was just about to post that link, props

 

And thank YOU Scott for pointing me in the right direction. Zeitronix FTW!

 

Anyone use STACK? I purchased it for my track car so I can keep an eye on things. I switched it over to the EVO again for referance but I know there are some usefull features. Only gripe is that it only reads to 16.
Now that I don't have anything in the R, I'm in the market for something to replace it. Should I pick up the LC1 ?

 

good info

 

well that answers my question. ha.

 

Innovate sponsored the Mike Kojima test and it was conducted using Innovates testing methods including the most technical method of spraying a gas from a bottle and holding the sensor near the nozzle.



Its a ridiculous article sponsored by Innovate who is in cahoots with the writer of the article Mike Kojima.

Their testing methods lacked proper calibration methods and were geared to make the results fall in Innovates favor.

Look at this mess:



For a company to have so many failures of their own product on their own forum, it should spell it out for anyone thinking about buying one.

Any good test would test the sensors one at a time and not in such a manner that different sensors are in different parts of the exhaust tract.

The test is flawed at best and greased at the worst.

 

Welcome to 2007. This is exactly the same "shoot at "article released to Ford Muscle 3 years ago. Total BS test performed by Innovate employs using Innovate equipment. Great if you like infomercials.

 

does the zeitronix WB have a 52mm type afr gauge?

 

Two different ones. ZR-1 for AFR and Lambda only:
http://www.zeitronix.com/Products/ZR-1/ZR-1.htm

and ZR-2 mulitgauge for AFR, Lambda, Boost/Vacuum, EGT, Oil, Water Air Intake Temperature, Fuel, Oil Pressure:
http://www.zeitronix.com/Products/ZR-2/ZR-2.htm

 

Zeitronix, do you guys plan on fixing the problems that netted you 2nd to last place? I would interested to see your revised version.

 

What problem? See 2 posts above.

 

What problems? This is a three year old test. Innovate sponsored the article. The writer is trying to get a website off the ground and propel his career so he took the same pics from three years ago's mag article and adapted the text to suit his site, pretending that the test was just done recently.

Its the same test Innovate paid him to do using their testing methods to try and sway the results in their favor.

Fast forward to the present and drop the greasy palm financial bias and you will find Zeitronix to be the leader in independent non-bias reviews such as the ones conducted in the most recent SEMA SHOW last quarter.

 

All the conditions for the testing were disclosed. It's up to the reader to consider if the test methodology is poor or not. It is also stated the test was from a while ago. Also, I do not agree that testing one sensor at a time would be best as it'd be very difficult to replicate the test conditions, namely the transient responses.

Makes perfect sense to test them at the same time to me. It eliminates the many variables related to run conditions, especially the transients.

Furthermore, AEM is an advertiser on the site and their WB worked for crap. Also, the picture you linked on using the calibration gasses on the sensor, you left out the caption: "We flooded the sensor tip with two different lab calibration gasses to do the first part of the testing. If this seems hokey, we tested this against the ultra accurate dyno sensor and got the exact readings that we were supposed to get." So, they also baselined the dyno sensor using this method and it all checked out.

If you think you have a better testing method, I'd like to hear it. Better yet, do the test yourself.

 

I was just having some fun

My father and I built a DIY wideband a few years back as he was a Prof. of Electrical Engineering and understood that stuff. We made many changes to the DIY design to make it more robust and improve accuracy outside of stoich, but the point is that I somewhat understand how the wideband controllers work.

I myself have the PLX device:
The PLX M300 included no logging software, and exhibited accuracy at +/1 AFR (the worst tested). The PLX is not calibratable.

The funny thing is the sensor is calibrated from the factory using the resistor, but I guess they assume an autocal function since all sensors tested were the same LSU4. The fact that all sensors were placed in the exhaust next to one another doesn't matter as that was only used to put miles onto the sensor using race gas.

The comparison was done at room temperature using the calibrated gas and I believe this could potentially create a problem for units lacking an autocal function. If I am not mistaken(not sure) the sensor accuracy is also a function of its temperature, at least that is what I remember for the 5-wire NTK unit, so if the unit doesn't have an auto-calibration function it will potentially read inaccurately if not at the proper temp. I tried to find the AFR accuracy as a function of temp, but could not locate such a graph.

I would also point out the PLX does measure free air and indicates such on the display so you can tell if the sensor is bad. All you do is turn the key to on and then wait for the sensor to reach max temp and it should then read "free air." You can also tell when they start to go bad as they no longer oscillate values while driving.

I would also point out that the all LSU4 sensors are calibrated from the factory in a lab environment. Therefore, if you have an Innovate system that auto calibrates it is actually making changes to the resistance values as set from the factory. Essentially, if you need to autocalibrate outside of the resistor set from the factory you are then compensating for a sensor going bad. I guess my point is that do you really want the wideband unit to override the factory calibration settings?

 

O2 sensors are also sensitive to exhaust back pressure, along with temperature. The induced inaccuracy from those two parameters is on the same order of magnitude as the error between the sensors seen.

http://wbo2.com/lsu/lsu4.htm
Look at that, 20% error due to exhaust back pressure not that unrealistic...

Oh, and then the fact that the manufacturer says flat out those sensors are +/- 0.2 or so AFR in the first place, completely seperate from the driver and measurement circuit accuracy.

A couple more reasons accuracy isn't that important but repeatability is.

I guess if you are in a lab and correct for temps and backpressure then accuracy matters.