cars with wide band o2 sensors for my lean burn conversion

[saand]

Hi All,
I am looking at changing my mazda 626 to run lean. This will involve getting a wide band o2 sensor, somehow fixing it to the exhaust system and making a circuit to convert the wide band signal to the desired signal for my ECU.

Everything on the list is easy for me except getting the actual wide band O2 sensor. I am wondering if anyone knows a list of cars (sold in australia) that have wide band o2 sensors. I have a wrecker close to me so i hope i can go there and just grab a sensor off one of the wrecked cars with little cost.

Also any help in how to identify a wide band sensor would be good.

Am i looking for only 5 wire sensors or are there different variants.

Thanks

[comptiger5000]

If you have trouble finding one, you could probably buy an aftermarket wideband, such as those used for tuning and datalogging in performance applications, and use that.

[saand]

thanks for the suggestion comptiger5000, i have considered that option but i dont see any of those costing less than $100 which makes the conversion pointless because i would never be able to make that back in savings.
Having said that if anyone knows of a source for cheap wideband sensors (less than $50) please post the info.
If i can do it cheaply then the method will hopefully be useful to others on the forum as well.

[t vago]

Go get a Bosch 17014 oxygen sensor.

[kamesama980]

Keep in mind it's not just the sensor, it's the controller too. Widebands don't output a voltage like narrowbands, they need a set of circuits to provide some voltage and current and that knows how to read the responses of the sensor. Part of the controller is also converting it into a signal for your ECU...the controller is actually the expensive part, replacement sensors for a wideband setup are in the $50-75 range.

What sort of ECU are you using? modified stock or aftermarket? that will also affect your options.

nitty gritty from memory of class last semester:
there are 2 types of wideband sensors, Critical Current and Ion Pump. critical current is a 4 wire (2 sensor 2 heater) and varies about .2V but the voltage isn't what's read. the ECU runs bias voltages on both lines to end up at 0V at the sensor at stoich. when the mixture goes off stoich, the voltage changes very little but what the ECU measures is the mA of current flowing and which direction to determine rich/lean and how much. These are actually made of Zirconia Dioxide like most narrowbands but they're physically different in the sensor. Exact voltages and mA readings will vary by sensor and car they're in.
Ion pump is a 5 or 7 wire (2 heater, 3 sensor, 2 trimming resister (may be in the plug or even the ECU itself)) and most expensive but also more precise. It's a completely different design. Again, you have bias voltages and mA is measurd to determine mixture but it's flip-flopped from critical current.

Either way it's not a simple circuit you're talking about building, unless you have oodles of time and like to figure out how to do things then troubleshoot and tweak for ages I'd just get the aftermarket wideband from race shops. you can get them for under 150 w/ gauge on ebay. That will be the sensor, controller (that can output a variety of signals for your ECU) and gauge.

[saand]

kamesama980 thanks for the info,
regarding the wide band o2 sensor circuit, my ECU is a stock ECU so it will only take in the narrow band sensor input however i am an electronic engineer so i will not have any trouble making up a circuit to deal with the wide band sensor, i planned to set the air fuel ratio on the circuit i make so i can set it lean or stoic then the ECU will just see the normal sensor input and attempt to compensate.

I went to the wreckers yesterday and i didn't find any wide band sensors at least of the 5 or 6 wire configuration. I saw plenty of 4 wire sensors but without knowing the info on them they could and would likely be just a standard sensor with a heater.

I am trying not to spend $50 on a o2 sensor because if it saves maybe 5% it will take several years to get the payback on that.

So i am going to experiment with how lean i can make my car using the narrow band sensor, i am going to make a circuit that will make the car try to control the air fuel ratio so the o2 sensor is outputting say 0.05v rather than 0.45v. I unfortunately cant find a datasheet on the o2 sensor i have or even a generic that has voltage output to air fuel ratio but i figure ill see the results on my fuel meter and see how it goes.

[t vago]

Originally Posted by saand

I am trying not to spend $50 on a o2 sensor because if it saves maybe 5% it will take several years to get the payback on that.

Er... actually, you could recoup that in a matter of months or even weeks. It depends on how much you drive, and the cost of gasoline.

Originally Posted by saand

So i am going to experiment with how lean i can make my car using the narrow band sensor, i am going to make a circuit that will make the car try to control the air fuel ratio so the o2 sensor is outputting say 0.05v rather than 0.45v. I unfortunately cant find a datasheet on the o2 sensor i have or even a generic that has voltage output to air fuel ratio but i figure ill see the results on my fuel meter and see how it goes.

Er... That's pretty scary territory there. How would you compensate for the effects of E05 or E10 or E15, given that they have differing stoichiometric AFRs, and given that normal heated narrowband O2 sensors are not at all accurate outside a lambda of 1? You might end up frying some engine internals if you're not careful.

[saand]

i fill up maybe once a month or less and petrol here is about 1.15 a liter so the pay back is a while.

regarding the effects of e05,e10,e15 i do not need to worry as i only fill up with standard fuel not blended (hopefully not) with ethanol.

the standard o2 sensors are not accurate outside of stoic but im not after accuracy just better fuel economy, doesn't matter how much leaner i am as long as the car still works and the fuel economy gets better.
I can check the fuel economy very accurately at idle and although the sensors are not accurate away from stoic I am hoping that the sensors are at least consistent.

yep i could very easily fry something in the ECU, but i dont plan to send a voltage over .9v to the ECU and have it current limited by a fairly conservative resistance so should not have too much of an issue.

either way whatever happens it will be fun to try and maybe ill be able to report an easy method for others to make their cars leaner.

[t vago]

Originally Posted by saand

yep i could very easily fry something in the ECU, but i dont plan to send a voltage over .9v to the ECU and have it current limited by a fairly conservative resistance so should not have too much of an issue.

While it's a good idea to buffer your signal so that you don't fry your ECU, I think you may still want to reconsider this idea. When I mentioned that you might fry engine internals, I meant pistons, valves, and rings.

Once you start operating lean, your engine will start to experience detonation. This is explosive quick (supersonic) combustion of the air-fuel mix, which is in contrast to the slow (subsonic) combustion of air-fuel mix at stoichiometric. The difference is that detonation causes a shock wave to propagate from the point where the detonation started. You can actually hear the shock waves hitting combustion chamber surfaces, as a knocking noise originating within the engine (hence, it's commonly called engine knock).

At relatively low loads, this is usually not a problem. However, under moderate or heavy loading, detonation may scour away combustion chambers, may shatter piston crowns, or may jam piston rings into their grooves. In addition, detonation may cause combustion chamber surfaces to become really hot locally, which will lead to pre-ignition.

Pre-ignition is when air-fuel mix combusts prior to when it's supposed to when the spark plug fires. As the cylinder travels upward on its compression stroke, the air-fuel mixture heats up. If this mixture is burnt while still in the compression stroke, combustion chamber surfaces will quickly rise in temperature to the point where piston crowns may melt, valves may melt or get burnt, or you might even have a nice engine backfire, where the air-fuel mix will actually combust while still in the intake manifold.

This is why you want to be very careful when running lean.

[cujet]

While I have not verified it, the race car guys purchase VW widebands for around 30 bucks.