NGK AFX vs. AEM 30-0300 CAN Comparison

[10_SS]

10_SS- I'd be a little concerned about this too...

Somethings going on with the scanner.... the pic below is what I see on my main logging computer and my work desktop computer... now wondering if the fueling error you guys see is messed up too...

Wots.PNG

[gmtech16450yz]

Somethings going on with the scanner.... the pic below is what I see on my main logging computer and my work desktop computer... now wondering if the fueling error you guys see is messed up too...

Wots.PNG

Mine is on "Highest", yours is on "Average". I prefer to look at highest when looking at knock. As you can see, if you only look at the average you end up missing some huge knock events. Especially on a long drive, you could have 12 degrees of knock showing as an average of .5 degrees if it only happened a couple times on a drive.

[La RPM]

(Larry Larson's tuner) Once told me that accuracy does not matter in a wideband as much as Consistency..

MY opinion however... is that AEM isn't exactly known for making quality products...

DSteck is right, You can feed the scanner some real skewed readings over CAN and it'd be no different.

OP hasn't even informed how often he calibrates his O2 in free air. as anyone who wants CONSISTENCY should.

and if you ask me, If a marketed wideband has no means of free air calibration, I wouldn't touch it with a 10 foot pole.

[schpenxel]

You do realize 4.9 sensors use a different basis for calibration now, right?

[La RPM]

You do realize 4.9 sensors use a different basis for calibration now, right?

I'm aware that they use a current pumping circuit, and from what I understand, That was done for longevity purposes because the 4.2 had an air reference, yet most companies will solder in resistors and chips to make them "self calibrating" and not able to calibrate in free air.

I don't know if AEM has setup the 4.9 that way. I know PLX does something similar.


on a side note, someone needs to come out with a CAN/Serial integration for the NGK wideband. Would be pretty kickass. I hate that little green block on the side of the interace

[10_SS]

I'm aware that they use a current pumping circuit, and from what I understand, That was done for longevity purposes because the 4.2 had an air reference, yet most companies will solder in resistors and chips to make them "self calibrating" and not able to calibrate in free air.

I don't know if AEM has setup the 4.9 that way. I know PLX does something similar.


on a side note, someone needs to come out with a CAN/Serial integration for the NGK wideband. Would be pretty kickass. I hate that little green block on the side of the interace

yes the 30-0300 with 4.9 can be free air calibrated

[La RPM]

yes the 30-0300 with 4.9 can be free air calibrated

Does the AEM controller allow this? or is it using something they built into their controller

[10_SS]

Dunno, it says you can run the resistor calibration or free air... whatever you wish. This is for the 30-0300

AEM_CAL.JPG

[GHuggins]

Gmtech - not trying to open up a can of worms here and just let me start by stating this up front... I haven't used this AEM wideband to know just how well it works or how accurate or inaccurate it may be... BUT, one very simple point here -

Headers are installed on most all vehicles out there - relocated O2's and can't always be tuned in perfectly due to tables not being available = this alone can, but not necessarily explain differences seen in stoic to wideband readings during closed loop operation... As Ed has hinted at and as I have personally witnessed - 450mv doesn't always mean stoic - O2's are very inaccurate and I can program them to switch at 450mv, but actually be something like a 14.9 A/F - I know in theory 450 is above all else supposed to be stoic, but you have to keep in mind how ethanol and everything else plays on this... I will only do the stoic switching comparison for dialing in the O2's, not for sanity checking the wideband - this of course requires the wideband to read correctly, but not worried about that too much with mine... Other known facts - O2's age and don't read or work as accurately - same can be said for widebands - this is why I try to do a 1hr free air burn off in between each tuning session - ECM will let them burn off overnight before testing - the ngk has been in this exhaust system for who knows how long and has been used for who knows how long - and we're comparing this to a brand new out of the box wideband... NOW something I would like to see - stock headers - wideband in same spots opposing one another and see which one works or reads more accurately - think this is the only 100% way to compare them...

Personally I believe affordable accurate widebands are needed and hopefully they're accurate, but for tuning professionally I'll keep using my lambda pro built by ECM......

Really not trying to get or keep anything going here - just stating my observations from the years...

[JBZ]

someone needs to come out with a CAN/Serial integration for the NGK wideband. Would be pretty kickass. I hate that little green block on the side of the interace

this!

I didn't buy the interface with the little green block so I need to find a way to log mine cheaper then $289.00 and no ac pressure sensor either. guess I just need to upgrade

[LSxpwrdZ]

Gmtech - not trying to open up a can of worms here and just let me start by stating this up front... I haven't used this AEM wideband to know just how well it works or how accurate or inaccurate it may be... BUT, one very simple point here -

Headers are installed on most all vehicles out there - relocated O2's and can't always be tuned in perfectly due to tables not being available = this alone can, but not necessarily explain differences seen in stoic to wideband readings during closed loop operation... As Ed has hinted at and as I have personally witnessed - 450mv doesn't always mean stoic - O2's are very inaccurate and I can program them to switch at 450mv, but actually be something like a 14.9 A/F - I know in theory 450 is above all else supposed to be stoic, but you have to keep in mind how ethanol and everything else plays on this... I will only do the stoic switching comparison for dialing in the O2's, not for sanity checking the wideband - this of course requires the wideband to read correctly, but not worried about that too much with mine... Other known facts - O2's age and don't read or work as accurately - same can be said for widebands - this is why I try to do a 1hr free air burn off in between each tuning session - ECM will let them burn off overnight before testing - the ngk has been in this exhaust system for who knows how long and has been used for who knows how long - and we're comparing this to a brand new out of the box wideband... NOW something I would like to see - stock headers - wideband in same spots opposing one another and see which one works or reads more accurately - think this is the only 100% way to compare them...

Personally I believe affordable accurate widebands are needed and hopefully they're accurate, but for tuning professionally I'll keep using my lambda pro built by ECM......

Really not trying to get or keep anything going here - just stating my observations from the years...

Ding ding. I do not trust narrowbands on anything that isn't stock. Completely stock!

Another thing many people don't realize using the NGK/ECM/Ballenger wideband is it's warm up test output. You can alleviate every single voltage issue by simply logging the startup of the wideband. It outputs 1v, 4v and 0v upon startup and simple calculations will show what the wideband "should" read at those voltages by back calculating the voltage to lambda output. My ECM AFM1600L (Converted NGK AFX) has a 0.60-1.30 lambda output range. It will output 0.740L, 1.160L then 0.60L upon startup. This will correct any minuscule variation in the output. So with that said, I would "TRUST" a properly calibrated NGK/ECM/Ballenger wideband over anything AEM ever makes!

[gmtech16450yz]

And STILL nobody is looking at the OP's logs and answering HIS question. You all just want to argue with me about what wideband is better and whether you can use narrowbands as a way to verify wideband operation. Try answering the OP's question guys. His SPECIFIC question about his SPECIFIC SITUATION. Not yours, or your experiences, or your personal bias, or brand preference, or emotion based responses. This is a little like saying a Z06 will turn quicker lap times than a base C6. NO CHIT! But what if PeeWee Herman is driving the Z06 and Randy Pobst is driving the base C6? Hmmmm, could be a little different story huh? THAT'S MY POINT!!! You guys are trying to convince me of what wideband tests better in the lab or works better. I'm NOT saying that. I'm working in the real world of this particular car the OP is asking about and NO OTHER. The greatest tool in the world is only as good as the person using it.

Here are the facts in relation to the OP's particular situation and question. His tune and logs show that the tune is a mess. Fueling is way off and there's a ton of kr. Who knows how much else is off. This we know by looking at the logs. Is there anyone that has a problem with that statement or assessment?

So, going on the fact that the car is not running properly, there are two reasons that I can see that could be the cause...
The first possibility is that the OP doesn't know what he's doing.
The second possibility is that he is getting false data and trying to tune based on that.

This is what should be being discussed here, nothing else. Why are you supposedly smart "experts" not willing to address the OP's particular problem and instead insist on talking in general terms? Again, there are two possibilities, either the OP doesn't know how to tune or the data he's using isn't accurate. Which is it? That's the question. What particular wideband you're fond of should have absolutely NO bearing on your assessment. Look at HIS data, not your own past experiences. After all, this WAS his only question...

So... after thinking about this for awhile.. which data log would you trust?

[LSxpwrdZ]

And STILL nobody is looking at the OP's logs and answering HIS question. You all just want to argue with me about what wideband is better and whether you can use narrowbands as a way to verify wideband operation. Try answering the OP's question guys. His SPECIFIC question about his SPECIFIC SITUATION. Not yours, or your experiences, or your personal bias, or brand preference, or emotion based responses. This is a little like saying a Z06 will turn quicker lap times than a base C6. NO CHIT! But what if PeeWee Herman is driving the Z06 and Randy Pobst is driving the base C6? Hmmmm, could be a little different story huh? THAT'S MY POINT!!! You guys are trying to convince me of what wideband tests better in the lab or works better. I'm NOT saying that. I'm working in the real world of this particular car the OP is asking about and NO OTHER. The greatest tool in the world is only as good as the person using it.

Here are the facts in relation to the OP's particular situation and question. His tune and logs show that the tune is a mess. Fueling is way off and there's a ton of kr. Who knows how much else is off. This we know by looking at the logs. Is there anyone that has a problem with that statement or assessment?

So, going on the fact that the car is not running properly, there are two reasons that I can see that could be the cause...
The first possibility is that the OP doesn't know what he's doing.
The second possibility is that he is getting false data and trying to tune based on that.

This is what should be being discussed here, nothing else. Why are you supposedly smart "experts" not willing to address the OP's particular problem and instead insist on talking in general terms? Again, there are two possibilities, either the OP doesn't know how to tune or the data he's using isn't accurate. Which is it? That's the question. What particular wideband you're fond of should have absolutely NO bearing on your assessment. Look at HIS data, not your own past experiences. After all, this WAS his only question...

Are you blind? Your responses help how? With some experience you'll see. It clearly shows within his logs to back up my claims. Look at the very beginning of his NGK log and you'll see part of the startup procedure I previously spoke of. He has a slight offset throughout the range that needs to be corrected.

With that said still I trust the NGK over the AEM especially after seeing the startup procedure. Even with sensor aging if he free air calibrates it very often at all it'll stay accurate. The tune isn't spot on even with the NGK log though... Even with his offsets on the analog line it's still reading more accurate than the AEM. If you are to blind to see that then I don't know what to tell you.

It seems you are trying to ensue that we aren't answering the question when we ARE. We are trying to tell you that youre being a hard head and that the NGK is more accurate because we DO have experience to back up the claim. So in an extravagant way I am saying I trust the ngk log the most. There's your answer.

OP is targeting a stoich based on 14.68 with a wideband with a stoich setting of 14.57. Switch to lambda and some of that error will get corrected. Some talk about the stability of the NGK vs the AEM within the logs, the AEM isn't responding to "actual" exhaust with that erratic behavior, it's noisy within itself. You put a quality wideband in, dial the fueling in, it will be somewhat smooth with very small oscillation which will come from the pulsing within the exhaust. The erratic reading you see on the AEM log is noise and lack of quality. If you don't believe me then show me WHAT is causing the fueling to read erratic. You'll find that fueling usually tracks what the airflow is doing and that will track similar to pulse width.

[GHuggins]

GmTech - I answered his question in my own "round a bout" way in my post with stating the "FACT's" of aging and so on... As I pointed out or guess should have pointed out better? It was YOU who stated the AEM was obviously more accurate via it's O2 mv comparison - Are you not reading what we are all trying to tell YOU? This comparison is worthless... 10_SS has headers if memory serves? I also thought the overly rich tune was for compensation to his boosting issue - better safe than sorry I thought - so never mentioned anything about it... What I get for assumptions - completely my fault... Admittedly I also didn't go through the log in as much detail - honestly the log wouldn't even display correctly when I viewed it - other's - even 10_SS mentioned he wasn't sure the logs were correct...

Again, not trying to turn this into a big argument - just stating the FACT's to keep people from trying to do things that your trying to push with O2 comparisons - utterly and completely useless unless everything is stock or the O2's have been dialed in in LAB grade conditions for ethanol, length from exhaust valve and so forth all we can do when we're lucky enough to have access to all of the tables is get them decently close!

Plus something else here - we're comparing a CAN network protocol to an analog - of course the CAN is going to comm quicker and show different spikes as it's hitting those cells - BUT if you don't slow it down through filtering somehow - good luck trying to get an accurately tuned fuel table - IF it will show the same differences in cell "cluster's" and it's accurate then I'm all for it - AS I STATED originally, I haven't used one of these - heard good things and have even talked to other people as to what affordable wideband I would recommend for them - I recommend one of these or a plx nearly everytime........

I don't think 10_SS is a bad tuner and yes it's more than possible that the wideband sensor that has been in the exhaust for who knows how long isn't as accurate as it once was - vehicle in question could even have another underlying issue that has contaminated it's O2's and wideband as far as that goes... Another thing if you weren't so caught up in yourself - 10_SS stated fueling has steadily been drifting more and more negative - another reason I'm not 100% that I would trust the O2 readouts currently being seen until the underlying problem or problems are actually found... Just saying don't use O2's to check a wideband's accuracy - if you do this and use voltage offsets or whatever to "dial" in your wideband - you WILL regret it!!!

I'm leaving this with this... I'm no "know it all" by any means and am ALWAYS willing to listen and learn - just can't stand it when someone is trying to justify or push something off of false information...

[gmtech16450yz]

Lol. No problem. Sorry I was being so blind and ignorant. Maybe someday I'll get enough experience to work out of a barn in kintucky.

[GHuggins]

I admittedly was a little ticked with my last response - posted before reading LSX's much better informed response... Think ALL of the evidence should have been made clear with that

[LSxpwrdZ]

Lol. No problem. Sorry I was being so blind and ignorant. Maybe someday I'll get enough experience to work out of a barn in kintucky.

I would expect that type of response when there was nothing else to say. As Forrest Gump would say, "that's all I got to say about that"

[matty b]

This is reminding me of, oh yeah thats it, a waste of time. What bothers me is that Mr Tech is leaving info for less informed individuals to find later on that makes them think that if the stock O2 reads 450mV and the HEGO reads lambda 1 and follows the curve, that it is ACCURATE. Mr Tech obviously doesn't understand what accuracy is or elese he wouldn't have made all these posts about it.

Accuracy is based on a standard, I have standards for my scale as well as my mics. They dont change if I use them in the conditions they should be ie: I dont let my standards sit in the shop and calibrate my mics that have been sitting in the air conditioned office with them. Even then, my standards have tolerance and my mics have a certain amount of precision. Mr Tech doesnt seem to understand that accurancy, precision and tolerance are not mutually exclusive.

[schpenxel]

Hopefully you guys will start getting a little rowdier, I've been wanting to delete some posts for a while

Kidding.. but seriously, we're all here to learn, no reason to be difficult or get into personal attacks over it.

[Montecarlodrag]

I trust the AFX more than any AEM. Properly calibrated and set you get very precise and consistent results.
I know this is not a wideband comparison, but I won't trust any AEM wideband to tune a forced induction engine, ever. I'd better use a Innovate instead of an AEM.
The AEM is good 30-0300 is great for NA tunes, stock or mildly modified cars only. Well, that's my opinion

Best Regards.