My analog output from my LM-1 shows 360 to 550 millivolts when simulating the narrow band sensor. This isn't enough variation to keep the PCM happy. After a few run cycles a P1153 and/or P0154 DTC occurs. The simulated voltage needs to go between 100 and 900 mv I think.

I believe LM Programmer can amplify that NB millivolt range but I'm not sure what numbers to use in the Volt and Lambda boxes. Do I need to move the Lambda range, the voltage range or both?

Can anyone enlighten me?

 

Tighten up the spread so the effect is greater. Try using lambda's of 1.1v=0.98 lambda and 0.1V=1.02 lambda. Also, make sure the response speed isn't too fast. You may need to set it to 1/6 or 1/3.

 

I tried all of the suggestions but the voltage spread remains the same as detected by the PCM. I am wondering if I have the wiring correct. All I am using is the red wire from the LM-1 #1 analog out. This is supposed to replace the NB O2 sensor high signal. I have tested the analog out by making both ends the same voltage. A volt meter verifies the programming is working and a specific voltage is being generated. I wondered about the ground offset but the difference between the copper ground wire off analog #1 is negligible to the cigarette lighter (LM-1 power source / ground and PCM ground).

Is there anything else that I might try??

 

There are 4 wires. This diagram is from a '99 Vette, but should still work:

On the car side, make sure your LM1/LC1 signal wire is connected to the "high" signal wire (labeled B) and that the "low" signal wire (labeled A) is connected to a good ground. Stock NBO2's ground the low wire by connecting it to the exhaust as seen in the diagram.

On the LC1/LM1 side, make sure all wiring is connected as instructed.

 

I hope you don't mind the flattery but I've always paid attention to your posts. In this example you have supplied the needed help to my very frustrating and long standing problem. I've had this Innovate controller for 3 years and haven't been able to figure out how to make the controller take the WB data and scale it out to a NB signal. Correct me if it's actually a simulation. My understanding is that it's an actual scaled representation.

I had been trying to understand page 17 of the LM/C-1 manual in regards to wiring the #1 analog into the O2 sensor loom. To anyone that stumbles on this thread; you're a genius if you can go by those few sentences in that assuming manual and get the correct result.

I don't know if I will have a DTC after a few run cycles but at least with the altered wiring the ECU is seeing the full range of millivolts from analog #1 output. At least that's what my ScanMaster is reporting on the #2 oxygen sensor. At times it acts a little strange because I need to move the throttle ever so slightly to get the sensor moving and it seems to be affecting the other sensor while in closed loop. But the really weird thing is, the LM-1 controller is now displaying believable A/F numbers while idling.

Anyway, I had the low signal connected to the #1 (the red wire is the tip of 3.5 mm stereo jack) THINKING it was the high signal. I can justify the selection of the TAN wire when testing for the high signal wire. And I would not think any of us electronically challenged people would ever think a low signal wire is actually a ground that carries half a volt.

Because the LM-1 controller is powered and grounded through the cigarette lighter circuit - which I believe is the same ground circuit for the PCM (and all the O2 sensor "low signal" wires). Maybe this isn't the best ground after all but it certainly is convenient and according to the manual, no ground is specified and once you get into the Innovate forum it's ALL about ground offsets.

Anyway, thank you kindly!!! Now I can get back with my friend who has been waiting for me to get this problem resolved so we can finish our EFILive tuning @ WOT.

 

Thanks for the kind words.

LMK if you can get it to work for sure. I'm still a little skeptical...

 

Skeptical about the grounding??

I am soldering bullet and blade connectors on the ends of wires now. The car comes down and I do some driving. As life goes, usually things get fixed at some obscure point in time but I hope not this time.

 

Take a look at the thirdgen.org site for and do a search for SAUJP. It has some OBDI bin files that will run the NB section along with WB logging. Sounds like what you are seeing is probably correct and your ECM is logging correct data but since it is so narrow it is giving the errors. Just a suggestion why don't you just turn the codes off and not worry about it while doing WBO2 logging? I am going to be there shortly but I still have a ways to go to get my car on the road and will be doing some WBO2 logging. I am not sure if I am just going to put a new bung in and log separately along with the regular O2's.

 

It's all bolted together now and I've attached a ground wire to the exhaust in case I need one.

It works the same as it did yesterday in that I'm getting the full scope of millivolts. However, I notice the scanner spends more time showing the 0 - 700 mv than it does any higher. I can make the top of the range appear by very slightly touching the throttle. It will go to 999 and hang there for a moment. Also, while in open loop the WB sensor is feeding 100 mv and below to the PCM until the engine goes into closed loop. I think I can fix that by de-selecting high impedance - the real NB sensor stays right around 450. Can you tell me how to adjust the mv range so i see more of the lean side from the NB display? Does that make sense?

 

I'd be lost using OBD1 bin files - wouldn't know what to do with them. I can't imagine my PCM would know either but I'm not getting a lot of things.

If I turn the codes off I think I'll screw up my LTFTs but I don'f that for sure.

Certainly a dedicated WB sensor would be best but in my case I'm not running headers and don't plan to. We have emission testing here so i have to work around that.

 

Please help me under stand better. I think I get what you're saying...kinda.

Ideally, you want to keep .55V = 14.7:1. So however you move the spread, that kinda needs to stay the same. The numbers above may have been too aggressive.

When is the last time you free air calibrated the sensor?

 

Maybe I did my math wrong above...try these:

1.1v = .98 lambda & 0.1v = 1.016364 lambda

1.1v = .97 lambda & 0.1v = 1.024545 lambda

I've got a good feeling about this one:
1.1v = .96 lambda & 0.1v = 1.032727 lambda

1.1v = .95 lambda & 0.1v = 1.040909 lambda

 

The sensor was calibrated a day or two ago.

I'm bringing the car back down, I think / hope all the work is done in that regard. I'm going to drive it around a bit and see what those numbers do in real traffic and make adjustments after that.

I wish I understood how you are coming up with those numbers. All I need is to understand the concept of how you are sliding the range, expanding or contracting it. ??? BTW, thanks again!

 

1.1v = max rich (lambda)
0.1v = max lean (lambda)
.55v = 1 lambda

For the example, we'll call the lambda value at 1.1v "L" in the equation below. If you go by .05v increments, there are 11 increments between 1.1v and .55v, which we know .55v=1. So, the equation for each .05v increment becomes:

(1-L)/11

If L=0.97, then the increments are .002727. There are 20 increments between 1.1v and 0.1v. So, multiply it out and add in the starting value:

0.97+(20*.002727)=1.024545


It's a lot easier than it seems...especially if you have Excel.

 

I wanted to follow this up so all interested will know the longer term outcome.

I didn't need to reprogram the controller analog output for the NB voltage but I may fiddle with them in the future, so thanks again for explaining how to do it. The Innovate default factory numbers are working, no codes after 2 weeks of testing.

I would be curious now to see how the car would do at one of our friendly e-check stations with the WB installed in place of the NB sensor.

 

Thanks for the update. Do you have a log, please...?

 

Lots of logs stored here joecar. I'm trying to think of what you would want to see. I would be happy to post an EFI or LogWorks3 log, a graph or an exported log in text format. What would you want to see?

 

Do you have some EFILive logs that contains HO2S11 and HO2S21 (i.e. both front NBO2's, where one of them is the LC-1) in addition to at least LONGFT1, LONGFT1, VSS, RPM, TP, MAP (captured at 10 frames/s)...

by looking at some waveforms it [sometimes] is possible to tell if the NBO2 signal behaviour (amplitude, frequency, rate of change, etc...) is inside or outside the limits required by OBD-II (lol... I did say sometimes... )

What did you do to make the DTC's to go away...?

Do you have long tube headers...?

 

I looked back through the logs. I have not monitored the O2 sensor voltage. All I was concerned with was the LTFTs so my logs do have LONGFT1, LONGFT2, VSS, RPM,TP and MAP. I will create a PID list and log just what you describe; just give me a little time because I'm caught up at the moment in trying to figure out why I have some huge spikes at open throttle but no spark knock. I will gladly create and post an EFI log with the two NB O2 sensor voltages though. I would be curious to watch the voltage from EFILive Scan myself. I just never thought to do it before. I simply watched the voltages from my in-dash ScanMaster.

As for getting rid of the DTCs, they (P1153 and P0154) went away when I finally got the right wires connected between the LM-1 controller analog output and the PCM. I used the Innovate factory voltage / lambda profile for the feed into the NB O2 circuit.

I do not have LT headers. I have the factory LS1 exhaust manifolds and catalytic converters which is why I needed to supplant the NB sensor in the factory supplied O2 bung. There is virtually no way I could put another bung in front of the cats and still get a wrench in there.

 

No worries, take you're time...

Yes the stock exhaust is rather cramped.