i recently took a look at this thread at nasioc and wondered if a wideband sensor absolutely should be used to tune our cars.

http://forums.nasioc.com/forums/show....php?t=1922848

bosch lists this as a replacement sensor for our front o2:

LSU “Broadband” Universal Sensor - as its name suggests, the
broadband oxygen sensor has the ability to not only indicate an
air/fuel ratio of Lambda = 1 but can measure as low as Lambda = 0.7
up to infinity

there are successful usage reports of Bosch universal LSU planar sensor on evos.
this leads me to thought that OEM mitsubishi sensor is similar in its construction and probably measurement.

did anyone do any kind of research with this or everyone blindly installs commercially available wideband sensors (even though Bosch claims the same 0.7 < lambda < infinity measurement for them)?

thanks.

 

Yes, a wideband should be used to get optimum power & the most accurate readings.

It's a great tool to have to make sure you're not going lean when running a bad batch or E85 either.

I'd invest in a couple hundred dollar wideband than a couple grand on a blown motor...

 

have you read the nasioc thread and have any actual data to prove your point? i have my couple hundred bucks invested in LC1, it lies in the box and before that goes on my car (for a second time) i want to hear someone who really knows lambda sensors, not the one who blindly copies common knowledge. thank you.

 

If it's listed as a replacement, it ought to work with no modifications to the computer. What does y'alls front O2 currently read? I know my front one doesn't read anything but like between 14.6 and 14.8 or whatever, basically just "too lean"/"too rich".

Placement would be an issue, even if your front is a "broad" or "wide" band sensor. Also, maybe the sensor can detect the range, but the ECU isn't set to interpret it any more than a narrow band. From that Naisoc forum, it sounds like it can be fairly innaccurate.

 

I'm tempted to say "if you have to ask, then you shouldn't be tuning your car," but that wouldn't be very useful since something like it has been said many times and, so, I'd just be cutting and pasting.

Here's why you want one. The narrow band sensors on the car will not tell you where you are when you're any distance from stoich. Estimates of AFR from airflow and injector pulse width are also inaccurate, plus the point of a wideband is to have redundant, as well as more accurate information. And you want to know exactly where you are (when in open loop) because the difference between 10 and 11 can be several horsepower and the difference between 11 and 12 can be several thousand dollars.

 

subie guys problem is that they have o2 sensor mounted before the turbo, not after. we have it right after the turbo from the factory, so that's on our side. i can tell you from my own experience that the best available wideband sensor is as accurate as the install location. if you put it in o2 housing, it's one story, if it's put near the end of the exhaust pipe - then it's the other (provided it's cat-free). there was a commercial wideband and oem sensor comparison plot, installed at the same place on nasioc, see post #26 of the link above. to me it looks like the sensors give the same output.

as to ECU which isn't set up for the task, i just don't buy it. subarus can log their sensor fine, i don't believe our ECU is set up in a different fashion. it can easily be either a proper MUT request or a scaling issue. i just think noone gave a proper attention to it.

 

Why do you consider measuring AFR later rather than sooner to be an advantage? Would you also want to place your EGT sensor after the turbo, rather than before? As to logging, that is a completely different issue. Logging low-resolution and lousy data is still logging low-resolution and lousy data. The issue - I thought - was whether you wanted a wide-band for tuning, not whether you wanted to log the data.

 

i only think that all these sensors can be inaccurate at the same degree.
i also think that our front o2 sensor can be wide band from the factory and not a narrowband as many people believe, just like in subaru world. subies have similar 4 wire planar sensors.

 

You are free to think whatever you wish, but the stock O2 sensor on all Evos is a narrow-band.

You are also free to believe that a narrow-band is as accurate as a wide-band, but that won't make that true, either.

 

the reason i started this topic is that i want some solid proof, not some piece of common knowledge. something really other than "you need a wideband to tune". i already knew that actually.

do you know how exactly our sensor is different from that of say MY04 wrx sti? they're both four-wire ZrO2 planar sensors. their can be used to tune the car (ok, if re-located to the right place), and ours not? please explain your point.

do you know the limits of accuracy of stock sensor? guys at nasioc did a comparison study, looks solid to me, if i was at their place, i would use that OEM sensor to tune without second thought. can you show me some similar test conducted with evos, in case i missed it?

also, if all else fails (which i doubt, because i still think the sensors are similar), can sti's wideband or replacement be installed on our cars? should be trivial, both are 4 wire, the pinout is known and exactly like ours.

 

All I know about Scoobies is what I can see in my rear-view mirror, so I can't help you on that. (Snide comments aside, I'm impressed that Scooby went with a wideband. Good for them. But, then again, Scooby has never been as shy about warranty work as has been Mitsubishi. No surprise that they went wide-band before Mitsu did.)

I have no idea how much you know, so I'll just add this and then walk away. The key difference between narrow- and wide-band sensors is the shape of the graph relating O2 concentration to output voltage. The curve for a narrow-band is close to a step function. It can tell you whether you are above or below stoich and that's about it. Great for setting the LTFTs for closed loop; worse than useless for tuning the open-loop maps. A wide-band, in contrast, has a nice shallow slope, such that it can give you a near-precise value (depending on the controller) from down around 9:1 to 100% "unused" air. And that's what you need for tuning.

 

our sensor and sensor on most cars a 0-1v making them narrow bands. wide bands are based on a 0-5v scale... some cars do use a wideband o2 like vw and audi and alot of other vag cars. there oem sensor is the same as the lc1 sensor. but our sensors are not for telling us how lean or rich we are on a afr scale but rather "hey your richer than stoich , hey now your leaner than stoich, hey your stoich...no now lean " thats the conversation btween the ecu and a narrow band 02 sensor (0-1volts)

 

i think it's the controller manufacturers who output the signal in 0-5 volts.

perhaps you will notice that subaru folks log current (in mA's) and not voltage in order to get a lambda value out of it. the question remains, do we know if an o2 sensor current can be logged off the stock ECU or not? if we can, we may go without aftermarket wideband controllers.

did anyone bother to check?

 

the front o2 sensor on our cars that is located on the o2 housing is narrowband and not wideband and no it can not be ran in wideband without the actual aftermarket controller to convert the wideband o2 sensor voltage signals into wideband readings.

the front o2 sensor is for closed loop operation on our cars and for fuel trims only.. while the rear o2 sensor directly behind the catalytic converter is there just to make sure the cat is still working efficiently.

Now there are mods we can do to our cars to run them in wideband only and never use the narrowband sensor again.. but it still requires an aftermarket wideband controller. our cars are not equip with wideband controllers and the ecu has no chips on it to know how to convert a wideband signal, therefore no its not possible.

the further away the sensor is located the leaner the AFR will be read.. so yes it is very important to have a proper measurement between the wideband o2 sensor .. and it is not comparable to a egt probe that needs to be right there to read temperatures in the exhaust manifold runners. also it effects the lifespan of the sensor as well depending where its located.

 

I guess I should have added a snark tag to that one.

Even with your post, however, $5 says we're not done yet.