FWIW Info on narrow band O2 sensors

[Sleeper]

I thought I would post this.
It answers a lot of questions for those curious about narrow band O2 sensors. It's from the Innovate WB board.

"Using a narrowband O2 sensor to measure/tune AFR or Lambda is pretty much impossible. The only Lambda number that will be correct is 1.0.
The reason is as follows:
A typical NBO2 sensor puts out a voltage between 0.1 and about 1.1V. Around Lambda 1.0 (14.7 AFR for gasoline) the voltage jumps from low voltage to a higher voltage as the mixture gets richer. At Lambda 1.0 the output is 0.45V typically. The curve at that point is very steep. Below Lambda 1.0 (richer) the voltage rises slowly the richer the mixture is.
BUT:
Although the slope of the sensor output curve at that rich region is fairly constant, the voltage offset is not. Depending on sensor temperature, an AFR of 12 can be 0.7V or 1.1V or anything in between. This means that even though you could make a millivolt table measuring AFR/Lambda vs. output of a NBO2 sensor (and program that into a meter), this table will be only correct for one particular sensor temp. for each of the measured points.
For example:
865mV on the NBO2 sensor can mean anything between an AFR of ~10 (high sensor temp) to an AFR of ~14 (low sensor temp).

If the sensor temperature is not precisely controlled (and it is not, typically), the millivolt outputs of the O2 sensor can tell you only rich, stoich, lean. Basically saying that the output is above, at or below 0.45V. And that is what it is designed to do ONLY.

Measuring AFR with a NBO2 sensor is like trying to measure battery voltage with a relay. If it clicks, voltage is there, but it's impossible to tell if it is 7V or 18V.

A lot of people believe there is a 'magic' mV number to tune while using a NBO2 sensor. This mV number is of course useless for the above reasons. It can change (by several hundred mV) even depending on the state of your battery and charging system (more or less heater voltage for the NBO2 sensor).

Regards,
Klaus"

Paul

[1badTTA]

Damn, and to think all of our cars are so far out of tune that it is surprising they even run :racer:

That information is useful for people who think that tuning by O2mv (narrow band) is an exact science and it shows that it isn't. Suppose it works pretty good for selling WB O2's also. But for htose of us who know how to tune and realize the downfalls of a narrowband O2, it is quite possible to tune with one.

BTW, exactly what A/F ratio will produce the most power on my car? I want it down to the hundredth also :SHOCKED: Also, how much timing will work best, and how much boost :fragez: If you can't answer that then you won't convince me to throw away my inaccurate NB O2 just so I can have a more accurate number that means almost nothing.

I am not trying to argue that NB O2's are anywhere near as good as a WB, just that when used with other sources of info such as reading plugs and an EGT meter they are just fine for most people and spending several hundred on a WB will not make the average person any faster except for the reduced weight of their wallet.

[mgmshar]

Another thought on this...

I use my NBO2 sensor more as a "sanity check" than an exact tuning tool. If you're using the same O2 sensor, and you do each of your 1/4-mile passes consistently, then your NBO2 sensor readings in third gear should stay fairly consistent from pass to pass.

Let's say I make three passes. On the first two passes, the O2 voltage is about 750mV at the end of third gear. On the third pass, the voltage drops to 720mV at the end of third gear. I then know that something was wrong on the third pass.

As far as using it to figure out exactly what the A/F ratio is at any given moment, well, you might as well sniff your exhaust gas and take a guess. NBO2 sensors are primarily a "rich-lean" switch, as Innovate suggests.

[Sleeper]

1badTTA
I guess I should have prefaced the post with... For those that are new to tuning, and want to know why trying to tune with NB O2#s alone, isn't very accurate... ;) It was just general info like I said "FWIW" :)
I'm not asking you to throw away your NB sensor, (our ECM wont read the WB anyway) and of course you can tune using EGT and plug cuts, combined with using the NB # as a guide line, but not all tuning is done at WOT. A little more accuracy when working on lean cruise economy, emissions, general tweaking, etc. wont hurt. And if you are a serious racer, it is going to help (ie. Fast) But I agree with you, if you are the average Joe :marchmell ... save your money.
To me it's just another tuning aid. Just like some are happy with ScanMaster, others prefer DirectScan.

Paul

[Bruce]

[]l[/QUOTE]

I've run WB's for years now.
I've run them at the same time as a NB, and data logged both.

In back to back, passes, at 810mv on a NB, read, 11.5:1 AFR, and 12.3: AFR. If you think that is close enough, then by all means look at the NB readings. IMO, that range of error isn't acceptable.

Has anyone noticed how often GNs blow headgaskets?.
Naa, no possible connection.....

The days of running static injectors, and guessing at what AFRs one's running are on the decline.

And for those the can solder, you can do your own WB's easy enough, and the Innovative one is very reasonably priced, IMO..

[1badTTA]

Sleeper and Bruce, I am not disagreeing with either of you that a wideband is much more accurate and a narrow band really isn't designed for what we are using it for. There is a reason that the ECM doesn't pay attention to it at WOT. I am just trying to show the flip side of the coin compared to the original quote which is definately one sided.

Simply looking at the graph of afr/O2mv will show how accurate a NB really is and how it isn't ideally suited for tuning the way it is being used. However if you can remember back about 20 years, there was something known as a carburetor used to meter air and fuel into engines that didn't even have a narrow band O2 sensor for tuning and some of them actually ran pretty good. My point is that if you know the limits of what you are using then you can make it work for you.

[Sleeper]

1badTTA
Appreciate your input, and for looking at things from the other side.
You are correct about being able to to tune with a certain degree of accuracy using the old methods. (Believe me, I'm learned on Q Jet's, and by messing with distributor weights and springs ;) ) Back in the old days, plugs were a lot easier to read, (Leaded gas) gas was better, and 1-1&1/2 HP/cu in, was thought to be the max possible. As you know... Technology has changed a lot since then. Why not use it? Why settle for running "pretty good"
No doubt that you can still set up your car to run really well without the WB stuff, and there are lots of guys out there that do. (Some of them are really sharp) But it sure can save a lot of time in trial and error guessing, as well as head gaskets ;) when you are pushing things close to the edge.
I believe that there is a lot of hidden HP to be found just by fine tuning.

Paul

[Bruce]

Simply looking at the graph of afr/O2mv will show how accurate a NB really is and how it isn't ideally suited for tuning the way it is being used. However if you can remember back about 20 years, there was something known as a carburetor used to meter air and fuel into engines that didn't even have a narrow band O2 sensor for tuning and some of them actually ran pretty good. My point is that if you know the limits of what you are using then you can make it work for you.

Most of the graphs for the NBs are inaccurate, few if any show the errors due to EGT, and backpressure.

I've spend many weeks doing custon emulsion tubes for webers on 12 cyl Ferraris, so I'm more then clear about carbs, and how much they can generate HP wise, and the drivibility they offer.

But, why spend the weeks, and the gas trying to get the calibration CORRECT, when for $350 you can call B+E Customs and have a WB in a few days.

A WB and any serious tuning is just an investment rather then a cost. If it's not for you, then you can oft it on eBay. But, once you get dialed in on using one, it will pay for itself in gas never mind the wear and tear and the time you'll spend tuning without it.

If the AFR changing alot a full AFR ratio is close enough for you, that's fine, but to infer that's it's OK, and tolerable for everyone is misleading, IMO.

Not to mention trying to guage Tip-In AE. You can spend days guessing at it, and in just a small percentage of that time, have it correct if you have a WB.

BTW, in oem form WOT fueling is mostly about keeping the converters alive, on the TRs.

And CARB the last I heard wasn't fully satisfied that the oem O2 sensors really are accurately sensing O2. If DIY-EFI ever comes back on line, there were some interesting threads about that.

And Plug reading is another issue. Needless to say, it's the only current way of being able to see into what's happening in a combustion chamber. In both my GN, and truck I can read the coloring just as well as several decades ago. It's still about the tune. With the right timing, and fuel I get the same beige coloring, other then with Shell. They're Platformate or what ever they call the Manganese additive just turns em pink.