Bruce
03-18-2004, 07:53 AM
There are two types of O2 sensors, Narrow Bands, and Wide Bands. The oem GN O2 sensor is a Narrow Band. Which means all it can effectively do, is tell you if the mixture is richer of leaner then Stoich., ie 14.7:1. Over the years, folks have come to think that 770-820+- mv., is a safe number for a GN to run at, when in WOT. Well, while for some cars it might be, it's not usiversally true. EGT, and Backpressure effect it's readings, so if things are just right you might being doing damage to you engine, when in fact you think your OK. A WB on the other hand, (in the case of the NTK's), operate at ~1,600dF, so unless your mount it in the wrong location, or ar running extremely high EGTs, EGT won't effect it's readings. They are somewhat Backpressure sensitive, so they're generally mounted downstream of the turbo to eliminate this error factor.
Now with a WB, you need to be more particular in how you mount it, to get max life, and accuracy from it. Like I've mentioned you don't want to mount it in too hot of location. You also don't want to mount it in too cool of location, since the heater controller is trying to maintain a 1,600dF if the exhuast gas has had time to cool off alot, then you might actually chill the sensor into not working, so rear bumper mountings, may or may not work, it would depend on the heater controller design. Then there is particlate matter, that can accumulate on the sensor, and ruin it. The combustion process generates alot of water vapor. Toss in some oil, carbon, lead, etc, and you can get kind of an oil ooze, in the exhuast. Usually it dry, and the flakes just go down the pipe, with no problem, but if things are kind of gooey, then this goo, can get into the sensor, and may or maynot get burnt off due to the sensor's oerating temp.. Cover the sensor with enough contaminates, and it will fail. So you also want the sensor somewhat down stream of the turbo so that the particulate matter has had some time to *dry out*. I mounted mine down by where the converter used to be, and even having broken a piston crown, and failed headgaskets, it's survived. It's just worked out well for me having it this far downstream. IMO, it's works out to the *best* average of what the sensor will have to deal with. You also want to mount the sensor in the top half of the pipe, ie between the 9 and 3 O'Clock positions, and with the sensor end, low, ie below horizontial. This will keep any moisture that condenses in the sensor from emersing the it. If the sensor (meaning the internal guts of the WB) is emersed in water, then when you apply power to and it begins to heat to 1,600dF, it will crack. And by the same token, you don't want to emurse it in water. In oem form they have to deal with water condensation in the exhaust system, so they are somewhat tolerant of water. So far in lots of test miles, I haven't seen a pattern develope of it mattering weither the WB is partically heated, unheated, or fully heated before starting the engine making a difference in sensor life, in normal street usage. If your going to be smothering it, with leaded fuel or other additives, then it wouldn't seem like a bad idea to let it come up to temp first, since the hot sensor is less *sticky* (from what I've heard). The time it takes for an NTK to warm up, again varies by controller, and it might be from less then 20 secs, to a full minute. Innovative Motorsports, has said the Bosch takes also little as 20, and the NTKs 30, but I have one that takes less then the alledged 30 sec min..
There are also two differenct WB sensors commonly being used today. One is Bosch and the other NTK. The Bosch is less expensive, but needs more temp compensation, then the NTK. It's sensing holes are tangent to the exhaust flow, where as the NTKs are somewhat of a flow thur design. If you were running some home brew fuel, shrouding the sensor might not be too bad of idea, and the Bosch seems to do this a little bit just in it's design. Thou using a spark plug anti-fouler, with a little trimming allows you to get the sensor out of the direct exhaust flow. The down side is that you introduce a little lag in it's response. The NTK design is again in use in current Honda, but there are differences in the heater part of the sensor, so you can't just just plug one of them in, where an older series one went, even thou the plugs are the same.
Now there are the normal caveats. With a large series turbo, exhaust back pressure *might* be several order of magnitude less then with a smaller one, so for some folks they might get away with mounting the sensor pre-turbo. There will always be the seemingly random failure of a sensor, but there are many reasons for one to fail, and they do suffer from the straw that broke the camels back syndrome. Silicone, can instantly poison one, and just as well only partically kill it, and 200 miles later it dies. And not all sensor safe adhesives are in fact sensor safe. When first breaking-in, a new engine, I run them for 200 miles without a O2 sensor, because I've had several new engines kill O2 in rather short order (NBs).
Now with a WB, you need to be more particular in how you mount it, to get max life, and accuracy from it. Like I've mentioned you don't want to mount it in too hot of location. You also don't want to mount it in too cool of location, since the heater controller is trying to maintain a 1,600dF if the exhuast gas has had time to cool off alot, then you might actually chill the sensor into not working, so rear bumper mountings, may or may not work, it would depend on the heater controller design. Then there is particlate matter, that can accumulate on the sensor, and ruin it. The combustion process generates alot of water vapor. Toss in some oil, carbon, lead, etc, and you can get kind of an oil ooze, in the exhuast. Usually it dry, and the flakes just go down the pipe, with no problem, but if things are kind of gooey, then this goo, can get into the sensor, and may or maynot get burnt off due to the sensor's oerating temp.. Cover the sensor with enough contaminates, and it will fail. So you also want the sensor somewhat down stream of the turbo so that the particulate matter has had some time to *dry out*. I mounted mine down by where the converter used to be, and even having broken a piston crown, and failed headgaskets, it's survived. It's just worked out well for me having it this far downstream. IMO, it's works out to the *best* average of what the sensor will have to deal with. You also want to mount the sensor in the top half of the pipe, ie between the 9 and 3 O'Clock positions, and with the sensor end, low, ie below horizontial. This will keep any moisture that condenses in the sensor from emersing the it. If the sensor (meaning the internal guts of the WB) is emersed in water, then when you apply power to and it begins to heat to 1,600dF, it will crack. And by the same token, you don't want to emurse it in water. In oem form they have to deal with water condensation in the exhaust system, so they are somewhat tolerant of water. So far in lots of test miles, I haven't seen a pattern develope of it mattering weither the WB is partically heated, unheated, or fully heated before starting the engine making a difference in sensor life, in normal street usage. If your going to be smothering it, with leaded fuel or other additives, then it wouldn't seem like a bad idea to let it come up to temp first, since the hot sensor is less *sticky* (from what I've heard). The time it takes for an NTK to warm up, again varies by controller, and it might be from less then 20 secs, to a full minute. Innovative Motorsports, has said the Bosch takes also little as 20, and the NTKs 30, but I have one that takes less then the alledged 30 sec min..
There are also two differenct WB sensors commonly being used today. One is Bosch and the other NTK. The Bosch is less expensive, but needs more temp compensation, then the NTK. It's sensing holes are tangent to the exhaust flow, where as the NTKs are somewhat of a flow thur design. If you were running some home brew fuel, shrouding the sensor might not be too bad of idea, and the Bosch seems to do this a little bit just in it's design. Thou using a spark plug anti-fouler, with a little trimming allows you to get the sensor out of the direct exhaust flow. The down side is that you introduce a little lag in it's response. The NTK design is again in use in current Honda, but there are differences in the heater part of the sensor, so you can't just just plug one of them in, where an older series one went, even thou the plugs are the same.
Now there are the normal caveats. With a large series turbo, exhaust back pressure *might* be several order of magnitude less then with a smaller one, so for some folks they might get away with mounting the sensor pre-turbo. There will always be the seemingly random failure of a sensor, but there are many reasons for one to fail, and they do suffer from the straw that broke the camels back syndrome. Silicone, can instantly poison one, and just as well only partically kill it, and 200 miles later it dies. And not all sensor safe adhesives are in fact sensor safe. When first breaking-in, a new engine, I run them for 200 miles without a O2 sensor, because I've had several new engines kill O2 in rather short order (NBs).