Ben

iTrader: (33)

Curious if anyone has successfully been able to simulate nb to their ecu with a wideband? I can't get my lc-1 to work with the stock 1.6 ecu; main grounds by the brake cyl, heater ground to cam cover, sim nb tapped to ecu pin 2n (red/blue). I called Innovative and spoke with a guy named Phillipe who was no help.

brgracer

iTrader: (14)

My LC-1 NB sim has been working fine feeding the stock ECU. I'd check the output settings first and make sure that it is on spontaneous and not 1/12 or 1/6 sec. If that is okay then, I'd change ground locations. I have my signal ground connected to the ECU ground and the heater/unit grounded elsewhere.

Ben

iTrader: (33)

brg,

Thanks for replying... Do you remember which ecu ground you used? From what I understand, the ecu references o2 ground at the body between the engine and cowl on the driver's side--that's why I used that location. Did you indeed tap 2N at the ecu?

Thanks

jayc72

iTrader: (14)

You have the WB in the stock greddy downpipe bung?

Can you describe what the car is doing? I moved mine out of that location because I beleive it was getting too hot and not giving me accurate enough NB signal to the ECU. Car would be find when cool, but if I got it good and hot I would get bogging and it appeared that I was running rich.

Braineack

iTrader: (62)

i can send to my gauge without troube. I grounded the analog ground and gauge off the ecu 2N port.

Ben

iTrader: (33)

Jay,

With the lc-1 a1 out programmed to default and connected to 2N at the ecu I have the following problems:
1 car fails to idle and will stall
2 logworks shows a/f of ~8:1 at 1.5-2k rpm (minimum engine speed required to run)
The problems begin immediately after the lc-1 shows o2 sensor is warm and reporting. The car will idle correctly before the o2 sensor comes to operating temp. With a1 disconnected from the ecu (no o2 sensor connected) I can idle and drive, of course it is very rich and accerates poorly.

I have tried altering voltage values in lm programmer. Eventually I was able to make the car stay running at idle, but it was at 10:1 afr and running poorly ("hunting"). With no o2 sensor I can idle at 13.9:1

Thanks

Ben

iTrader: (33)

Brain,

Are you sure you grounded at 2N? I thought 2N was o2 V+ ecu input!? Wiring diagram shows grounds are 2A & 2B http://www.madracki.com/miata/images/wiring/91diag.pdf

I can get a1 to send signal to my old nbo2 and dvm without problem. BTW, I am now grounded to the motor itself (cam cover bolt)

jayc72

iTrader: (14)

I wonder if you have a bad wideband sensor? I'd do another free air calibration. I had my LC-1 randomly recalibrate while I was idling the car. Did it when I attached it to my laptop.

If that doesn't work I'd run both your NB and the WB, using only the WB to monitor things. At least in the short term. Because at this point you don't know if the AFR the WB is reading is out too lunch because the car is running like poo. Or is the car running like poo because the WB is borked and sending a high voltage to the ECU.

Jay

jayc72

iTrader: (14)

Becareful which ground you grab off the ECU harness. They don't all go to the same place. One of the grounds goes to the igniter and the other to the block under the MC. I had all sorts of trouble with my emanage when I had the O2 sensor and the EMB grounding to different locations. I talk about it in my trouble shooting thread.

Jay

brgracer

iTrader: (14)

Ben,

I have mine grounded at 2B (black wire).

What readings are you getting on the LC-1 when you are not feeding the ecu and the car is running rich? My first LC-1 had a digital convertor failure and had a fixed output at around the afr you mentioned and it did exactly what you described. It would start up and slowly die b/c the ecu would see a rich reading and continue to pull fuel b/c it was fixed at a rich setting until the car would stall. I could see the fixed reading on the gauge and contacted Innovate. After a bunch of diagnostic tests, they said it was a bad convertor and replaced it.

Would also repeat the free air calibration as mentioned.

Tom

Ben

iTrader: (33)

Thanks. Right now I have all 3 of the lc-1 grounds and the xd ground going to the engine with a 16ga. Earlier I had them grounded to the factory ground point on the body under and to the inside of the brake booster (except the heater ground went to the engine). The ecu grounds on the same panel, but on the outside of the brake booster. I never tapped directly to the ecu ground because I did not know if the correct ground was A or B at the ecu, but I did know that the ecu grounds under the clutch cyl.

I did a full calibration at installation and a second full calibration yesterday. It is absolutely running rich, regardless a possible incorrect calibration. There are two large, black streaks on my driveway under where the pipes were, and it stinks and pops when you let off throttle. It will smoke if I hook up the a1 to the ecu and maintain throttle to keep the engine running.

With no o2 sensor, the afrs I get are:
warm idle: 13.5-13.7
cruise, 20": low 13's
cruise, 10": high 13's
cruise, 0": mid 14's
boost, 1psi: mid 12's
boost, 2+ psi, 10's going to 8's @ 6psi


With a1 hooked to 2N, default settings,
idle: not possible
2k rpm: mid 8's, low 9's
Obviously I didn't drive anywhere like this, so no other data exists


With a1 hooked to 2N, but voltages at .6 and 1.6,
idle: mid 10's, car "hunting" and rpm swinging from 600-1100
2k rpm: mid 10's


Setup:
91 1.6, stock injectors, vortech 12:1 (will get my begi fmu installed once the lc-1 is going correctly--but not before), walbro 190lph hp

I called innovative for support. Phillipe told me that I shouldn't simulate nb to a stock ecu. Thanks, I'm glad that they advertise it.

Braineack

iTrader: (62)

oh i dunno remember anymore, whichever one mine bipes and o2 clamp is grounded too as well.

y8s

iTrader: (8)

I just welded an extra bung in and ran the NB too. Seemed easier at the time. I have a TechEdge though which will output the NBsim signal but I never bothered.

brgracer

iTrader: (14)

Not trying to get you riled up, but I think Phillipe is giving you the run around b/c that is one of their selling points and in fact, I think one article cites how running the NB sim got someone better gas milage. Furthermore, the developer klattin and other people on the forum have been talking about/doing this for awhile now.

I get close to 29mpg with the LC-1 feeding the stock ecu with no issues. Granted, I think that it is just easier at this point to just put a NB to the ECU and get a second bung, but that is total bs that it should't be done or not work.

Just for reference, here are a few threads from their forum where one of their main developers is talking about using the NB sim to feed the ECU:

http://www.innovatemotorsports.com/f...&highlight=ecu
http://www.innovatemotorsports.com/f...&highlight=ecu

bripab007

I ran my Innovate LM-1 like that (simulating NB to stock ECU) for several months. Worked great!

Braineack

iTrader: (62)

yeah I would assume a heated o2 sensor would be better, and if it can simulate a NB signal, why not. I was just too lazy to have to find the signal wire in the ecu and wire in my output2 to do it. one of these days I will.

jayc72

iTrader: (14)

I used the harness for my original O2 sensor. Cut off the O2 sensor and wired it into my Lc-1 analog out. Made no connections at the CPU at all.

I'd get in touch with someone else at Innovate. That is total bullshit that you shouldn't run it to simulate narrowband. The factory default configuration on Analog 1 is setup to do just that. And I've seen mention a number of times on Innovate's forums talking about using the NB simulation to tweak fuel on the car. Talk to someone else.

Jay

Ben

iTrader: (33)

Update:

After screwing around grounding the crap out of everything, I welded the included bung into my midpipe. Now the LC-1 and ecu seem to be getting along. The car idles smooth and steady at an indicated 850 rpm (with the nb sensor it would bounce between 825-850) and it will accelerates again. And the CEL has not come back on.

Problem: Poor idle/run after simulating narrow band to factory ecu
Solution:
1> Reference common ground (ecu pin 2B)
2> Check LC-1 settings, correct as necessary--Do NOT trust factory defaults
3> Install new sensor bung further from turbo

Note: After installation of the wb controller, I acheived an average of nearly31 mpg on a daytrip to the north GA mountains, including running through the twisties and using plenty of turbo. This was mostly back roads; we were on the interstate less than 45 miles total.

y8s

iTrader: (8)

the reason most places suggest not to use the NBsim is response time. from what i understand, the wideband sensors are much slower to respond than the standard narrowband version.

i dont personally know how the response times compare but it may have an impact on how the stock ecu adjusts fueling.