Recommended spot to mount Wide Band O2 sensor?

[Mr Starlet]

Hi all

Mounted my Innovate wide band O2 sensor on my NA a few weeks back (ready for the ECU) but after a bit of a read on life span affected by heat, I'm considering relocating the sensor to a cooler spot but not sure where and how much cooler to worth the bother. The sensor needs to sit somewhere on the extractor/header. Innovate says it needs to be before the catalytic converter, must not be after, supposedly the lag in the air flow and chemical reactions at the cat will affect the reading, but then tuning shops reads from the exhaust tip, is this because it's easier for them?

So you would think further away from the engine but apparently the cat itself run pretty hot by nature as it burns wasted fuel and harmful gases. Does it makes much difference at all if you mount it say 2-3 foot away from the current position shown in the pic below? or is the heat pretty uniform across the entire header away so there's no difference?

Also the stock header has the sensor right at the collector (where all 4 pipes join) whereas my aftermarket Tanabe (supposedly reputable) has the sensor on the output of the first cylinder, good/bad idea?
good that it see less heat as gas only pass through once per cycle?
bad that it only monitor the first cylinder and not the combine engine as a whole?

What do you think?

Cheers
Minh

[Steampunk]

Ideally it would be good to mount it on the secondaries section, for the reason you mentioned, but the only problem fitting it way down low is it would be difficult to get to. Remember the LC-1 likes to have regular "free-air" calibration.

Thinking about it, the optimal place may be on the top of the collector section cause you may be able to get a spanner on it whenever you need to.

I recommend you make and install a heat-sink as per Innovate's suggestion.

If you look at it in another, more optimistic, light Minh, this can be just a short-term problem, ie. once you have found a good fuel and ignition map, you can remove the Lambda sensor altogether, plug up the hole, and turn your MS into continuous open-loop mode.

As for dyno shops sticking an 02 sensor at the tail-pipe, yes, I think it's more likely for ease.

[Steampunk]

http://www.miataturbo.net/

[Mr Starlet]

True, I guess once it's tuned you'd run open-loop.

[ampz]

I mounted mine at the 12 oclock position underneath the word 'around' on your picture.
Free air calibration takes 90 seconds with the sensor out of the exaust stream (it needs ingition pulses in free air to recalibrate).

The controller module sits on the other side of the gearbox, zip tied to the ppf and i routed the wires with the reverse light switch wires. To enter the cabin, I pushed it through the speedo grommet.

Why so far down the exhaust? the first bits usually glow red

[Steampunk]

The controller module sits on the other side of the gearbox, zip tied to the ppf

[Juffa]

Mine is in the same position as AMPZ stated, and also at the 12 oclock high position. This was recommended to me by the manufacture ( www.wb02.com ). From memory they said it needed to be about 2 feet from the exhaust valves, and before the cat. By being at 12.00 it reduces any issues due to condensation. I have enough cable so that I could run it into the cabin via the engine bay. My controller is attached to the floor behind the passenger seat.

As for removing it after the tuning...that is true...but I like to have it so I can keep an eye on the AFR. I have already paid for it....might as well use it.

J

[Mr Starlet]

I mounted mine at the 12 oclock position underneath the word 'around' on your picture.
Free air calibration takes 90 seconds with the sensor out of the exaust stream (it needs ingition pulses in free air to recalibrate).

The controller module sits on the other side of the gearbox, zip tied to the ppf and i routed the wires with the reverse light switch wires. To enter the cabin, I pushed it through the speedo grommet.

Why so far down the exhaust? the first bits usually glow red

Does mounting the sensor so far down introduce lag making it difficult to tune? or is it negligible?
Also, from experience how often should you need to do a free air calibration? I've done mine once when I first install it but not sure how often you need to do it, if it needs to be done regularly then I might have to sacrifice some longevity for easy access and mount close up where I can get access from the engine bay.

Cheers
M

[Mr Starlet]

On 2nd thought, would it be a bad idea to mount the bung on that second piece (see photo), after the main header part, and just before the cat?
So much easier to remove than having to remove the main header part, my only concern is that the cat gets really hot (hotter than the rest of the header?) and heat is radiated back to the sensor.

[GS]

I used this guide on the FM website when installing my AEM Wideband. Obviously its focused on FI applications, but it may help. FM dont seem concerned about the proximity to the cat, but then again the approach with FI engines is to move the sensor as far away from the engine as possible:

http://www.flyinmiata.com/support/instructions/ecu/AEM_WBO2_sensor_gauge_type.pdf

[ampz]

free air calibration on rich running cars, usually once every 6 months.
I don't see it really difficult to jack up the car to remove the sensor, but opinions may differ.

The cat converter does not get as hot as the headers, but i usually do my calibration on a cold engine.

[Mr Starlet]

Free air calibration takes 90 seconds with the sensor out of the exaust stream (it needs ingition pulses in free air to recalibrate).

So you calibrate yours with the engine running? instruction says to just have the ignition switch to on.

[timk]

You'll still want at least a narrowband O2 sensor for cruise so it can run closed loop there. Do our bit to save the trees and be closer to legal.

[ampz]

For some reason (it may be the way i wired it up) my lc1 needs ignition pulses to recalibrate.
But there isn't mention in documentation that the engine needs running.

For the couple of minutes that it takes to calibrate i don't mind the noise (it isn't very loud at all). I've tried just leaving 'reds' on but the sensor locks up.

The lc1 has the advantage of programable outputs, I've been able to display a wideband o2 signal on a narrowband sensor (led sweeping bargraph) which makes reading the instant tune a breeze.

There's a delay in sensor reaction as well as in the ecu to determine the tuning latency (mines set at 1/2 second delay, it tunes in a narrower window than when it was instant)