in closed loop. If it ever indicates rich, that sensors sole purpose in life is to throw a CEL.

Just to verify- the wire you tapped into was covered in a black sleeve, correct? Might now hurt to get a digital voltmeter and verify voltage there. You won't be able to get much of an accurate reading, but you should see something in the range of .2 - .8 volts.

One option if you can't get it working at the FI/C would be to just use an O2 clamp.

Anybody have an idea as to why AEM is calling for a 10k resistor to be installed in series between the sensor and the ECU, prior to the tap into the FIC? Seems odd.

You mean closed loop, right?

0420 code sucks. :td:

I'm pretty sure it was a shielded wire but i'll re-check it tonite. I checked several wiring diagrams and double and triple checked my connection and it's in the right spot.

This thing is driving me nuts. The miata ECU strives for 14.7:1 harder than any ECU i've ever worked with before.

Uhm, yeah. That's what I said, right? Check my post. (just ignore the place where it says "last edited...") :bigtu:

Any updates?

I'm the one with the 95. Apparently SloS13's o2 input magically started working this morning, but he can tell you more whenever he comes here (not that there's much that can be said about magic).

Yes Braineack that is the pin I chose, 2N, red wire with blue stripe. This is driving me crazy. I'm gonna hook up the bypass harness and see if it continues to throw me the "O2 output voltage too low" CEL code. If it continues to throw me that code then there might be a possibility that my o2 sensor decided to die RIGHT before I hooked it up to the F/IC.. since I wasn't getting the CEL before I hooked it up nor was I getting 20 mpfg!

I originally had a 10K resistor hooked up but I spoke with an AEM tech yesterday who said the reason I'm not getting any input from the o2 sensor could be a result of the resistor being too strong... So I tried a 1k and no resistor at all and I'm still getting nothing... ARGHHH!

on the aem forum they were talking about using different resistors then the 1 k that is shown ... how is your table set up? mine is set up as map load /voltage then i filled in the cell while in boost with 0.3v and it seems to work fine. i noticed sometimes while at low rpm in 6th gear and as it goes in to boost it wants to pull fuel

When I was testing the setup with and without resistors yesterday I had the table on O2 input and Offset mode with a zeroed out map so it wouldn't mess with the incoming signal... however, there is no incoming signal. So today I tried running it with the bypass harness to run the ecu in stock mode and I still got the CEL. I'm starting to think the o2 sensor is busted. AFIK it was in perfect working order before I installed the F/IC. My gas mileage had been down by 3 mpg but after the install and now the CEL it's in the low 20s (down from 28-30).

I bought a multimeter today so I'll be testing the voltage tomorrow morning to make absolute sure there is no voltage coming from the o2 sensor. Then I'll go through and check all the connections again... damn this is driving me nuts.

Mine magically started working. Vac to boost is much smoother now. Narrowband's sweet-spot makes it a bit difficult to "target" a particular AF in closed loop situations but its definitely better than not working at all.

Not sure why it didnt work one day then started working. I went out to my car, plugged in the stock O2, started the car, hooked up a voltmeter and got a bouncing voltage, looked at the screen and saw a bouncing voltage. First I did an offset of like -0.7v and the car started idling really rich, so it was working.

Update:
In the instructions it calls for a 10K resistor right before the F/IC wire splices into the ECU o2 wire so I thought that might be the cause. Tried it with a 1K... threw a CEL. Then I tried it with no resistor... threw a CEL. Each time it's the same code too. The output voltage of the o2 sensor is zero. I hooked up a voltmeter and read 0.00v. So then I decided to replace the o2 sensor, but there was no way I was gonna pay $150 for a stock Miata o2 sensor when I have an LC-1/DB wideband setup waiting to go in as soon as I figure this problem out. So I looked around the intraweb and discovered the early dodge neon has the same type of sensor as the miata and it's only $50. So I bought one, soldered the new sensor to the old harness and put it back in. CEL? yes... 0.00v reading on the sensor? Yes. So I decided to read the resistance to see if the sensor the neon uses is indeed the same. The resistance on the miata o2 sensor is supposed to be 13 ohms... the resistance on the neon sensor? 4 ohms I'm not really sure how the other guy got his CEL to turn off with a sensor with different resistance. This is driving me crazy. Luckily this isn't stalling any major installation yet as I still have a lot of other separate prep/fab work to do before the turbo goes in, but this is definitely wasting a lot of time figuring this out.

Does anyone have a spare o2 sensor they'd loan me? I'll even paypal you a security deposit until I mail it back to you.

EDIT ~ I'm wondering if the local dealership would be able to temp-install an o2 sensor just so I can find out what the hell is going on.

unless I'm wrong, besides the plug, arent all narrowband o2 sensors the same thing electrically?

I assume you were measuring the resistance across the heater, and not across the sensor, yes? The sensor itself has a nearly infinite resistance, and on pre-OBDII cars the resistance of the heater is relatively unimportant. You could even run a 1-wire unheated sensor in your car.

At 13.5 volts, 4 ohms would pass about 3.4 amps, resulting in a dissipation of around 45 watts. That's a lot, however in your '95 the power feed to the O2 heater is not through the ECU but directly from the "Meter" fuse. So long as that guy isn't blown, you should be ok.

Do you get any reading at all on the Red/Blu O2 sensor wire when everything is disconnected? You'll need a good digital meter that you can manually set to low range in order to get a reading, but you should be able to measure the output of the O2 sensor with everything else disconnected. Know of course that when the mixture is very lean, the output of the sensor will be near zero anyway. At normal idle you should see maybe 0.3 to 0.7 volts.

That's good to know about the resistance... I was basing my reaction off the enthusiasts manual as it states if the resistance is not 13 ohms, the sensor is junk.

When I check the voltage I disconnect the sensor harness from behind the valve cover, take the yellow/green plastic harness protector thing off and place the positive probe in the pin of the blue sensor wire coming from the sensor itself. I take the other probe and ground it on any part of the body in the engine bay. It goes to 0.000v on the 2vDC mode... there is no fluctuation, just 0.000v. It doesn't change if I rev the engine, let it idle, hold it at XXXX RPM... it doesn't do anything. I'm starting to think it's the harness itself and somehow the sensor wire came loose from its pin in the harness and so no matter what I do it just won't give me a reading. But that also means it mysteriously came loose as soon as I hooked up my F/IC.

*Bzzzzzzt* Sorry, but thanks for playing.

If you have disconnected the O2 sensor completely from the vehicle's wiring, it may not work because the heater will be off, but more importantly you've got the wrong ground reference. Assuming a standard 4 wire config with two black, one white and one blue, you need to put the (+) probe on the blue wire and the (-) probe on the white wire, not chassis ground. Additionally, you should apply +12 and GND to the two black wires to power the heater (polarity does not matter.)

But the critical thing is the placement of the (-) probe. 4 wire sensors are not referenced to chassis ground- they are isolated.

alright so since i just soldered in the new sensor im just gonna take the heat shrink wrap off and measure the voltage directly coming out of the sensor instead of the harness.

That'll do. I would suggest that if possible the output of the sensor initially be disconnected from the AEM, the ECU, etc., just so you can validate the raw sensor output.

Then start adding things back in- first make the connection to the ECU, with the resistor, and measure the voltage on *both* sides of the resistor. It should be roughly equal.

Then you can put the AEM back in, and ensure that it is not affecting the voltage when it should not be. My understanding of the system is that the AEM is acting as a pulldown on the line when active in "clamp" mode, thus explaining the resistor- the AEM is trying to act as one leg of a voltage divider network.