I'm super interested in this but that link to the testing is exactly the opposite everyone here is saying. I have a PowerFC running closed loop with no 02, it was tuned before. I'd just like to get some accurate 02 readings to I can adjust my fuel.

 

I'm not arguing that AEMs aren't bad or anything, but I am very curious what everyone finds bad about them? Same as bigD, mine has been perfect, so of course I wouldn't know. And also the same as him, with no potential for calibration, now I'm also considering getting a new sensor just for comparison's sake to make sure mine is still giving me good information.

This is CONFUSING!

 

Thats why I said somone musta got payed ...

I've seen AEM widebands that worked fine then I've also seens several that were off.. Alotta the cars I tune have them installed but I always hookup my own wideband for tuning, and the AEMs are consistantly inconsistant... On top of that their analog output is not linear which makes it tough to get a logged value that matches the displayed value Its not the sensors themselves, most of the widebands now a days use the standard LSU 4.2, its more in the way the sensor is controlled, and the integrity of the signal on its way to the controller and how its handled once its in the controller and gets converted to your display..

 

ive had issues with the lm1 and the aem, plx NEVER had an issue with

 

What test are we refering to above, I dont see an NGK being tested or reporting being testing in any of the replies above.

The NGK comes with a Bosch but can use the NGK sensor as well by just recalibrating to the new sensor. Even using the Bosch sensor the AFX unit has proven to be pretty damn good from my experience with it over the last year.

 

It's this test. Seems to be a VERY well done test, other than for the fact that the results seem kind of backward to eveyone's experiences with the various brands/models:

 

So, just to confirm, the PLX would be the most compatible with Hondata right?

 

The thing is, all you guys that say "mine is working great", what do you have to base that on? If it was half a point off on a/f ratio, would you know?

 

I'd say not. . . .

Do you have a suggestion, or just highlighting the fact that a lot of people who are happy with their widebands (including me) might fall into the "ignorance is bliss" category?

 

Exactly. I see so many wideband meters that are off compared to my Motec PLM that I have lost faith in most of them. Unfortunately, without something to compare against, you are just assuming that your readings are correct.

 

Tony when you worked your Texas magic on mine, I think my AFX was reading pretty close your Motec if I recall... but I dont remember exactly. I am sure you would remember better than me.

Edit: To clarify so no one misreads that... the Motec is an amazing unit and in no way am I saying the AFX is on that level. Just saying its accuracy was pretty close I thought but the Motec is far superior for tuning. The AFX is a good end user cost effective type gauge for checking a/f in my mind, I dont tune... I leave that to the pro's with better toys.

 

You must have seen a lot of widebands come through your shop in cars you were tuning, have you noticed a brand that seems to be less often wrong when compared to the rest? (or even more often "accurate" when compared to the rest?)

 

TONY1, since i will not be needing a WB to tune with, and you cant compare any to motec, which WB would you suggest or often sell to customers?

 

Same here. I went with the plx because of my hondata and its great.

 

I've used the PLX m300, and it can read past 14.9? It'll read up to 18 (or 19) until it gives you a LEAN reading.

 

find the cheapest aem you can find, and then call summitt racing. they offer a price match plus whatever percentage AND free shipping after the price match. my buddy got the aem uego for under 180. or buy a used one that is known to work.

 

well i just ordered the PLx m-300te......sooo well c how good it is when i hook it up..

 

The whole article is full of crap.

The accuracy test of holding the sensor the tip of the cylinder is not what I would consider a a repeatable test. Depending on how you hold the sensor and the flow of the gas you can easily make the sensor read +/- 0.5 AFR without even knowing it. I know because I have done the same. They need a rig to mount the sensor so that it is always in the same position, and also need to measure the pressure of the gas around opening of the sensor to ensure that the pressure not too high or too low. Too much gas pressure will make the sensor read overly rich, wideband O2 sensors have a dependency on pressure. Too little pressure and free air mixes in and makes your readings too lean. If interested, I can make a video showing huge AFR changes just do to a slight change in the position of where the sensor meets the output of the flow of test gas.

The response test is also a bunch of bunk. The test is not repeatable, you can not tell which controller has the fastest response time based on having multiple sensors with its own controller in the exhaust. There is no "god" curve to compare the output curves of the various controllers to determine which one most closely follows the "god" curve.

The test for response time should be a chamber filled with normal atmospheric free air. A solenoid connects a tank of test gas to the chamber. At T0 the solenoid opens and high velocity test gas fill the chamber. The time it takes for the output of the controller to match the test gas free of oscillations on the output would be T1. The response time is T1-T0. A pressure sensor should also be mounted in the chamber to ensure that the pressure is not too high to cause large errors in the readings. This test is repeatable with each controller. The T1-T0 will not be the actual response time as you do not know how long it take the test gas to reach the sensor and how much of the free air in the chamber mixes with the test gas before being totally pushed out by the flow of test gases. However those things will be constant for each controller test, so the result of T1-T0 is valid to determine which controller has the fastest response time, you just can not determine with absolute certainty the exact response time. You can use the data from the pressure sensor mounted near the tip of the sensor as a reasonable indicator as to when the gas actually reaches the sensor and have a pesudo absolute measure of response time.

Furthermore for the accuracy test, they run the unit on a bench with a bench power supply. If you are working a lab and need to measure test gases, this would be a reasonable setup. However for all of us we have the unit mounted in the car, the power supply is shared with the stereo system, AC, injectors, ignition coils, HID lamps, etcc.. Also the power supplied by an alternator. All this means there will be lots of noise on the power and ground, and interference from high current injector switching and ignition switching. The Accuracy test with the test gas should be run with the unit inside the car and the unit powered by the cars electrical system. There also be variation in the age of the electrical system and the components and loads in the electrical system, also the car should be strapped to a dyno so the accuracy can be measure during idle, part throttle, and WOT. I am strongly convinced that one of the higher ranked units in the shootout will fall greatly if it is subject to the typical noise and power fluctuations of a typical automotive enviroment.

a high school lab session has much more rigorous standards than the test run in the article.

 

That's all well and good, but it's also totally unusable for almost all of us on here. Do you think you could offer some useful information to help people make better choices than they are currently?

Looking forward to it, thanks!

 

everyone says the plx, that test said it was the worst?

and i have an i-phone and innovate has an app for the wideband badass butits only obd-2 :-(

 

There really is no way to know for certain which controller has the best accuracy and response time, unless someone does rigorous testing and publishes the data. Since an article of that nature does not exist, I would buy based on features and price, just assume accuracy and response time is the same among all controllers.

Also Free Air calibration is 99% BS, the sensor has 2 distinct operating regions; rich and lean. Each region has it's own distinct chemical reaction. Free air calibration will only tell you characteristics of the lean operating region, which for most of us is useless because we all tune for rich. The only way to calibrate the rich operating region is to use rich test gases.

 

From my limited knowledge with tuning I would have to recommend the LC1. We compared that with the AFX and the LC1 was the better of the 2, hands down!

 

Could you maybe elaborate a little on what ways you found the LC-1 to be better please?

I kind of think that if we can collect some details of peoples' comparative observations that might kind of focus us in on exactly what wideband really does work the best for normal users like the people on this board.

At this point I think I'd sooner try the LC-1 than the AFX, but observations like yours could be really useful to confirming that suspicion.

Thanks!

 

Well compared to what the laptop was showing via Hondata, the LC1 appeared to rely closer to that then the AFX. It also updated faster and did not freeze like the AFX. Again this is just what a buddy and myself found.

 

We've had multiple AEM widebands crap put and sit at 14.5-7:1 and not move, replaced the sensors, then they worked again, all the ones that died were under 6 months old. the are very hit and miss