explain like im 5.... too tired to think.

Is it possible to use a higher grade lab type sensor with a.... mtx-l ? :)

That's exactly where I am now. Everyone says theirs' is the best but no one says why. There are no comparo's or hard data on why any of these are any different. At this point I'd rather have feedback from the community indicating whether their car blew up or the controller would act a fool at the most inopportune times.

I had an lc-1. Two controllers died. And both would love to give incorrect figures right before their deaths. The company told me to buy more crap and when I wouldn't they told me to go kick bricks. I didn't like that. I want my next wide band not to do that.

And now I'm getting to the point of wondering if either this degree of precision is necessary for our applications or most of these companies are clueless.

The other day I saw this:
(watch from 10:00 to 11:00 or you'll definitely have a stroke ). I was under the impression that tail pipe o2's are about as accurate as tuning a carb by ear. But apparently that's good enough for a tuning company that has the market on lock for quite a few models.

Regardless, who am I supposed to believe: a bunch of engineers that have more brains and schooling that do what is generally frowned upon or the enthusiasts that build cars out of wood and hopes that can take double the power under track abuse?

Buy a higher end unit that just focuses only on precision and accuracy with future sensor options. No.

Don't assume that COBB uses this method for the calibrations they send out en masse. You can run a tailpipe sniffer with compensation. Usually you see this with customer cars on dynos to get the cars in and out quickly in an economically fashion. The preferred method is closer to exhaust header and installed in the pipe.

Even the worst wideband is dramatically improved from a few years ago, even the Innovate stuff. For a naturally aspirated or low boost general daily driver car nowhere near the edge, you should be ok with any of them for general tuning. Read your plugs and use other validation methods if you go with the cheaper units. You can knock out a sensor quickly on even the best unit if your tune is so far off you are dumping raw fuel, have backfires, put the sensor in disconnected, etc. Be sure to get in the general ballpark before putting in any system.

Best of luck in your search!

Get an aem, connect 3 wires, and move on with life.
I refuse to use innovate until they figure out a way to not throw the stupid thing into "warmup" every single time you re-start it regardless of how hot the sensor actually is.

I mean my AEM and MXTL both are off in TS, and required the same amount of effort to get going. Between the 2 I like my MXTL a bit better, just because I like the gauge better. I would not suggest an LC1 to any one any more since now there are better ones out.

Just because the AEM doesn't actually show that it is warming up, doesn't mean that it isn't. The fact that it always shows 14.7 for the first few seconds of operation should give you a hint.

I'm not disputing that.
But 3-4 seconds is much less time than 10-15.
Maybe the LC2 is improved in that regard. I've not tried it yet.

It's definitely not 3-4 seconds.

On a hot restart the old LC-1 only takes 4 or 5 seconds to come back online. I'm usually not in boost within the first few seconds of cranking so I don't see how that matters to anyone... :dunno:

...because I'm not tuning while doing LeMans starts.

https://nineelevenophilia.files.word...mans-start.jpg

Sub 10-second warmup on an LC-1 here. Never noticed if the AEM in the last car was any faster, so clearly it didn't bother me.

Reference a really old post of mine. my LC1 warmup time was 7 seconds.

Too long. I grow a full beard in that time.

i couldn't grow a beard in 7 years, so Im good. and im not stuck with the worst gauge in the world.

The installation complexity is the same for nearly all controllers. Get XX unit, then connect power, ground, & sensor then read data.

All wideband sensors require warmup and none return accurate data when cold. Where they stop the countdown is a design side choice and arbitrary. The sensor is not warmed up in 3, 7, or 10 seconds and faster is certainly not better in this case. This is from another thread ( https://www.miataturbo.net/ecus-tuni...4/ ):

Warmup
All wideband sensors have a fixed heating element and need to be in their working temperature range to function properly. Most controllers go full power on the heater element immediately and then modulate down based on feedback. The Innovate LC-1 specifically modulates heater circuit in a way that causes premature sensor failure.

Whether a manufacturer chooses to show you a countdown or just immediately show values is just a choice. Functionally, at the sensor level, most are doing the same thing. Ones that show a countdown, I would argue, are the more intelligent manufacturers because they get the everyday people who don't know this to provide some basic sensor warmup, extending sensor life. No wideband sensor will work when you immediately fire it up cold and if you want your sensor to live longer and provide accurate readings, you should let it warm up electrically before firing the engine. We don't all have 5+ minutes to wait before firing our vehicles so, as long as you understand that the sensor needs to be up to temperature for accurate readings, you will be good. If you put the sensor in a place that is too cold, the heater element will have to overwork and my not be able to stay in its working temperature range, providing poor readings.

[¹] Citation Needed.

You might crack the top 10. The early failure rate of the LC-1 was catastrophic and the LM-1 was so much better. Maybe they resolved many issues mid-run. It is up there with old "wideband" gauges and sensors that either just extrapolate the analog range out of a switching narrowband or one of the Bosch highly tolerant narrowband designs.

The student designed controllers and breadboard controllers from people who don't understand the function of the pump cell typically take the cake. There are a surprising number of those projects out there. I still get frequent calls from people who think that wideband sensors just output voltage on a 0-1.2v scale like a standard zirconia cell or just magically dump 0-5v and don't think a controller is needed at all. So maybe that is the worst.

In the late 90s and early 2000s there were a few projects that cost thousands of dollars using the Horiba/NTK/Honda sensor that were pretty awful.

What part of the early failure rate over 10 years ago concerns me?

My first LC1 failed in 2007 actually--it was installed in a horrible location and is probably what lead to the failure. I had a replacement in a week which worked well until the day I pulled it out of my car in 2014. I plan to use it again eventually in my WRX.

the AEM uego gauge itself is awful--that's what i was referring to: it's ugly and bulky, and the refresh rate is slow.

Two things.

1. They improved it in later product.
2. I'm not in the habit of telling companies exactly what mistakes they made in design and telling them exactly how to correct it. Not for free that is. It is a function of my business. Design engineers don't, as a habit, give away IP for free... The exception is student events like FSAE where I do give away free design advice to students.

The failure rates on the early LC-1s were catastrophic, which they obviously worked to resolve. Most of the engineers I spoke to in the industry at the time were aware of the problem and just waited for Klaus to work it out.

Ok. More anecdotal proof on the early LC-1 issues but otherwise, I thought that is what you were referring to and did not choose to research your LC-1 replacement history. So, miscommunication it seems.

my other friend's LC1 died because he installed it near the turbo and it melted to death.

more proof.

You gehs want a video or something? It warms up and starts showing actual afr in 4 seconds or less :dealwithit:

My MTX-L on the other hand, was retarded slow. And I got it new. And from a reputable vendor. And it wasn't defective.

It's whatever, I don't even care what everyone else uses lol