Turbonator cal: WB2NB use issues

[wheming]

Trying to utilize the Innovate MTX-L wideband as the sole O2 sensor on MeanMini. On the recent rebuild the 0-1V output was used to give the O2 signal to the smec. I was trying out the 0-5V output and trying to configure the WB2NB feature in the turbonator cal.
The issue I've found is during a warm restart of the car. The wideband output is basically nil when the car first starts due to the sensor heat-up cycle. It signals the smec with basically a lean condition which appears to make the car try to correct.

Is there something else that needs to be adjusted when using this feature? I think the work around might be to alter the open loop mode to be activated even on a warm restart.

Any thoughts?
Is anyone else using just their wideband to read to a turbonator smec? The other thread was discussing using the 0-1V output.

[Force Fed Mopar]

You may have to give the sensor a 10-15 seconds to warm up.

[ShelGame]

Trying to utilize the Innovate MTX-L wideband as the sole O2 sensor on MeanMini. On the recent rebuild the 0-1V output was used to give the O2 signal to the smec. I was trying out the 0-5V output and trying to configure the WB2NB feature in the turbonator cal.
The issue I've found is during a warm restart of the car. The wideband output is basically nil when the car first starts due to the sensor heat-up cycle. It signals the smec with basically a lean condition which appears to make the car try to correct.

Is there something else that needs to be adjusted when using this feature? I think the work around might be to alter the open loop mode to be activated even on a warm restart.

Any thoughts?
Is anyone else using just their wideband to read to a turbonator smec? The other thread was discussing using the 0-1V output.

Yes, you might need to change the warm re-start so that it stays in open-loop longer. I get the same thing on my car. I get a CEL for a few seconds on a warm re-start due to the warming sensor...

[wheming]

Yes, you might need to change the warm re-start so that it stays in open-loop longer. I get the same thing on my car. I get a CEL for a few seconds on a warm re-start due to the warming sensor...

Is it more than one table that needs adjusted?
I'm looking at the O2 Controller section but I'm not sure which. Is it the "LowLimitForO2RampSwitching"? Or is it " DelayTimeToClosedLoop"
I don't want to mess with the wrong ones.
Thank you.

[ShelGame]

I'll have to check later tonight. But, I think you don't want to mess with the limits. DelayTimetoClosedLoop, maybe...

[wheming]

If i wanted it to stay in open loop, to test not getting a O2 feedback, can i just lower the 03LTHTIM "IfNoO2SwitchForThisTimeThrowFault51or52"? It is currently set to 43.
I'm trying to test running without the O2 to test a possible bad wideband sensor/gauge/output/O2 sensor wiring.

Thanks for the assistance.

[x.Gen]

I run an LC-1, so I'm not as familiar, but I checked the manual for the MTX and it looks the same as the LC-1...

you can use the (innovate) LM programmer for the MTX to set warm-up output to whatever (volts) you want, but only for the analog output. it is in advanced settings, but the factory default is 0. I know this doesn't help you get where you want, but how it should help - I'm going to guess the WB output does the same thing (puts out 0v until the sensor is warm). so changing anything in the cal probably won't help if you are using the WB output of your sensor. only thing you can do for that is let the sensor warm-up cycle complete before running the engine, seems like it usually takes 5-10 seconds.

[Vigo]

so changing anything in the cal probably won't help if you are using the WB output of your sensor.

If you keep it in open loop it will IGNORE the sensor for a certain amount of time, which is what is needed here.

[wheming]

That's good to know x.Gen. thank you.
I had not looked to see if i should configure the wideband differently. I was trying to diagnose an issue where the car would run a bit then stumble and die. Warm/hot restarts were a problem too. I wanted the car to stay in open loop to test if was a bad controller, sensor or output.
As it turns out, might have solved the issue this afternoon by replacing the ignition coil. It was a very difficult thing to diagnose!
I ran the car today with the 0-1v output and ran fine. Seems a bit lean for a time after throttle blipping, but it eventually settles out.

[wheming]

I do want to make sure the car stays in open loop long enough for the sensor heatup to occur and read. On every start the sensor goes through a heatup.
So that is where i want to be sure all of those tables which relate to this are adjusted properly.

[x.Gen]

If you keep it in open loop it will IGNORE the sensor for a certain amount of time, which is what is needed here.

I got that, just trying to add a few missing pieces to the puzzle. I'll leave the right answer to Rob, but I'm guessing he might have to add some code to accommodate a start-only condition like that.

[x.Gen]

That's good to know x.Gen. thank you.
I had not looked to see if i should configure the wideband differently. I was trying to diagnose an issue where the car would run a bit then stumble and die. Warm/hot restarts were a problem too. I wanted the car to stay in open loop to test if was a bad controller, sensor or output.
As it turns out, might have solved the issue this afternoon by replacing the ignition coil. It was a very difficult thing to diagnose!
I ran the car today with the 0-2v output and ran fine. Seems a bit lean for a time after throttle blipping, but it eventually settles out.

you're welcome. It looks like the MTX doesn't have a switch/LED setup to display controller errors (like the LC-1), but displays them on the gauge? you should also know the LM programmer will allow you to (separately) specify analog output for errors, so you could just modify the value of the error condition only to make it obvious if it is the controller. if I remember correctly (haven't thought about it recently), although the analog output will act however you program it, on your logs you should still see the actual AFR, not what the analogs are generating. another way you could play around to see what is going on without going too crazy. see page 9/10 of the MTX-L manual.

it might not hurt to calibrate the sensor if everything else seems good but you still have issues...

[wheming]

...- I'm going to guess the WB output does the same thing (puts out 0v until the sensor is warm). so changing anything in the cal probably won't help if you are using the WB output of your sensor. only thing you can do for that is let the sensor warm-up cycle complete before running the engine, seems like it usually takes 5-10 seconds.

I did try that. Key on and let the wideband sensor heatup until it displayed a value and then cranked engine. After the start, the sensor went back into heatup.

I'll just need to find which table to change to increase the open loop time a bit for the warm restart.

When the o2 output was disconnected the ecu default o2 was like .53V as read on a drbII. Maybe i should just set the same default in the MTX-L? At least this way any fueling issues should be minimized?

[ShelGame]

I did try that. Key on and let the wideband sensor heatup until it displayed a value and then cranked engine. After the start, the sensor went back into heatup.


You must have it wired to an accessory circuit. Find a circuit that doesn't power off on start.

[Force Fed Mopar]

X2, my AEM stays on during startup and if I warm it up for 10-15 secs will show actuall AFR on startup.

[wheming]

Hmmm... i'll have to see where it gets the power.
Which circuit are you tied into?

I was just looking at setting the default warmup output. From the factory as stated earlier it is 0V. Would it be better if was set to 2.5V? At least then if came out of open loop, it wouldn't be adding fuel if the sensor was still warming for a few more seconds.
Its apparent there are several ways to skin this cat... just wondering the best option.

[ShelGame]

Hmmm... i'll have to see where it gets the power.
Which circuit are you tied into?

I was just looking at setting the default warmup output. From the factory as stated earlier it is 0V. Would it be better if was set to 2.5V? At least then if came out of open loop, it wouldn't be adding fuel if the sensor was still warming for a few more seconds.
Its apparent there are several ways to skin this cat... just wondering the best option.

It should make it run better.

But, it might still set a code. Once the ECU goes into closed loop, it expects to see the O2 signal toggle.If you set it to 2.5v, it won't toggle. And, it might light the CEL. On my car, the CEL does go away once the O2 signal is working as expected.

[wheming]

Thanks. I will give it a try.
I looked for the power wire and it might be tied to some power with the radio. Don't really feel up on ripping it all out to get to it. Just put the gauges back in after connecting the MTX-L programming cable, and running a wire for using 0-5V output.
I'd rather be driving it! ;)

[wheming]

Got a new question.
I was doing some logs of a cold start and warmup. I noticed the MPScan AFRwb seems not only to lag the actual from the meter (Innovate MTX-L) but also might be off slightly.
I have checked to ensure the gauge in MPScan is ranged properly (again in this case 0V=7.35 and 5V=23.5)

I was going to try this test if i can borrow a signal driver from work, or rig up a adjustable voltage device.
If i send a constant known voltage signal to the ecu, and find that what MPScan shows as a value is low, can i use the wideband offset to calibrate the input to actual? That is of course if the error is linear. If it is correct at some point in the range it might be more difficult. I would then have to make a compromise as to where i want the error.

[wowzer]

looking at the mtx-l settings it looks like the range goes from 0v=7.35 and 5v=22.4 and i assume you are using 100% gasoline and not an ethanol blend? and you confirmed the actual gauge output range setup using the innovate software is the same as mpscan? i know with my lc-1 i had a small ground offset i had to program in to my settings.

btw, can the mtx-l gauge display lambda? might be easier to use in the long run.

[wheming]

Yes, you are correct. I mistyped. The 5V=22.39 AFR.
I did not change from the default gasoline fuel type. I'll have to see if they have something for the 10% EtOH blended gasoline. But my fuel stations say "may contain..." so you never know. But basically we can assume if 10% ethanol is cheaper they are using it.
I will verify again since i mistyped, but yes i thought i had all the ranges scaled the same both in MPScan and my MTX-L.

And yes, the MTX-L display can be programmed to display lambda. And i am just starting to get used to thinking in afr, i don't know lambda, its Greek to me! ;)

[Ondonti]

looking at the mtx-l settings it looks like the range goes from 0v=7.35 and 5v=22.4 and i assume you are using 100% gasoline and not an ethanol blend? and you confirmed the actual gauge output range setup using the innovate software is the same as mpscan? i know with my lc-1 i had a small ground offset i had to program in to my settings.

btw, can the mtx-l gauge display lambda? might be easier to use in the long run.


Yes, you are correct. I mistyped. The 5V=22.39 AFR.
I did not change from the default gasoline fuel type. I'll have to see if they have something for the 10% EtOH blended gasoline. But my fuel stations say "may contain..." so you never know. But basically we can assume if 10% ethanol is cheaper they are using it.
I will verify again since i mistyped, but yes i thought i had all the ranges scaled the same both in MPScan and my MTX-L.

And yes, the MTX-L display can be programmed to display lambda. And i am just starting to get used to thinking in afr, i don't know lambda, its Greek to me! ;)

For understanding purposes, you don't modify wideband settings for fuel mix changes. Spray a bunch of methanol and you don't think about the STOICH for methanol. Its not like the MTX-L is converting for E10 and the MPScan is not. There is something off.
Fuel blend does matter when sizing fuel injectors in some setups like Megasquirt. Or if you were running flex fuel.

[wheming]

looking at the mtx-l settings it looks like the range goes from 0v=7.35 and 5v=22.4 and i assume you are using 100% gasoline and not an ethanol blend? and you confirmed the actual gauge output range setup using the innovate software is the same as mpscan? i know with my lc-1 i had a small ground offset i had to program in to my settings.

btw, can the mtx-l gauge display lambda? might be easier to use in the long run.

I hust found MPScan did not save my change to the gauge setting fot AFR_o2v, it was back to the default 20.9. I changed back to 22.39 and saved (there is no idication that MPScan performed an action. I hit the save button a couple times). Then i closed MPScan and reopened to verify my change was saved, and it was. I must have missed saving my change the other day.
I will try a log out later today. And verify the settings in MTX-L.

[wowzer]

Ondonti is correct - i should not have included the comment on gasoline blend since it was not applicable to the issue at hand.

wheming - i'll add a message when you save the data. it was an oversight. sorry.

[wheming]

Ondonti is correct - i should not have included the comment on gasoline blend since it was not applicable to the issue at hand.

wheming - i'll add a message when you save the data. it was an oversight. sorry.

Hey, its ok. I didn't mean to sound overly critical if i did. I appreciate and thank you for your support.
I knew your comment about the fuel wasn't the issue for the observation i had. It's a "instrumentation calibration" issue in my opinion. However, since now i found the MPScan had the incorrect range I'm hoping that fixes it.
I was worried it was an error due to different grounds, gauge ground and smec ground, and then how to correct. Data is only meaningful if it is accurate!
I'm going to try this shortly. Just woke up.