More Wide Band through AC Problems

[5_Liter_Eater]

My cheapness may come back to bite me in the ass but I picked up an Innovate MTX-L. I'm holding out hope that it will prove more reliable than I have seen the LC-1's be.

[pontisteve]

The Bosch sensors seem to crap out a little too often with the Innovate stuff. From that point of view, sounds like the NGK AFX unit would be the hot ticket there, since the control strategy for the heater has good engineering from ECM. But at least the Bosch sensor is cheap. Get mine from Amazon for like $55 shipped.

The MTX-L is just basically a sensor and a gauge, right? Does it have analog outputs? Can you change the output with LM Programmer for fuel stoich and AFR vs Lambda?

[5_Liter_Eater]

Yea the AFX is the bee's knees but for some reason I have to make things harder on myself. I went with the MTX based on price point and the fact that it will fill the hole the AEM leaves in my pillar. It's got two analog outputs that are fully configurable just like the LC-1. I like that as it makes it easier to verify the offset and transfer function. I also like that there are two outputs as I will send one to the Aeroforce and one to the AC for input to HPT and not have to worry about splitting the signal. I also like the ability to do a free air cal. You can even configure what ranges you want the green, yellow and red parts of the gauge to represent. On paper it seems good. I just hope it's not just an LC-1 + XD-16 in one as I have seen many an LC-1 lose its config, not take a firmware flash, bork sensors, etc. You may see me post up a "I shoulda had a V8, err AFX" thread in a month or two but hopefully not.

[pontisteve]

I haven't had much trouble with the LC1/XD16 setup, other than eating up sensors. And some of that was probably due to my tune being rich, which kills them prematurely.

I have had some quality issues with the XD16 gauge bezels fading out and looking bad, but I personally haven't had issue with flashing, firmware, etc. It could be that some people's trouble with that is in their USB to serial adapter.

I did also have problems when I tried to run two separate LC1/XD16's without daisy chaining them together. The Innovate software didn't like that it kept finding different components. I finally daisy chained them and all works fine. I wanted a permamently installed twin wideband setup in my car, but I actually much prefer using my LM-1 for tuning.

If your setup has all those options, then what do you get with the XD16 expensive gauges that this setup doesn't have? Configuration for displaying other sensor data that I never use anyway? I do really like that the XD16 has calibration buttons built into the gauge face.

[5_Liter_Eater]

Thats good to hear that you have not had those types of problems.

I have no idea why they would even still have the LC-1 + XD-16 on the market.

Yea, to call the MTX-L you need to power it up without the sensor attached, shut it down, and power it back up with the sensor attached and in free air.

[ArthurJGuy]

I have nothing to add to this but I have a question that is semi related 5_Liter_eater.

What pin on what connector is your wideband connected to, or did you go under the hood directly to the sensor connector?

[5_Liter_Eater]

Its soldered in right at the PCM. I don't remembwe which pin it was.

[pontisteve]

I think the XD16 is more high end. It has min/max, free air calibration button, and can be configured in LM Programmer to output any thing in the daisy chain. Using an LM1 or something with analog inputs, it can display lots of different things like boost, etc. None of which I ever use it for. But the CAL button is worth it. Just hit it after the car has been parked for a long while, and you're good to go.

[5_Liter_Eater]

Heres the E40 AC PCM pin info: http://forums.corvetteforum.com/c6-t...g-diagram.html

[5_Liter_Eater]

So I got the MTX-L installed over the weekend. I have it installed the exact same way the AEM was. Its powered from the alternator via a relay and gounded to the body. I haven't plugged HPTuners in yet but I have one output of it going to the Aeroforce with the Aeroforce slope and intercept set exactly as they should be based on the configuration of the MTX and it matched up exactly across the board. Didn't have to mess with ground offset or anything. So I'm not going to call it good yet but it's looking promising.

[pontisteve]

Powering up the gage and wideband should be your preference, but the analog ground should be hooked up to the same source as the HP Tuners ground, if you're inputting it into a MCVI Pro. And since that's built in to the HPT device, and the only ground the HPT device has is the OBD2 port, then it's the PCM ground that we're really talking about here.

I check analog output often on the SCT stuff. Their datalogger setup has an analog ground wire coming out of it. My understanding of how to deal with it, according to Klaus at Innovate, is to run the analog ground from the wideband and from the device to an engine ground, and to pick a good block ground that doesn't have anything else trying to ground thru or near it. Come to think of it, that's really not any different than the HPT device, because the analog ground probably grounds thru the OBD/PCM on that as well. Anyhow, Teeing that analog ground to the engine block always seems to result in less of a ground offset. In the SCT software, I can input the entire wideband formula to my specs using a formula like (V*3.008) + 7.35 for the default 7.35-22.39 AFR range that comes loaded in an LM1.

Using the LM Programmer software, I set up the wideband to put out 2.5v at 7.35 and 2.5v at 22.39 AFR. That way, it spits out 2.5v all the time. Then I just use the formula "V" in the softare, so it datalogs simply the analog voltage raw. I should always get 2.5v, and if I don't, it's a ground offset issue.

Normally when I do this, I get between 2.38 and 2.6 volts. Rarely is it not somewhat of a problem. I did that math once, and found that typical ground offsets are usually affecting the wideband a max of .2 AFR, but that depends on the AFR range you set the wideband to.

Here's the article Innovate asked me to write on the subject, if anybody is interested. It's handy for understaning ground offsets, and setting up analog formulas, if nothing else.

http://www.innovatemotorsports.com/r...2_tutorial.php

The formula is a little different and more difficult for HP Tuners, since they use division instead of multiplication in their formula. The only difference would be to take the multiplication number (3.008 in the above example), and divide 1 by that number. 1/3.008 = .33245. That would go in the box in HPT. And you can't add any voltage offset numbers in the software, but you might be able to skew the AFR up or down the right amount by changing the other number on the right by the amount you're off, in AFR.

[pontisteve]

So I got the MTX-L installed over the weekend.

What exactly is this new wideband all about? Is it an LC-1 with a different gauge? Innovate already had a cheap display gauge, and they had a top of the line XD-16 fancy gauge that's programmable. So where does the MTX-L fit in?

[5_Liter_Eater]

Thanks for the info Steve. I have the MTX-L grounded to body under the dash near where it's installed. This happens to be very close to the ALDL. My thought was to tee off of pins 4 and 5 in the ALDL which are it's grounds and ground them to the same spot the MTX is as well as where they lead to now. The thought being that they are then all tied into the same multiple ground points. What are your thoughts on that?

Even if I cannot eliminate it I can adjust the offset to fit. Whats important is that the offset and slope remain static which they were not with the AEM.

[5_Liter_Eater]

What exactly is this new wideband all about? Is it an LC-1 with a different gauge? Innovate already had a cheap display gauge, and they had a top of the line XD-16 fancy gauge that's programmable. So where does the MTX-L fit in?

I'm not sure really, but I like it so far. It's like an LC1+XD16 minus some of the fanciness, all in one, for a low low price.

[pontisteve]

Thanks for the info Steve. I have the MTX-L grounded to body under the dash near where it's installed. This happens to be very close to the ALDL. My thought was to tee off of pins 4 and 5 in the ALDL which are it's grounds and ground them to the same spot the MTX is as well as where they lead to now. The thought being that they are then all tied into the same multiple ground points. What are your thoughts on that?

Even if I cannot eliminate it I can adjust the offset to fit. Whats important is that the offset and slope remain static which they were not with the AEM.

Well, without seeing a wiring diagram, I'm guessing that the OBD2 port isn't grounded to the body itself. I would think it's only ground source would be wiring going to the PCM, but I could be wrong.

And if it goes to the PCM, that PCM ground is what we really want to ground the analog ground to.

PCMs are sometimes grounded to the car body, right in the kick panel area, at least on a mustang. While that is convenient for them, I've always wanted to try jumping that PCM ground with an additional large gage wire to the engine block, to make sure the PCM can get a really good ground, seeing as how it's pretty important.

I don't think the OEMs care, because all the sensors are grounded inside the PCM to the PCM ground anyway, so there wouldn't be ground offsets in the sensor outputs. That's the whole reason they ground sensors to the PCM and not to the block anyway, and that's also why when testing voltage at a sensor, we should be using the sensor ground for the negative side of our multimeter, not the engine block. Although if there are no ground offsets, that should make much difference anyway.

With the PCM grounding to the body, it has to go thru the body to get to the frame or engine block or whatever.

FYI, there are usually 2 ground cables on a battery. The big one goes to the engine block, and it's primary function is to provide a ground path for the starter, since it draws huge current thru the block. (The starter is grounded via it's bolting up the block, and from there thru the neg battery cable up to the battery). But that much smaller ground wire on the negative battery post goes to the body, usually right near where the battery is mounted. THIS small ground cable, probably 8 gage, is the ground for everything in the car but the starter, including headlights.

Knowing this, we can see how grounding to the engine block is a better ground than most chassis grounds! When the starter is not engaged, that whole big negative battery cable is not really being used up as a ground path, so its wide open to being used by other things that we might ground to the block. Ultimately, we really want to be grounding important things to the battery itself. Some also argue that it's actually the alternator housing we should be grounding to, but I believe it's the battery. So does MSD, Big Stuff 3 and others, who recommend their electronics get grounded right to the battery directly.

By analog grounding the wideband and the HPT to the block, we're basically getting a pretty straight ground path thru the block, and thru the large negative battery cable right to the battery.

These grounding kits that are sold (mostly to import guys) just add a crapload of ground straps every where, giving multiple chassis connections to the block, battery to the block, battery to the chassis, etc. They work because no matter what you ground your crappy import electronics to, they find an easy ground path back to the battery. Sort of like having 20 roads going to the same town, you can't miss.

I do believe that electronics could be improved in a car by providing two larger negative battery cables, one going to each side of the car's chassis. Think of ground paths as highways. Small highways with lots of cars on them cause delays due to traffic. We can either reduce the traffic (run less electrical loads), or increase the size of the highway, and reduce the distance of the highway.

In my trunk mount battery for example, I ran a large ground cable from the battery to the rear frame rail. Then I ran a separate negative cable from that same lug on the frame rail up to the engine block. This way, things that need to be grounded can take whichever path is shortest, chassis or block ground.

Knowing this, I think jumping the PCM ground (while still leaving the factory ground path to the chassis) right to the engine block with a good sized cable would make for much less ground offsets in the PCM. In our widebands case, that might mean the analog ground circuit may have less offset issues, without having to ground the analogs separately to the block. And that can't hurt other PCM things as well.

[pontisteve]

I'm not sure really, but I like it so far. It's like an LC1+XD16 minus some of the fanciness, all in one, for a low low price.

Can you do a free air calibration via a button right on the gauge? That's the one thing about my XD16's that I really like. Wait for the car to sit over nite where the exhaust fumes are fully dissipated, and press the calibrate button on the XC16 before firing up the car. No need to pull the o2 sensor out of the exhaust to calibrate.

The rest of the programmable crap on the gauge, I really never use. But it can be used to display whatever logged item you want in the modular tuning serial path.

[5_Liter_Eater]

No, I have to disconnect the sensor, power up, power down, and reconnect the WB in order to calibrate it.