That's the point, ~0.45V isn't an acceptable voltage, in normal operation the ECU shouldn't see it. The sensor is always swung one way or the other if it's working. Now go away and get a degree in electronics
N/A systems go open loop at full throttle, sometimes before and forced induction cars can go open loop as low as 60% throttle.
No, it's just the way narrowband sensors work. It's also why people pay a lot more for the nice linear AFR/voltage relationship of a wideband, that's essentially the difference between the two, with a wideband you can read the voltage and know it's a certain AFR - narrowband can't be that specific.
I fully understand that stockiometric covers quite a large voltage range on a narrow band sensor but not that an ECU will disregard it as a bogus volatge.
On this page, under how does an O2 sensor work. I'm off to get a degree in English to help explain things more concisely EDIT: and learn to spell Edited by: martyn_16v
martyn I have discovered the answer. We are both right! Having been a right I've descovered that the sample voltage and the ECU's use of it depends on the construction of the probe. It appears Zirconia based probes generate their own voltage that the ECU just takes as an instantaneous voltage reading. Then their are Titania (sounds rude ) based probes which are fed by a digital reference voltage from the ECU. Having descovered the latter your wave distortion argument makes complete sense but at the same time explains why I disagreed, the Zirconia probe is the older design, pardon me for being old As for this reference page I would try to stay away from this kind of material. The tell-tale signs are:- auto repair manual 600 degrees F professional auto repairs All of these of course pointing to the fact the author is a yank and therefore couldn't find his arse with both hands and a map.
Totally agree about the yanks OK I concede defeat I thought I was talking about zirconia sensors. I know there's a very good reason why you can't just connect a pair of sensors together, but damned if I can find it now. Let's stop before this kind of spreads and we end up like the vortex
Just to clear up a few items posted; There are two types of transducer used for measuring the oxygen content of exhaust gas; Chemo-voltaic zirconia probe which produces an emf (voltage) porortional to the partial oxygen pressure. The transducer gives an output change of about 1v over the range from lean to the rich side of stoichiometry at 700K. Although the output is temp dependant, the measuring instrument is generally only looking for a rich/lean input. Don't have any info on the wide band version. In all probablility, the ECU injects 0.45v at the sensor so that no current flows most of the time, the sensors don't like supplying current. Then the ECU *probably* measures a V. small current flow in either direction which indicates rich/lean. This aproach has the disadvantage of too slow reponse time for high engine speeds. The info i have suggests they are OK upto 2200 oC. Chemo-resistive which is a tiny oxygen sensitive resistor made of titania, TiO2. The small size of the sensor gives a potentially faster response time than the zirconia probe. This approach has the disadvantage of enormous temperature coefficient (4% per oC), so some form of temp' compensation is needed. This type of transducer is happier operating at exhaust temperatures i.e. 650 - 750 oC, the performance drops off above 1000 oC, not sure about damage tho. With this type of transducer the ECU will *probably* have to inject a constant current, to be able to read a voltage proprtional to transducer resistance and hence O2. As for the original thread topic, it depends on the sensor type from the two above. I would imagine/guess the self heating types are less fussy about position. Reading only two cylinders is normal, the system does not have the ability to evaluate a signal of one cylinder playing up. In the absence of any thorough specs for the transducer you have, same distance from the head as the corado seems good, shame about the tape......
It`d be a lot easier to weld it to the manifold and let it sniff 1 cylinder, but does it expect a reading from a pair? does it matter that it`ll only get 1/2 the amount of combustion gasses through than the original fitment? The sensor doesn`t work at the throttle positions where it would act as insurance against an injector failing/ running lean so if it`s just idle and part throttle mixture contol I`ll take a chance on just 1 cylinder I think? Anyone any advice on Mig welding stainless? Will be using Argoshiled gas, maybe mild steel wire or is that a no no?
My guess is that it'll be fine on just one cylinder, the flow/quantity of gas is not important - imagine the difference in flow between tickover and redline WOT! I think temp is a much more important factor. MIG welding low grade (304) stainless is fine, i used the standard mig wire (copper colour ?) on some about 7 years back and have no probs yet. With pub bottle CO2 too. There is a reaction that takes place between the mild & stainless which means it's not that good in the REALLY long term, but for the mickey mouse low grade stainless, i think you'll be OK.
Although the ZrO2 technology works upto 2200 oC, Bosch ZrO2 lambda probes are only good for upto 850 oC, 600 oC is optimum.
Well there it is, picked number 4 because it was the easiest to weld to and I could get the sensor into a bit of decent air, wrapped the manifold while I was in. Piece of urine to take the manifild off with the head out of the way! Double gaskets knocked the cam timing out about a 3rd of a tooth, luckily I`d got a pegged vernier from BBM with the rest of the kit. Got the old loom half wat out but due to the pooty weather the car`s locked away `till next weekend now [:^(]
Sorry Keith I should've said, I've never found those stupid metal straps stay on very well, normally junk 'em for decent hose clips.
Will see how they go, a ball bearing is supposed to lock them off but don`t think it`s a good design, bent the end of the tabs back so they should be ok (it`s never ever ever coming of again )
