MS3 Extra 1.4.1 FW (though similar occurrences under 1.5.2B4)
1999 Stock Engine
LC-2 Controller and 4.2 Sensor.
Advanced Controller settings of 0.77V for warmup and 0V for error
Calbrated to match TS with Logworks
Sensor Calibrated

The issue: The AFR reported by TS log shows that it stuck at 14.5, as the AFR table required 12.6 or so. Thus, MS saw about AFR Error > 2 for 2 seconds, and AFR shutdown was initiated. Status 6 and CEL status verify that it was AFR Shutdown.

This occurred at 296 second mark in the attached log.

Thoughts?

 

Nothing much helpful to add other than I have lots of logs of my LC-2 / 4.9 combo doing very similar things. I also have/had it hooked up through a custom board to log everything so I could see error codes from the LC2 in logs. More often than not it would do what yours does there, completely stall before snapping back into shape as if nothing were wrong. It happens over canbus, and it happens on the analog output too. It usually does not throw an error code but eventually it will throw an E4 error after the stalling gets worse. I went through a couple new sensors before I scrapped the piece of junk.

It can also apparently forget it's programming, but I never had that happen other than when I was trying to send the serial calibration request to the LC-2. Somehow the stoichiometric value got switched to 7.8 or something. Have you checked to make sure the error value isn't defaulted?

 

I checked the settings before I posted.

Did you go AEM?

It’s funny, I have had the same 4.2 sensor for about 5 yrs now and it never changed readings with an air calibration. And never an error code.

Thanks for for at least letting me know I’m not nuts.

 

Been tuning a car with an MTX-L for a little over a week, and my opinions aren't favorable. It does multiple wonky things. I much prefer the AEMs I typically use.

 

I got a 14point7 spartan 3 ADV, looks like I got it around the end of May and haven't had any issues since. I also wired up an interrupting relay so it's never on if the engine isn't, for whatever that's worth.

I went insane thinking I had somehow broke the LC2 since I was doing my own thing with it. Wasted more time than I care to admit redoing things several ways to make sure it wasn't on my end. Turns out just reading things off the serial port doesn't bother it at all, it just sucks!

 

My MTX-L plus also had hanging readings and errors after 500 miles, new bosch 4.9 fixed it. Not impressed by that, but when it works, its awesome.

If the new sensor burns out in a few months, I'll go AEM

 

I gave up on the LC-2 years ago. It was continually doing weird things and it got to the point that I couldn't trust it for anything. I even went through multiple Bosch sensors before I finally threw in the towel. I finally settled on an AFR500 from Ballenger (after tolerating a PLX Devices box for a year or so) and haven't looked back. It's been reliable and stable - only requiring a calibration once a year.

 

For now, I have set AFR Shutdown wait to 5 seconds to ride through.

AEM with CAN seems a good choice, but, as many know, I just reverted to 1.4.1 FW. I think I'd like to wait for 1.5.2 release before going back up.

I am planning on a new FMIC, so I may need to do some minor tuning after that.

I admit that I have been enjoying leaving the laptop at home. I have an unbelievable amount of time in tuning this car to get it like I like it. It's been fun, but part of a good tune is that it works right no matter the weather, hot starts, cold starts, high TPSDot, etc. Some items are iterative, so there is some elapsed time requirements.

But I digress. Thanks for the input guys. I will research a bit more. I cannot live with intermittent issues, when a cure is available.

 

@der_vierte Did you upgrade the controller FW at the same time? Could that have made part of the difference?

EDIT: Deezums had the 4.9 and still saw the problem. Oh well. I did send an email to Innovate, but I'm not expecting a solution from them.

 

@deezums How did you measure and account for any offset error? With the LC-2, I could connect Logworks and get the digital reading to compare to MS3 and put in offset. I also note that they are not presently selling the 3 ADV?

 

I didn't, it can be configured to broadcast at a specific CAN offset and baudrate so it should be able to work with CAN receiving even on 1.4 right out of the box. The directions do say you need 1.5 which I think added CAN EGO fields to CAN receiving. Otherwise you can configure the analog output using bluetooth, or at least I am pretty sure you can. I wired it up to the analog EGO input just because, but only because megasquirt is still pretty ******* stupid and won't let me use that "spare" analog input for anything else. I could try and offset it using the can output, but eh...

Any innovate serial device runs at 19200 and only sends packets every 83.5ms or something like that, then adds on whatever latency from your serial to can converter board. The ADV3 pushes 20ms updates over canbus. It's a good deal faster. No need to mess up megasquirt canbus protocol either trying to force 1kbps baudrates. You can still use a tinyiox, iobox, or whatever other canbus hardware you have.

And it looks like they are out of stock again till December. That sucks, probably the biggest downside of the thing is it's a smaller manufacturer. I've been trying to get a buddy to order one to replace a LC2 I suggested that has ate 7!? 4.9 sensors within 2 years. All of these "dead" 4.9 sensors I have work just fine on a 14poin7 spartan 2 OEM, and I'm up to around 10 of em now. I should buy three more of these cheap little spartan2 OEM things and do full per cylinder correction....

 

Good idea, I should do that.

For now I have a switch on it, to make sure it is powering up after a minute or so and hope its the heater/condensed water thing, that kills the sensors. Read something about that it is a good idea to NOT heat up the sensor before the car is running, but you guys know that of course.

Fingers crossed, I really like it somehow, but if it burns sensors for 200$ a year, it's going in the bin

edit:
crazy that "innovate burned out" sensors actually work on other controllers, that goes beyond my understanding

 

If you dig deeper into it I think innovate uses Bosch sensors but tries to extract useful info from them in their own proprietary way. Almost everyone else uses Bosch defined protocol, sometimes on a Bosch supplied control IC, CJ125.

The Spartan2 OEM doesn't appear to use the bosch CJ125, but I think most all AEM devices do. But basically, there's more than one way to get "data" from a wideband sensor. Fastest or most accurate way is up for debate.

 

as soon as I see any piece of trash innovate product being discussed in any thread, I just close the thread so as not to waste time discussing trash.

step 1: throw innovate in the trash
step 2: get aem
step 3: not have this discussion again

 

Thanks to all.

I will go with AEM 30-0310 in line (no gauge) controller with Bosch 4.9 sensor. I will dart with analog, and change to CAN if I upgrade MS to 1.5.X FW.

I don’t use any other CAN so AEM non-compatibility should not be an issue.

/Thread

DNM