View Full Version : wideband sensor accuracy
740ATL
12-07-2008, 08:08 AM
-
http://bp1.blogger.com/_X0qiKiqN1Tw/RpF95oIow_I/AAAAAAAAABY/v3kUR86FC-I/s400/chart_lg.jpg
I never really was able to get the JAW to work. It kept giving me super shaky readings and liked to just plain not work some days. I ended up going with the AEM. I love it, its so simple and does exactly what I need it to.
740ATL
12-07-2008, 09:36 AM
Yeah, talk to me more about the AEM. The LC-1 may have the most accurate readings, but after listening to the horror stories, I doubt I'd ever own one except maybe if I found one.
Where did you get your AEM?
EDIT... wow... Nick, is this the unit you're referring to?
http://www.aempower.com/images/categories/d7a7e743-9297-4035-a0e9-d583f68561e6.jpg
m4ff3w
12-07-2008, 10:19 AM
http://www.zeitronix.com/press/WBshootout8-600.jpg
Import Tuner AF Meter Shootout (http://www.zeitronix.com/press/afrshootout1.htm)
740ATL
12-07-2008, 10:44 AM
Thanks for the link. I wonder why did they chose to run the tests on a vehicle that went down to an afr of 10.9 if one of the sensors they were trying to test went outside its limits past 11:1. :)
Mike, the AEM is as basic as a wideband can get. It comes with a pre-calibrated sensor so no more fee-air calibrations. Simple 4 wire hookup. In a car full of uncertainties, the wideband is one thing I can be 100% sure about. MS reads it just fine and my numbers in MT are the same as the readout from the gauge. I bought it through work using my employee discount when I got tired of fooling around w/ the JAW. IIRC, it came out to 230$-ish after tax, which is what you can find them for around most online retailers.
I was skeptical to try something made by AEM... I've heard good and bad, so we'll see. A month later and its working great.
Also, FWIW, my AEM never goes 'off line'. It keeps reading all the way down to 10:1... I dont know what ImportTuner was doing wrong, but I don't see anything in the documentation that indicates that the sensor stops working at 11:1.
http://www.aempower.com/images/products/Installation%20Instructions%2030-4100.pdf
I can post graphs if you want to see some data logging w/ said wideband, but it definitely works well into the 11-10:1 range.
740ATL
12-07-2008, 01:45 PM
Cool. You've definitely piqued my interest in the AEM unit. I was looking for something inexpensive to begin with, the fact that it comes with a decent looking gauge makes it that much better.
Nice to hear that it measures down to 10:1. I imagine they've made a few changes since that article appeared.
On a side note... why is it that the tech edge sensors are never included in anyones' tests?
birchfur
12-11-2008, 12:21 AM
I am also have aem wideband gauge. Been using it for about 9 months now and has been working great. I've only used it with kjet so far, as i'm in the process of switching to megasquirt. The gauge reads down to 10.0 and never goes off line, it just bottoms out at 10.0. I also have the aem trueboost gauge/controller to go along side it. So far has been a great boost controller, little tricky to set up though. I purchased both on ebay, dont recall the price though.
500dollar744ti
12-11-2008, 02:09 PM
mine (NGK) has automatic and manually adjustable calibration. i used the read out from my dyno run to fine tune the calibration.
740ATL
12-11-2008, 03:01 PM
Is there a standard one could use to compare against to verify that a wideband is working properly?
I suppose one could hook up a nb and see if it's reading 0.5v when the wb is at 2.5v (14.7:1).
m4ff3w
12-11-2008, 05:58 PM
Is there a standard one could use to compare against to verify that a wideband is working properly?
I suppose one could hook up a nb and see if it's reading 0.5v when the wb is at 2.5v (14.7:1).
You could always do a dyno run and see if they match?
Lord Tentacle
12-11-2008, 06:09 PM
I am a large fan of the PLX SM-AFR
no free air calibrations, just simply works
made tunning the tentacle wagon super easy
The Aspirator
12-11-2008, 09:05 PM
My tech edge is working, but I have no idea about its accuracy sometimes. I have this haunting/gut wrenching feeling that it's reading too rich than it actually is.
MikeHey Mike I'm in the exact same boat as you are. I've been successfully using my Tech Edge wideband for over 4 years I think, it works fantastically and has one of the fastest refresh rates of any gauge I've seen (I can't stand the slow ones). But I'm curious and wary about its accuracy.
I've got a JAW waiting for me at a friends house, so I just have to solder it up and wire it in, but I would need to buy a second sensor and weld in a new bung before I could test the two gauges side by side. I would like to do that, so maybe one day I will. But the JAW is like $70 unassembled with a gauge, I couldn't turn it down.
740ATL
12-11-2008, 09:19 PM
Hey Mike I'm in the exact same boat as you are. I've been successfully using my Tech Edge wideband for over 4 years I think, it works fantastically and has one of the fastest refresh rates of any gauge I've seen (I can't stand the slow ones). But I'm curious and wary about its accuracy.
I've got a JAW waiting for me at a friends house, so I just have to solder it up and wire it in, but I would need to buy a second sensor and weld in a new bung before I could test the two gauges side by side. I would like to do that, so maybe one day I will. But the JAW is like $70 unassembled with a gauge, I couldn't turn it down.
That's where I'm at right now. I'm tempted to buy a 2nd unit and sensor to "test" the original unit.
Taking it to a dyno shop to compare isn't always great b/c they usually stick the probe into the tailpipe to test, which prompts inaccuracies due to the catalytic converter.
I've tried reading the plug to compare, but due to the ethanol (up to 10%) in the gas, plug readings are ambiguous. I'm not seeing any little flecks on the plugs and there doesn't appear to be detonation, but there's those situations where the knock sensor is going off when it says I'm at 12:1afr....
There is a specific gas that they sell specifically to test widebands.... Fordmuscle.com used it when they did their wb shootout.
It's weird, I'm a slob in real life, but when it comes to megasquirt I nitpick.
740Weapon
12-11-2008, 10:15 PM
i support the narrow band as a second check and a knocksense as a third check.
740ATL
12-11-2008, 10:22 PM
i support the narrow band as a second check and a knocksense as a third check.
nb yes, but the knocksense isn't working that well on this car. For some reason, this engine is noisier that my 16v and I have the sensitivity knob pegged.
kildea
12-11-2008, 11:59 PM
Anyone going from the LC-1 to a JAW?
i'm going from an lc-1 to a jaw.
mostly just out of curiosity, but i will mention that when karl fixed my wiring for me and i started seeing 13.5v rather than 11.5-12v, the sensor started reading way rich.
i tuned it with a target of ~12:1 out of boost this week (and 11:1-9:1 up to 200kpa) and after recalibrating it today i found that it put me closer to 14:1 in the same regions where i had been around 12:1 prior to the recalibration.
so my point is that the lc-1 appears to be rather sensitive to the stability of its power source, probably obvious, but i figure it's worth mentioning.
as for the knocksense it doesn't work on my car either unfortunately, honestly the plugs are probably your best guage, but the nb is worth having around... to know if you are getting scary lean and staying there. (assuming you can't trust the wb)
The Aspirator
12-12-2008, 03:33 AM
It probably doesn't help that I can't remember the last time I calibrated my sensor :).
crazy0000
12-12-2008, 03:40 AM
My knocksense goes off at 3600-3800 blinking but no knock, its like a resonance thing, and then goes off at 4000 to redline (haha I really need to turn it down).
Anyway, I'm loving the JAW, also I don't know if you guys follow the JAW forum, but you can hook up 3 LEDs to it, one to tell you when its on, one to tell you when the sensor is too hot(I think it means it reads leaner slightly) and the other LED, I have no idea I can't remember. There has been a firmware update, a few people don't like it, I personally like it, it made the readouts more stable and also changed the layout of the JAW edit program. Which is the program you use to program the JAWs outputs and whatnot. I'm thinking of hooking up the LEDs on my JAW shortly so Ill let you guys know how it goes.
Also tuned a bit more the other day, threw a bit more at my engine, I'm liking it haha
EricF
12-12-2008, 11:43 AM
I just got my Autometer Ultra Lite installed. It's got a cool WOT switch peak lean recall function, so if you punch it and don't want to stare at the gauge while you are wide open, you can just recall the most lean it went after you let off the throttle. You of course need a WOT switch though. It also has a datalog output for EMS, but I have it in my 850 now so I'm not using that. Also has an alarm function
One way you may want to augment your gauging is EGT, if you are skeptical about the wideband.
The autometer:
http://www.autometer.com/cat_gaugedetail.aspx?gid=3604&sid=11
Was 240 shipped on E-bay.. Only reason I went this route instead of AEM was to match my other gauges :oops:
500dollar744ti
12-12-2008, 01:28 PM
My knocksense goes off at 3600-3800 blinking but no knock, its like a resonance thing, and then goes off at 4000 to redline (haha I really need to turn it down).
calibrate it in neutral so that you can revv it all the way up and not flash but be on the breaking point of flashing. like keep turning it slowly until there's no flash while revving at a stop.
MikeSr.
12-12-2008, 01:44 PM
Hey, Mike-for calibrating testing equipment, you can obtain small cylinders of test gases with guaranteed extreme accuracy in mixtures. Then you could build a little manifold to mount the sensor in and add test gas. Probably need two different mixtures to obtain some kind of slope to the sensor's response at extremes of operation. Dry gas of course. An alternative would be to use nitrogen and oxygen mixtures with some accurate flow delivery. These automotive sensors are relatively crude. In medical testing, we use sensors capable of accuracy to 1 ppm, and calibrate them to known dry gases.
vBulletin® v3.8.4, Copyright ©2000-2010, Jelsoft Enterprises Ltd.