Got a question ... had my car on AMS' new dyno this weekend and we noticed that, after a few disappointing pulls, the AFR was right around 10 (this is on a custom map tuned for low 11s). A couple days before the dyno, I swapped my cat back in, and noticed that I'm still running my "old" O2 sensor, which looked pretty bad (pitch black). I have also dialed down the boost by ~4%, since I was starting to go past 24 psi with the colder temps around here. So I guess the questions would be:

- How does SMART compensate for a flaky/bad O2 sensor? Does it go pig rich if it doesn't get a valid signal from the LC-1?

- Would adding the signature series HF cat throw the AFRs off at all? (seems unlikely)

- Would pulling boost (4%, i.e. 24-25 back down to 22-23) throw off the AFR by that much?

l8r)

 

I'm not the expert, but i'll give these a try.

Depends on your smart fuel map. If your sensor is reading way lean or way rich then the smart map is going to modify fuel appropriately. However it won't go beyond the authority range (top or bottom row).

Uhhh, i don't see how it could so long as your LC-1 is in the downpipe where god intended. I think you'll probably ok in this area though.

It shouldn't matter, methinks. Here's my argument:

The ECU is in open loop mode, so fuel delivery is totally dependent on MAF reading. Whatever you do with the boost, the MAF sensor will see it and the apply fuel appropriately. I didn't think that SMART fuel was loaded by FreqMAF, just RPM. So the smart fuel trim isn't going to be affected by your new, lower boost pressures.

EDIT: I also see 10:1 under full throttle on my very stocklike car. I don't think that SMART has improved my power output at all. One thing that was never resolved in my head is whether that 10:1 target is too rich. It probably is, but I've been too busy with school to care. I know that I almost never, _ever_ see knock -- even on 91 octane.

d

 

Just checking to be sure... Are you running a SMART fuel map, etc?

 

Boy Ludi, a few months ago, we seemed to be enjoying our cars, NOW LOOK! I cant help you here. but good luck. We got all winter to solve our issues...................and mine is still LOUDER when its not on a tow-truck or on the lift!!!!!

 

I'll just throw in my $.02, maybe something will help.

First, i'd make sure the O2 is calibrated correctly. I would recommend a true fresh air calibration - take it out of the downpipe, let it warm up, and hit the calibration switch. I have the XD-1 gauge, it helps alot with making sure the LC-1 is doing what its supposed to.

Next, i had an issue with my LC-1 just after install on one of the first tuning runs i did. for some reason, it did not boot up correctly. the LED on it was flashing, and the xede was not reading a correct value. The part where this comes into play, is that (in my experience at least) when the LC-1 was on the fritz, the xede was actually seeing a 50% voltage value: so it was reading a stoich AFR. So, instead of the xede seeing a 'broken wideband' it sees 'hey, i'm running a 14.7 AFR right now.' You can imagine that when the smart map kicks in, and sees a 14.7, its going to dump fuel in that thing like a sum*****

have you datalogged the O2? are you sure its getting a good reading? Do you have the LC-1 calibration switch and LED hooked up, and is the LED doing anything funny? I assume your O2 is installed in the recommended spot, just before the cat in the downpipe.

Also, i dunno how "old" your "old" O2 is, but i thought i saw a recommendation by Innovate to replace them every year or something like that. Its good insurance on a smart car for $80. I'd go ahead and get a new one just to be sure, or maybe grab a friends and plug it in.


In regards to the ~10:1 AFR of the SMART map in the xede: the official statement of shiv is that there is no exact science for the xede reading the LC-1 voltage. there have been variaces in signal, enough to create the smart map on the safe side, thus a low AFR. That said, shiv has also posted up an advanced smart map for higher AFRs, or people running higher (93, 94) octane fuel. Personally, I created my own smart fuel map, somewhere inbetween the two.


Like i said, hopefully something i said here helps


EDIT: and when you say "custom map tuned for low 11s" was this pre-smart? if so, and you just enable smart with the original smart fuel map, it is definately going to add in some fuel and pull it down into the low 10s. If you highlight the entire smart fuel map then do a run, it will highlight the boxes yellow that it passes through. if its passing though a bunch of cells with positive numbers, its adding more fuel to your tuned fuel map.

 

Great answer Sackett, that otta help him out! very informative stuff.

 

Sackett, thanks for the info.

I should have been clearer in my initial post. My SMART setup was tuned by Shiv at Dyno4mance, and it dynoed exactly where it was supposed to (low 11's AFR).

However, now, about a month later, it's running a full point richer. My O2 sensor has been in the car for over a year, looks filthy as hell, and has been abused from time-to-time (one tank of leaded, and a few hundred miles of driving w/out the sensor being pre-heated). So I'm fairly sure that it's the sensor that's starting to go ... BUT ... my main question really is "what does the SMART system do when the sensor starts to get flaky?" Is there some kind of failsafe built-in, or do we just have to be really **** about making sure that we have good sensors...

l8r)

 

But when the sensor fails, doesn't it fail in the other direction (towards lean)?

[quote]
... BUT ... my main question really is "what does the SMART system do when the sensor starts to get flaky?" Is there some kind of failsafe built-in, or do we just have to be really **** about making sure that we have good sensors...
[quote]

SMART can never add more fuel than the top row of the smart fuel table. In the examples of I've seen (and use), that is about 3-4%. I believe that corresponds to about 1 point. What would _really_ be useful is some kind of warning mechanism to alert the driver that smart has been working overtime lately.

d

 

the smart fuel table makes changes based only on the signal its getting from the LC-1. so at anytime, if the LC-1 is out of calibration, not working, going bad, etc. etc., the smart system is going to have unpredictable results. Like i said, when i noticed a problem the xede just thought the car was running stoich all the time, so the xede will try to adjust accordingly and add that extra 3-4% of fuel to try to get it back down.

so my guess would be a bad sensor, or maybe something else going on with the LC-1. When you get a new sensor, i think there's a procedure to get the LC-1 calibrated to the new sensor too, make sure you look into that.

This is another reason i recommend that XD-1 gauge for the LC-1. You know when your 02 is broken, or when you are hitting 10s when you should be in the 11s. It also has a serial connection to the LC-1, so its always got the correct reading, when the analog signal of the xede may be off a bit.

 

Yeah, I did get the XD-1, just haven't had a chance to hook it up yet. I also have an LM-1, which I have to figure out how to daisy-chain into the LC-1/XD-1 combo.

I really would like to know what the SMART system does when the LC-1 generates an error signal (i.e. bad sensor, sensor too hot)...

l8r)

 

I vote for: "Danger Will Robinson! Danger!!"

l8r)

 

So if you really think that the sensor is bad, plug into the LC-1's serial output and at idle you should be getting ~14.7. If the calibration is off you will see something different. If LC-1 errors are present the Innovate software will display them as text on top of the AFR reading.

Another thing to take note of is that once I had my LC-1 loose its mind. Specifically the analog output setup changed to something that was very wrong, all by itself. The LC-1's configuration tool is your friend to determine if the default analog output configuration is still setup within the LC-1's memory.

Could you share with us a screen capture of your smart fuel map?

 

I'll try to post my fuel table later when I have access to my laptop.

As far as reading stoich at idle, I have two problems with this:

1.) My car's AFR varies quite a bit at idle. As the ECU "hunts" for stoich, the AFR goes anywhere from 11.5 to 16 or so. It does stay around 14.x alot, but not long enough for me to assume anything about calibration.

2.) I was under the impression (from Klaus at the Innovate forums) that even a sensor that is going bad and/or miscalibrated will read stoich correctly. It isn't until you get into the richer AFRs that the miscalibration will become visible (and then only when tested against a known gas mixture).

l8r)

 

When the sensor fails, it goes to zero row, meaning that it will have no effect on fueling. However, if the sensor begins to act funny during the failure process (incorrect readings) The car will feel very inconsistent, and the sensor will most like err on the side of lean. However, after talking to the guys at innovate, they said it generally will just narrow the spectrum that the sensor is able to read.

 

There really is three different issues that are possible: bad sensor, miscalibration and a configuration error. To be honest I don't know about sensor failure, but with calibration and configuration errors I have clearly seen an offset in the output.

I have to admit that on my car at idle the ECU is able to hold to a small range of air fuel ratios as one would expect. I didn't know that you were all over the map.