Ok, well... i spent a good 20 hours working on the MX6 in the last couple days, and she's all buttoned up except for one problems.
I'm still throwing two CELs. One is code 15 - bad o2 sensor.
In the past, i threw this code all the time because i didn't even have the stock o2 sensor. On Saturday, i wired in the Analog Output 1 to the ECU o2 input. Factory o2 sensor was a 1 wire unit, so no weirdness with heater or ground or any of that.
I did double check in the software when i connected to the LC-1 that Analog Out 1 is in fact set up for a 0.1-1.1v signal, and the "high impedance" setting is enabled.
Still no love.
Has anyone successfully used narrowband simulation on this unit? In searching the interwebs, it seems like i'm not the first to have issues with this, and Innovate seems to have taken their forums down as well.
A theory i have is that the car runs rich enough that it doesn't spend much time in the 15.0:1 area so the voltage being sent to the ECU wouldn't change when cruising. (Which is when the light will trigger) Cruising AFRs are in the 13.8-14.2:1 range. Does this sound like it might be an issue?
The cruising AFRs do sound a bit off to me if we're talking about steady state cruising with no or not much boost. I (and probably the ECU) most likely will want to see the AFR hovering around stoich or probably even slightly lean at that point.
BoxheadTim wrote:
The cruising AFRs do sound a bit off to me if we're talking about steady state cruising with no or not much boost. I (and probably the ECU) most likely will want to see the AFR hovering around stoich or probably even slightly lean at that point.
This is actually leaner than the stock ECU was, fwiw... 
I really can't lean it out any more with the simple setup that's on the car at the moment.
But it does sound like it's possible that it's simply throwing the CEL because the ECU hasn't seen 15.0:1 or higher AFRs within a certain amount of time?
I can modify when the Analog Out 1 sends each voltage in terms of AFRs, but i'm a little worried about trying that because it would have an effect on how the ECU behaves. The only thing i'm trying to do at the moment is keep the CEL from happening so the car doesn't pull timing and start dumping a bit more fuel.
Ah, OK. I was more thinking along the lines of the CEL coming on because the ECU doesn't see the O2 sensor's voltage switch at all...
BoxheadTim wrote:
Ah, OK. I was more thinking along the lines of the CEL coming on because the ECU doesn't see the O2 sensor's voltage switch at all...
That's what i'm thinking...
It won't come on if i'm always driving through traffic or constantly decelerating/accelerating.
But once the car is warmed up, and i cruise for more than 5 minutes, CEL gets thrown.
Have you cheacked to see that the lc1 is putting out a signal? It sounds like when it goes from open loop to closed loop and looks for the O2 sensor it is getting screwed up some how.
Haven't checked that yet. Guess i'll throw the multimeter on it in the next couple days.
The bad news is that if it's not putting out a signal, that's a bad controller and i'll just be throwing this thing in the garbage.
92CelicaHalfTrac wrote:
Haven't checked that yet. Guess i'll throw the multimeter on it in the next couple days.
The bad news is that if it's not putting out a signal, that's a bad controller and i'll just be throwing this thing in the garbage.
Instead of the garbage - throw it in a box to me if it comes to it.
m4ff3w wrote:
92CelicaHalfTrac wrote:
Haven't checked that yet. Guess i'll throw the multimeter on it in the next couple days.
The bad news is that if it's not putting out a signal, that's a bad controller and i'll just be throwing this thing in the garbage.
Instead of the garbage - throw it in a box to me if it comes to it.
I'll keep you posted. I don't understand why something like this has to be such a pain in the ass.
Verify with your Helm's manual or equivalent that the .1v-1.1v range is indeed what the factory ECU is looking for. It may be looking for a different range, causing it to throw a code because the default is out of spec.
I appreciate everyone's help so far. 
However, in checking the Mazda FSM (i was working off a Ford FSM at the time i put the car back together), the Ford FSM has a different wire color listed for the wire i need to splice into.
I spliced into a blue with black stripe wire per the Ford FSM. Unfortunately, i doubt my cooling fan relay really cares what sort of signal it's getting from the LC-1.
I don't mean to laugh at your pain, but: hahahah! I hope your fix takes care of it.
The entire experience of working on this car from the period of Thursday to Saturday afternoon was pretty much a train wreck, so laughing at my pain is really all i can do as well.
Laugh away!!!!
I need to make a thread for this thing finally... but here's midway during the tear down. We had it down to just a block/head, and trans sitting in the engine bay, everything else out.
Turned it into this:
And i have clear headlights again so you can all stop complaining about it.
But yeah, this stupid Code 15 is the last thing i need to get figured out. It's also throwing code 42, but that won't light up, and it doesn't do anything. (Wastegate solenoid malfunction)
I will say that it's HUGE amounts faster than it was before. If i had to hazard a guess i'm looking at mid 13s @ 7psi. 
The only real way to see what's going on is with an oscilloscope (can you borrow one?)
The narrowband signal should switch between approx 0 and 1 Volt, at a rate between several times per second to every couple of seconds. As long as the signal is crossing approx 0.5 Volts each time, the ECU should be happy.
If the signal is not switching, or crossing 0.5V, then the ECU may see that as a dead O2 sensor, and throw the code.
OTOH, many ECU will not reset codes immediately,even when the problem has been corrected.
Have you been resetting the codes in the ECU between tests?
Carter
I liked the yellow headlights.
erohslc wrote:
The only real way to see what's going on is with an oscilloscope (can you borrow one?)
The narrowband signal should switch between approx 0 and 1 Volt, at a rate between several times per second to every couple of seconds. As long as the signal is crossing approx 0.5 Volts each time, the ECU should be happy.
If the signal is not switching, or crossing 0.5V, then the ECU may see that as a dead O2 sensor, and throw the code.
OTOH, many ECU will not reset codes immediately,even when the problem has been corrected.
Have you been resetting the codes in the ECU between tests?
Carter
The car has only ever thrown codes 15 and 42 and they both come back every time i reset the ECU. This past time, i've actually got an entirely new ECU on it, same deal.
It's wired into the wrong wire unfortunately, so i'm not going to bother with further testing until that's fixed.
turbojunker wrote:
I liked the yellow headlights.
I did, too. I think for now i'm going to grab another stock set of corners and see how they look all clear, or i'll paint those amber.
Then after the Challenge i'll spend the money on some LaminX and re-do the yellow properly.
Man look at those sexy injectors.
Osterkraut wrote:
Man look at those sexy injectors.
They are pretty sexy... but not the ones you sent me. Those are Supra 440s in the car right now. The 460s might go in soon just to see if they work with this particular ECU. Definitely will be used with the FCON though.
