how far off is not having a wide band

[MaTTSuN]

using the stock O2 sensor how far off is the stoich reading? Im thinking its better than nothing for tuning and its all i got just wanted to get into a ballpark range

thanks

[mobythevan]

On a non-turbo car a narrow band sensor may get you running good enough to drive to work. On a turbo car, do yourself a favor and do not try to tune with a narrow band sensor.

 

When tuning with a NB02 sensor you ask yourself this question: "Do I want 14.7 AFR throughout the entire VE table?"

 

If yes, then use NB02. If you want 11.8, 12.5, 13.2, 15 AFRs or anything other than 14.7 then tune with a WB02.

 

Some people say you can tune different AFRs with a NB02, but people also say that V8 swaps suck, R180 rear end will hold and you don't need an ACT clutch.

[MaTTSuN]

kool thanks for the reply moby.

ill just hook it up for looks i guess the GF will love it haha

cant you use the WB02 with an autometer gauge or do u needthose cheesy light boxes?

those are not very slick units

[MaTTSuN]

so you have to use those lame wideband boxes or can i use my autometer unit ?

[Clifton]

You have to use a wideband controller with a WB sensor. Those Autometer and other NBO2 gauges just measure voltage. You could do the same with a DMM.

[MaTTSuN]

thanks but whats a DMM?

[Clifton]

thanks but whats a DMM?

 

Digital multi meter. NBO2's put out 0-1v dc.

[badjuju]

I've got an innovate LC1, they've got 2 outputs, fully programmable. I have one outputs set to 0 volts at .5 lambda value and 5 volts at 1.5 lambda. The other one I have set to 1 volt at 10.0 air fuel ratio and 2 volts at 19.9 air fuel ratio. The lambda output is going into megasquirt, and the air fuel ratio output is going to a digital volt meter that I mounted in the face of my change dish. Looks pretty sick, I'll try and take a pic today

[510six]

If you`ll look on this site there is a kit for conversion of Autometer A/F gauge in a true WBO2 gauge.

 

http://www.dynotunenitrous.com/store/scripts/prodlist.asp?idcategory=116

[JMortensen]

Digital multi meter. NBO2's put out 0-1v dc.

That's what I did. Worked pretty good for tuning in my Mikunis. I have some friends who did the same thing and found they were dead on when they put their cars on the dyno and compared the multi-meter to the WBO2. There is an old post that you can search for that gives voltage values for a/f ratios on the NB. The light displays are fairly useless IMO, but if you drive around for a while with the DMM you can actually do a fair job. I wouldn't use it on a high strung turbo car where things are critical, but worked great for me.

 

Oh, and moby, I have an ACT, an R200, and someday will have an LSx as well...

[strotter]

I have to agree with those that are saying that an NB can successfully used for much tuning - if you're careful, and you're aware of the limitations. I've had both an NB and a WB on my Z for a couple of years now, and have spent many hours watching them out of the corner of my eye. Kind of hypnotic, as a matter of fact.

 

The thing to remember with an NB is that they are really a binary device - they indicate "hi/low", below stoichiometric or above stoichiometric, and that's about it. Nothing in between, not even perfect stoich. They just toggle up & down. If you're tuning for stoich, which is the vast majority of the time, you're fine - you'll see it oscillating back and forth over a narrow range of values more-or-less centered. That's good. If it's stuck in the rich zone, or more dangerously in the lean zone, you're not done fiddling. NBs don't really give you any more information than that, but that's a bunch, really; most tunes start out more than a little rough, and need some fairly heavy-handed adjustments to start. And bear in mind, 99% of your actual use will be in that range of performance - just easing around, stop-and-go, waiting at lights, 70 on the freeway, that sort of thing. It's the 80/20 rule: for 80 percent of the results, you spend 20% of the effort. It's that *last* 20 percent of results that take 80% of the effort.

 

That last 20% is when you're not tuning for stoichiometric, such as during WOT operations, or under boost, or trying to get a little extra fuel at idle for that lumpy cam; that's when you're out of luck with an NB. All they'll show you is "rich" or "lean", and that's not good enough: you need to know "how rich" or "how lean". That's where the WB comes in. It will accurately indicate mixture (or more accurately combustion product results of mixture, which is a little different) from, what, 10:1 to 18:1?

 

Of course, there are a couple of downsides to the WB. The first one is expense, they cost 7 or 8 times what an NB costs. The second is integration: If all you want to do is put a display on your dash to watch as you drive around (as I have done) it's not too bad, all you really need is a second bung in your exhaust and a couple of hours to install. But if you want to use one to control the fuel/air mixture directly, as a replacement for the NB, it's more involved. Some WB units offer an "NB simulator" output, that you can wire to your ECU and will give it the NB-like signal it needs, while simultaneously driving a display. The Mega-Squirt DIY ecu can (I believe) read the WB signal directly, and use if for all calculations. That's especially great, 'cause you can tune non-stoichiometric behavior in the ECU quickly and easily, no guesswork and little trial-and-error. In the GM ECU world, there are hacks of the EEPROMs that can do the same thing, though in more roundabout fashion.

[olie05]

The Mega-Squirt DIY ecu can (I believe) read the WB signal directly, and use if for all calculations. That's especially great, 'cause you can tune non-stoichiometric behavior in the ECU quickly and easily, no guesswork and little trial-and-error. In the GM ECU world, there are hacks of the EEPROMs that can do the same thing, though in more roundabout fashion.

 

I don't believe that the MS can do this yet, you still need a WB system (innovative lm-1, lc-1 etc) to do this. They are working on a PWC that will be able to hook up directly to the WB sensor, and do the calculations.

[JMortensen]

The thing to remember with an NB is that they are really a binary device - they indicate "hi/low", below stoichiometric or above stoichiometric, and that's about it. Nothing in between, not even perfect stoich. They just toggle up & down. If you're tuning for stoich, which is the vast majority of the time, you're fine - you'll see it oscillating back and forth over a narrow range of values more-or-less centered. That's good.

 

I don't think that's correct strotter. When you look at the voltage signal of an NB O2 that's being used on a typical FI system at idle it does fluctuate wildly. That's because the a/f ratios of most FI systems fluctuate at idle, cutting fuel and adding fuel constantly trying to hover around stoich. If you put a NB on a carbed car the numbers stay consistant and don't fluctuate at all. When I drove around for a while with the DMM in my car tuning my carbs, my goal was to get the DMM to show .8V all the way through a pull. .8V = 13:1 a/f ratio. The refresh rate of the DMM is the real limitation, so that's why I kept doing it over and over and over in an attempt to get it as steady as possible. IIRC the voltage also stops fluctuating in the FI systems when the system goes out of closed loop. I'm pretty bad with FI but that's what I seem to remember when I was trying to help a friend tune his turbo truck FI.

[Dan Juday]

...IIRC the voltage also stops fluctuating in the FI systems when the system goes out of closed loop. I'm pretty bad with FI but that's what I seem to remember when I was trying to help a friend tune his turbo truck FI.

 

In open loop the O2 sensor is taken out of the fuel control process. The voltage from it will still fluctuate with changes in the air/fuel ratio. It's just that the ECM doesn't see them anymore. The reason is that there are times (like Moby points up) when you don't want a 14.7/1 mixture. These times would be when the engine hasn't reached it's operating temperature or when it's at wide open throttle. These times you want the AFR richer so the ECM uses it's other sensors and it's tables to control fuel flow.

 

Just try disconnecting your O2 sensor and see what it does for fuel economy. (Disclaimer: With the current price of gas this is an expensive experiment!)

[JMortensen]

I think you missed my point Dan. The point is that the voltage fluctuates because of the FI in the closed loop mode. Strotter had said that the NB only indicates high or low, and that isn't the case, but it is what you would see when you look at the voltage signal for an older FI system in closed loop mode. The closed loop mode is constantly going rich then lean then rich then lean. Once you're out of closed loop it follows a map, and it won't fluctuate. It will change, but it won't go up and down unless the fuel map says to go up and down. That "hunting" is the effect of the closed loop system, not the effect of the O2 sensor itself.

[strotter]

What JM says is correct , the ECU swings the afr from rich to lean several times a second. There's a variety of reasons for this, not the least of which is to keep the catalytic converter happy, but it's also due to the nature of the voltage vs. lambda output of the sensor itself - a difference of a few tenths-of-a-points of mixture can make a huge difference in voltage output, so much so that you can't really tell where you are unless you're dead-nuts on stoich. Even without the ECU's oscillating behavior, such as when you have a sensor on a carburated car, there are very rapid changes to afr under normal driving conditions, happening in milliseconds, which I personally can't keep up with - but I have slow eyeballs, it's true. I've never had much luck with NBs on carbed cars, though I've tried it a couple of times. Once you get out of the stoich range and closed-loop, they aren't accurate enough to help you - if you need to tell the difference between 13.0 and 13.5 afr, you're looking a few mV change in output, which will be swamped by the normal variation and "noise" associated with combustion. On my Z, which is injected and run by a GM ECU, I was *shocked* at how inaccurate, or maybe "misleading" is the correct word, my NB was. Keep in mind, it's a pretty modified motor, but the same limitations apply to any NB. I discovered, for instance, that the difference of .3 afr, from about 13.8 to 13.5 during WOT on the WB, gave a *significant* improvement in performance, though the change was all but invisible on the NB.

 

There have been *years* of debate on this very issue over at thirdgen.org. The general consensus is that NBs are good for tuning near-stock motors under closed-loop conditions, but aren't accurate enough for WOT (or open-loop) tuning, or with wide-overlap cams, unless you're doing lots of plug-cuts (which I *hate* doing, standing on the side of the road with my ratchet, people slowing down to look, all sweaty, dressed in my working-on-the-car clothes, just not a pretty picture), in which case you don't really need the thing anyway.

 

BTW, something I learned from a buddy of mine a few years ago was to use an *analog* voltmeter with the NB - rapid fluctuations are damped out by the slower response of the needle, and you can mark points of interest on the face of the gauge with a sharpee type pen.

[mobythevan]

I don't believe that the MS can do this yet, you still need a WB system (innovative lm-1, lc-1 etc) to do this. They are working on a PWC that will be able to hook up directly to the WB sensor, and do the calculations.

 

To reiterate both points:

 

Megasquirt can use a wideband unit's output to correct fuel providing a closed loop mode(target AFR tables in MSnS_Extra), but Megasquirt cannot control wideband sensors directly so you still have to buy the controller and wideband sensor.

[JMortensen]

Well I'm sure my next post isn't going to convince strotter or anyone else who believes that NB isn't accurate, but I can say that I've one of 4 guys who has tuned a carbed car quite successfully with a NB. I will say this too, there is a difference in which brand O2 sensor is used, so there is a pretty good amount of variation between the sensors.

 

I started with a Bosch sensor because that's what I was told to do. One friend of mine bought the Autometer light gauge and wasn't having any luck at all. He then swapped the O2 sensor out with a Bosch sensor, and suddenly his gauge was accurate. In a semi-related way, this kind of bears the not all sensors are created equal: I have a friend who was getting 15 mpg in his Toyota 4x4 truck. He was running the cheapo generic O2. He switched to the $250 factory sensor and immediately got 23 mpg. Another time I saw a guy change the O2 in his Pontiac Grand Am. It didn't need to be changed, but he swapped it out thinking he'd prevent some problem down the line. He installed a Bosch sensor, and the car ran like dog ****. He swapped in a new AC Delco sensor after taking the car to the dealer and voila, car was fixed.

 

I know this only helps prove the inaccuracy of the NB in general, but those of us who have used the Bosch on a voltmeter or gauge in my group of racer friends have had very good results. I had mine tuned to where you could start at 3000 rpm and punch it and it would go all the way to redline and never go below .800 or above .850.

 

There was an old thread that showed what voltage equaled what a/f ratio, but it is gone now, it was by ZmeFly, and apparently when his little scam got discovered he went back and erased all of his posts.

[Dan Juday]

I think you missed my point Dan. ...

 

Yup, I missed it. I had my brain turned the other direction.

[JMortensen]

I was trying to say that if you watched the voltage on a NB O2 sensor with a DMM when it was in open loop the voltage wouldn't be up and down and up and down, because it does that as a function of how the closed loop system works.