Theory v. Practice: Wideband sensor location

Dan_Gyoba

Turbo Swapper
Aug 9, 2007
1,836
0
0
Alberta
www.gyoba.com
Okay, I see lots of pictures of downpipes, turbo elbows and other things with a wideband sensor right next to the stock sensor location.

I've read a number of documents that say not to place the sensor right next to the turbo outlet, place more than 15 degrees from horizontal and 10 degrees from vertical, and that failure to do so will cause early failure of the O2 sensor.

So, there's some discontinuity there. The parts manufacturers, are likely only doing what's easiest for them, so having the O2 sensor right next to the stock location is obviously easiest.

I'm more curious about what early failure means, in real terms.

The sensors appear to be rated for about 160,000 km. That's kind of short for a sensor, considering that's the expected lifespan of a typical platinum tipped spark plug, and I kind of expect spark plugs to be wear items, and sensors not to be.

So, if I were to get the wideband bung welded downstream in the exhaust, about 1m from the turbo outlet, but still protected from damage from whatever would be under the car. Route the wiring away from excess heat and strain, and do everything perfect, I could expect the sensor to last about as long as a set of OEM sparkplugs.

What if I don't? If I were to pop the wideband in that bung that came with my elbow/downpipe about 4" from the turbo. It's still more than 10 degrees from vertical, so at least that's still in the good range. Tech notes seem to suggest that recessing the sensor a little back from the exhaust stream will help prolong its life in that situation, but I don't know how much, or even how much it's life expectancy will suffer from being out there in the first place.

So... how much will it suffer? Would that elbow location mean instant death for the sensor the first time I go WOT at 16 PSI? Would it mean that life expectancy goes down to half? I mean, I assume that reasonably low EGTs (Where my car will spend most of its life, but by no means all) will hurt the sensor less, since higher EGTs seem to hurt it more. They seem to be so picky about other things, too. (If it's that sensitive to silicone, maybe I won't install the wideband at all until after engine break-in, just in case there's anything from the FIPG at the head gasket location, or somewhere else in the engine where it might hit the exhaust. (The stuff I used is labelled as "sensor safe" so maybe it's okay?)

Or maybe I just budget to replace a $78 sensor every couple of years. Do it when I flush coolant or something.

Lots of people seem to be running wideband sensors close to the turbo outlets, many with upgraded turbos, running elevated boost. How long DO they last?
 

Dan_Gyoba

Turbo Swapper
Aug 9, 2007
1,836
0
0
Alberta
www.gyoba.com
Information on life expectancy, and "ideal location" taken from things said here, and this document

Interestingly enough, Bosch (The maker of the sensors) seems more interested in telling me where I can find the OEM sensor, and how to replace it with one of theirs, in the same location.

At first, I thought that I'd just put the wideband in the factory location, use the simulated output and ditch the factory sensor altogether. I still kind of like that, except I'm not too fond of the idea of a code 21, and I do plan on placing the sensor down lower. not sure how I'm going to run wire down there safely, but I'll figure something out.
 

Poodles

I play with fire
Jul 22, 2006
16,757
0
0
42
Fort Worth, TX
100,000 miles is normal and intended (as it's when warranties run out on modern cars...)

You're going to cook that sensor that close to the turbo, but if you're using a cheapish sensor that's not a huge issue (hell, some of the OEM wideband O2's are cheaper than a replacement for our narrowband...)

You are correct in that the parts manufacturers put them close to the stock position because it's easier and they don't really have to think.
 

Dan_Gyoba

Turbo Swapper
Aug 9, 2007
1,836
0
0
Alberta
www.gyoba.com
Well, more reading, and my conclusion is that lower is indeed better. I'm going to aim to have it near the #1 crossmember. I really like the idea of running the wires through the shifter hole, but up by the starter with the reverse lights and ABS speed sensor wires should be a reasonably safe location, too. I'll have to see how much length I have with the wire harness. Kind of dissappointed in the aftermarket manufacturers putting wideband bungs in the turbo elbows.

I was looking for the stainless bung that I bought... And I can't find it :mad: I was sure that I had it with my targa seals. I bet that I put it "somewhere safe."

I hope I can find one locally. Maybe Mopac will have one. It'll probably cost me 3 times what I paid from Driftmotion though. :cry:

For the cost of the sensor, it seems that the Bosch wideband is an OEM replacement for a lot of makes, and it is cheaper than an OEM narrowband for the Supra. I'll basically figure on replacing it when I replace spark plugs. I don't expect to be burning oil any time soon with a freshly rebuilt block, but who knows with turbo seals and the like.

The documentation says that the sensor isn't going to like silicon much. I see "sensor safe" on the silicone products that I use. Bullshit, or true? Is Toyota FIPG safe, or should I break in my engine without the wideband installed?
 

IJ.

Grumpy Old Man
Mar 30, 2005
38,728
0
0
61
I come from a land down under
I think the Silicone Safe thing refers to using it as a thread sealant Dan, (don't understand why you would need to though) I use an Antiseize on mine and the washer seals fine.
 

suprarx7nut

YotaMD.com author
Nov 10, 2006
3,811
1
38
Arizona
www.supramania.com
I read the distance warnings on my AEM, but I have a Raptor Racing Exhaust, fully ceramic coated. The wideband location is in the elbow. I wasn't about to have that ground down and welded again. More than one year and 5k miles or so later and it all seems fine. I'd be curious to see how the accuracy declines with it that close.

I'm my install right now, the wideband isn't used for tuning or for any kind of informative signal to a piggyback so accuracy doesn't mean shit. It's just an indicator for me that the engine isn't going horribly lean or rich.

In the future, I'll have it run on a dyno and compare the signal from the shop's sensor to mine and see how much of a difference it really makes.
 

Dan_Gyoba

Turbo Swapper
Aug 9, 2007
1,836
0
0
Alberta
www.gyoba.com
@IJ: I don't plan on using any kind of sealant on the sensor, there are just dire warnings about any silicone used as a sealant anywhere in the intake/exhaust stream somehow contaminating the sensor. Anti-seize is probably a good idea though.

I know that household silicone outgasses some pretty nasty stuff as it cures. I don't know if that issue applies to "sensor safe" silicone as well. Well, everything that I've done so far has had a good long time to cure already, too.

@suprarx7nut: Accuracy is one thing, so long as it is translateable, I'm fine. I'm not planning on doing any super-fine resolution tuning based on it, mostly it's to get an idea. I'm more concerned about the coal-miner's canary data of AFR and EGT for keeping my engine safe. I don't really see myself as the type to lean out the mixture to squeeze a few more ponies out of an engine. (Though I'm not interested in washing the oil off of the cylinder bores with gasoline, either.)

I'm still thinking that I want to be using the Zeitronix programmable output to the ECU. I have my field harness set up so that I can connect either the stock O2 sensor, or the Zeitronix programmable output to the ECU. If I encounter any problems, I'll go back to stock.
 

IJ.

Grumpy Old Man
Mar 30, 2005
38,728
0
0
61
I come from a land down under
Yep think it's the Gases while it's curing that are the issue Dan, very acidic.

My understanding of the NB Emulation is it's quite slow and you can end up in a feedback loop while it's playing catch up...
 

GrimJack

Administrator
Dec 31, 1969
12,377
3
38
56
Richmond, BC, Canada
idriders.com
Household silicone outgasses some nasty crud that will trash oxygen sensors. The sensor safe stuff is supposed to be fine - I've used it around my sensors, and never had a problem.

I'm running the AFX wideband, with the sensor mounted in the elbow. I've been running it as a daily driver, with 550+ rwhp, for years, and it's still fine. I suspect when they talk about reduced life, they are talking small percentages.

Also, when mine goes, replacing it is easy as hell. I'd rather have something easy to work on that I have to replace more often, than something that's a huge PITA that I only have to work on half as often.
 

Poodles

I play with fire
Jul 22, 2006
16,757
0
0
42
Fort Worth, TX
IJ.;1953628 said:
Could be worse... my current toy has 6 Sensors, 2x WB and 4x NB's ;)

4 cats?

Also, the O2's that monitor the catalytic converters' function are quite easy to eliminate :p They serve no purpose other than to flag a code and a MIL for emissions reasons anyway...

Honestly, I've never seen a new O2 sensor not come with antisieze on the threads (all protected by a plastic cap). If there isn't some on there, definitely use it.
 

IJ.

Grumpy Old Man
Mar 30, 2005
38,728
0
0
61
I come from a land down under
Poodles;1953631 said:
4 cats?

Also, the O2's that monitor the catalytic converters' function are quite easy to eliminate :p They serve no purpose other than to flag a code and a MIL for emissions reasons anyway...

Honestly, I've never seen a new O2 sensor not come with antisieze on the threads (all protected by a plastic cap). If there isn't some on there, definitely use it.
2 CATS but the car runs 2x pre and 2x post, as you say easy enough to delete and shut off the codes but it's an instant fail if you get pulled over here...

My antiseize comment was more for used sensors than new, as you noted they all come pregooped, mine are in and out like a $2 Hooker so I keep some AS handy as the $2 Hooker will tell you some sort of lube is essential :D
 

Dan_Gyoba

Turbo Swapper
Aug 9, 2007
1,836
0
0
Alberta
www.gyoba.com
Ah, so that's what the plastic cap is for. I haven't removed it yet. I hope never to be able to make an informed opinion on a $2 hooker. :p

I'll have to see what "slow" means. More research. Is it the wideband that's slow, or the switching circuit for the simulated output? I can think of reasons why either might be slow, but the circuit should be able to compensate, even if artificially crosscutting when the sensor hasn't told it to. It's already faking it, after all. It would be kind of nice to be able to stretch gas mileage by running a tad lean in closed loop operation, but it's not worth causing problems. OBDII isn't going to cause me any problems in the near future on any of my "fun" vehicles, as they're all pre-1996, only the wife's DD has it (And it likes to light that MIL for the cat O2 sensor, which was forcefully removed by a pothole... Along with the cat and about 3' of exhaust pipe. Nice roads we get here...)

Thanks for the reasoned information!
 

IJ.

Grumpy Old Man
Mar 30, 2005
38,728
0
0
61
I come from a land down under
SeeUSmile;1953638 said:
I was under the thought that having the sensors close to the manifold/turbine is to get it to its operating temperature quicker during cold starts.
Correct, good for emissions and good for Sensor manufacturers as it reduces sensor life ;)