I am reluctant to remove the intake manifold to check the injector connection, but I feel like its inevitable.

Attachment 152293

Which cylinders are wet? Is it wired for sequential fuel?

Red line in composite log is lost sync events. So you get a couple while its grabbing sync. And then it stops because it syncs.

No sequencial. I was trying to avoid fucking with too many things until it actually ran.
After more testing, here is what I have found.


Cyl.1- no spark- wet
Cyl. 2 weak spark-dry, brown
Cyl.3 lighting type spark- brown
Cyl.4 no spark, wet.

I don't understand. Is this good, bad or normal?

Interesting. 1 and 4 are the same trigger from the ECU. 2 and 3 are different. Are you using output test mode to test the coils?

Nope. I could figure out how to make it test. I set it up like it says, but the start test button never becomes available. Just grey. Don't even really understand what should happen. Directions a vague at best. Do I crank it, do I sit there and wait for a test to start?
Instead I just removed all of the plugs, grounded them and cranked.
Anything from efi analytics seems to just have microspheres of information. I emailed them about directions on how to use the mega log viewer, and they said that they never wrote any. Well that's nice.

Make a little video of how to do it.


Might be a bit for it to process

Thanks man, I really appreciate the help. I really do. Ill try it later tonight. I have to take a break from this thing for an hour or two, Im really frustrated and need to get my head straight.

Always a good idea. No use getting pissed off at it.

Just wanted to throw out there, that if anyone is interested in my coil mount, I can provide a dxf file. It only cost me 40 bucks to have it cut.

Yes please

Update..
Testing the coil trigger wires.
With the ignition on-
Brown =240 Mv.- this is the circuit that actually fires a coil
BROWN/YELLOW= 8 Mv.- this circuit will not fire a coil.

Ignition off=
NeAR zero for both.

What the fuck does this mean?

I'll get it monday. It's on my work computer. Don't tell.

Holy fuck. It started! Bad connection at the crank/cam plug. Shit.

Well, it starts bit it doesn't run well at all. I don't know what to do, and am at the end of my rope. My wide band that worked fine six months ago, no longer reads correct. Calibration does nothing. I didnt change anything aboit it. There are so many new features on ms3 that I just don't know how to set up or use. I thought all of my ms2 knowledge would transfer over, but it just doesn't.

I am now going to literally start all over. New project, new map, and start from step #1 in the ms3 guide.

Yeah, I tried this. My "start" button is non existing. It's there, but greed out. FOR EV VER. FOR EV VER. No combination of actions make it accessible.

That's really odd. Same settings as I posted?

Yep. Same ones. I am a paid TS user, soooo. I emailed them to see what they say.

Got TS test modes working. All looks good too. Now on to,

FUELING
If my required fuel is set for 1000cc it comes out to like 3.1 or something. That being said, will my old fuel map work (from 265s) or do I need a completely new one. I thought that the idea of the requed fuel was to compensate injector output for their size, to an existing map.

That's the theory, but hardly ever reality. req fuel only linerally shifts the entire fuel table, and injectors aren't always a linear upgrade.

It's usually close enough to autotune back into place.

10 4. That's kind of what I have gathered.

I just don't get this! Everything tests good. Everything. Fuel map, i have started at 0 and increased every cell by 1 up to 100. I cannot get an idle.

Go higher, reqfuel or VE values.

What were you using when it started the other day, wideband still dead?

I can get it started, but no idle. I'm back to my original map, with original results. WB is still dead. Doesn't show an error code, calibrates fine. No reading. Nothing changed from when it worked fine.

ego corr is off right? if wb dead and ms see's pegged lean or pegged rich it will do funky tings

Yeah, TS pegged one way, Guage is the other. I'll get this sorted out first. Ergo is off.

O2 is working now. I don't even want to say what it was. Doesn't really matter.
Done for tonight. Must eat.

Sounds like progress! :likecat:

You had the wrong wire hooked up to the MS?

I bought the ms second hand, and didn't check the harness myself. PO had installed the O2 sensor wire in an auxiliary, instead of the factory location. I used the factory sensor wire (red w/blue stripe on my 95) So my WB was going to know where.

That would do it.

Three more days of fiddling around with this WB, and I still cant get a good read in TS. The actual gauge reads reasonable, but the TS gauge sits on 11 then swings around like its on meth. I remember a thread saying that a Brain built ms3 used a different pin for the WB, but now I cant find it.

Is it an LC? Sounds like the output is programmed as narrowband.

mtx l. been through the set up 100 times. project setting is set to wideband, then in TS I am using the custom setup with .22 v= 7.23 (or something close to that) and 5v = 22.9( or something close to that as well.

But the mtx l needs to be programmed too I think. It can be programmed for narrowband output

Not sure I understand. I has a wire used for analog output (yellow) and they say to use the settings like I did in TS. Again, this same gauge, wired exactly as it is now, worked fine with my ms2

The mtxl can be programmed to emulate a narrowband output. On my lc-1 it was the default setting and I even had it reset back to that on me.

Test the voltage at the input pin of the ECE. Engine off it should read the full lean, and somewhere in the middle when running.

Your sensor could be dead too.

Ill check that! ^^^^ could a shitty sensor read good on the gauge and bad in TS?

EDIT: mtx L only has analog wideband output. thats how I read the directions anyway. they dont ever even mention a narrow band anything.

Do you have a general engine settings page, it should have lag factors for map, rpm, lambada and more.

With my digital (reverant/mtx_l) wideband interface I get pure gibberish unless I use 10 for a lag factor instead of the default 50. The gauge will read true, the MS not so much...

It's a long shot, but what the hell.

What he said too. I still think narrowband setting.

Again, the TS gauge just sits still on 11 regardless of engine speed, then just randomly spins around like its possessed. Then when I turn the engine off, it goes to 19. I'm stumped.

Ill try this too!

That sounds like the behavior I'm used to, but that don't mean much.

But to clarify, the wire colors on the MTX_L output are useless if the gauge has been purchased second hand.

From the factory it is yellow=WB, brown=sim.NB

There's a data cable though, and software called LM programmer. You can write a narrowband voltage curve to the yellow wire, and the gauge will continue showing true wideband readings. The yellow wire is no longer wideband, it's whatever custom curve the PO programmed.

That's why Aidan is talking voltages and stuff.

Again, the WB is the same one I purchased brand new, that worked flawlessly for two years with my ms2. Every problem I have encountered in the last month, has had to do with this ms3, in one way or another. I am so burnt out on this thing. I feel like I'm one more wasted weekend from selling it all.

I hate that after building this car for a year straight, I am being brought to my knees by an electrical issue. One that I am certain, that a more experienced person could figure out.

Is there anything in the ms itself, that could fail and cause such a problem?

No, shouldn't be. It's a pretty basic voltage comparison circuit I imagine, not a lot to go wrong.

You could open it and follow the o2 terminal on the board, make sure it's going to the same db-whatever pin you've got it wired to on the other side. For it to read 11 and 19 means it's reading in the middle of the range you've programmed, something is happening.

I'd also go back into the wideband config in Tunerstudio and change it to LC1, just to be sure you've got the proper volt/afr curve.

I used the default volt/afr settings that Brain suggested, along with what mtxl says to use. I have also tried every Innovate WB choice in TS, just to see. I've learned to not post here without exhausting every other option.

Have you tried reading the voltage of the WB signal going into your MS?
One thing that's crazy about this problem is that going from 11 to 19 is pretty much what the reading of a narrowband sensor does, but if the brown NB output of the gauge were hooked up, it would put out .1-1.1V, so the AFR would only jump from ~7 to ~12 (with the LC1 calibration).
Could there be a problem with the O2 circuit somewhere? There would be some problems if it were grounded or connected strangely.

New findings! I am getting .5 volts fed into the O2 sensor signal wire, only when connected to the MS. Harness only, clean open line.Connect to MS with key OFF, .5 volts. WTF? how can this be?

EDIT. Now it seams that I have intermittent interference, or my meter cant be trusted. Excellent.

voltage from the O2 sensor seams right. comparing voltage to sensor signal to my mtx gauge reading, it makes sense. Tuner studio gauge is just wacky. The crazy part, is that if I set TS to narrow band, I get a somewhat believable reading. how is this possible??? I have every aspect of the the tune set to read wideband.

For the most accurate reading, you'll want to use the sensor ground in the MS. Probably a bit overkill, but good practice.
It could be hooked up for narrowband, .5 is right in the center of its reading and it'll drop to .1v when it reads lean (like when the engine's off)
I'd check the voltage with the engine running too.
https://www.miataturbo.net/megasquir...bration-73818/
double-check your wiring, this guy had his mtx hooked up to the output2 (narrowband) and it was acting just like yours.

oh, didn't see your post there, are you getting a reading above 2V?

Ill check the voltage while running, but....I have literally checked every connection, continuity, voltage and setting more than ten times.. this is what I have.


Problem- TS gauge does not agree with wideband gauge.

- All settings in TS, including project properties are set to read Wideband.

-TS WB setting is set to custom linear .22v=7.35 5v=22.36 (or something close to that)

- Wideband voltage on signal wire agrees with actual gauge reading.

-I am connected to the YELLOW wire on the WB signal output. this is the wideband analog output as stated in the MTX manual.

-If WB gauge reads 20, TS says 7.9 (or there about) this was with the engine off, after running, that's why the 20 on the gauge.

-If i set TS to Narrow band, it almost matches up the two gauges.

- This is the same gauge and wire connections that I had used problem free on my MS2.

Thoughts?

yes

Well if you're getting 0-5v on the gauge's output, it's pretty safe to say it's a MS problem. Looking at the MS schematics, it looks like it's pretty much just a filter, some over voltage protection and then it goes straight to the CPU. If the resistor value for R10 is too large, it could lower your readings, but putting the wrong part in there seems like a long shot. The spec is 1k Ohms, so you could find and check that. Like I said though, longshot. What about firmware? Is that up to date and compatible with everything?

I'm taking the ms apart now to look and see. Yes all of the firmware is up to date.