Wideband Issues

[TiredGXP]

I have a weird issue with an AEM UEGO wideband. The gauge is providing consistent readings over time, but it looks like the output voltage signal is progressively deviating from the spec sheet.

I've adjusted the offset down twice now, but it looks like it has deviated another 0.5 in the last week, for a total offset required in the custom pid of 1 AFR.

Anyone encountered this? Solutions?

[5_Liter_Eater]

Mine seems to be doing the same thing. I had it and HPT jiving and then all the sudden the gauge reading was off from HPT by a good bit.

Since there is no free to air calibration on it the only solutions I can think of are to by another sensor or another WB alltogether.

Not so impressed with my AEM.

[gold98Z28]

Not so impressed with my AEM.
thats an understatment mine has never worked the first time.

[TiredGXP]

5LE - I think things are a bit different (maybe) between our issues. Only the output to the EIO interface seems to be changing, not the gauge display. I originally had to set an offset of 9.7 to get the the gauge and HPT in sync, but I keep having to change the offset in the HPT custom PID to keep the two in sync.

The theory is that 0V = AFR of 10:1 according to the wideband spec sheet, I'm now at 0V = AFR of 8.9:1 to get the gauge to equal the scan. At this offset, the gauge, scan and NB all seem to be consistent, but it's a moving target - how do I trust the scan from day to day?

Like you, I'm not terribly impressed. Since we're in the middle of a cold snap (-20*C) I'm in the middle of gathering data for Bias analysis, but a week's worth of data just went out the window.

Oh well, back to closed loop until summer.

[EC_Tune]

How about verifying with a Digital Volt Meter? Or put a jumper from the #1 input to the #2 input and log both AFR and voltage at the same time (different channels though). If your wideband voltage matches what the gauge says it's supposed to then it's the HPT input. If the output voltage doesn't match the gauge, then ignore the gauge.

HTH

[TiredGXP]

EC - good thought.

I'll definately try the jumper thing to see what voltage is being picked up in a different input.

[5_Liter_Eater]

I think we're talking about the same thing; having to change the offset to get the gauge and HPT to be the same after they had already been the same for a while. That and the fact that my WB and NB's have never agreed on stoich makes me doubt the accuracy very much.

My next WB will have a free air calibration. Probably the NGK AFX.

[EC_Tune]

Bill,
I've seen quite a few widebands do what you are reporting. There are some that are just not accurate at stoch. My PLX was probably the worst offender: 15.3-15.7 at Stoch on a lot of vehicles not just one. I've had exceptional accuracy with my DynoJet: within a tenth at stoch, that's as good as the sensor. The AutoMeter seems to be pretty accurate as well. NGK: Haven't used one yet but from all reports it's very good.

My only beef with some of the displays is that it's digtal. When I'm tuning I look for movement and a digital display is always moving so it's more distracting then helpful. More options on controlling the "LED rich lean" displays would be welcome.

Regarless of anyones pet theories, narrowbands (lambda sensors) switch at Lambda 1.0 (14.7 for gasoline). That's why the OEM's use them, they don't deviate, even with temperature.

Can big cams affect lambda sensors? Not so much on the exhaust side but on the MAF end when the MAF is too close to the throttle (Camaros have it the closest of the V8's). The early & late (not card style) MAFs measure air in both directions so if you have a big cam and a close MAF you will get fueling errors because the MAF doesn't know direction, just flow.

Are there areas where they are inaccurate, yes, generally at very low MAP, when you get big cam idle missfires, or if there is excessive oil present in the exhaust.

[5_Liter_Eater]

But if it's not accurate at stoich who's to say it's accurate anywhere else?

[EC_Tune]

Yep, and without calibration gasses, who knows...

[Ben Diss]

Bill,
I've seen quite a few widebands do what you are reporting. There are some that are just not accurate at stoch. My PLX was probably the worst offender: 15.3-15.7 at Stoch on a lot of vehicles not just one. I've had exceptional accuracy with my DynoJet: within a tenth at stoch, that's as good as the sensor. The AutoMeter seems to be pretty accurate as well. NGK: Haven't used one yet but from all reports it's very good.

My only beef with some of the displays is that it's digtal. When I'm tuning I look for movement and a digital display is always moving so it's more distracting then helpful. More options on controlling the "LED rich lean" displays would be welcome.

Regarless of anyones pet theories, narrowbands (lambda sensors) switch at Lambda 1.0 (14.7 for gasoline). That's why the OEM's use them, they don't deviate, even with temperature.

Can big cams affect lambda sensors? Not so much on the exhaust side but on the MAF end when the MAF is too close to the throttle (Camaros have it the closest of the V8's). The early & late (not card style) MAFs measure air in both directions so if you have a big cam and a close MAF you will get fueling errors because the MAF doesn't know direction, just flow.

Are there areas where they are inaccurate, yes, generally at very low MAP, when you get big cam idle missfires, or if there is excessive oil present in the exhaust.

I have no experience with anything else, so take this for what it's worth.

I'm using an NGK. One thing that's nice about it, is that on power up it outputs stoich for a few seconds, goes to zero volts for a few and then gives you the measured output. This way it's easy to know at the beginning of the logging session that it's working correctly.

[EC_Tune]

Well, that will tell you the min & max and Stoch output points of the D/A converter but it won't really tell you if the sensor control system is accurate. And the control circuit is what is important.

[LBSBLOWN]

I have had the AEM WB for two years and the Gauge and HPT have always agreed with in .1 afr I can check it at 14.7 on the gauge and at that moment mark a comment on HPT scanner at it shows 14.68 and at WOT the gauge will show 11.6 and hpt will log it at 11.5 and its done this with this accuracy for two years now.
Are you guys confirming that you have good low resistance grounds at the sensor, gauge and for the EIO?

[05HD]

AEM widebands have been pretty reliable in my experience, aside from the really early ones that were suceptible to display malfunctions in cold weather. Also, like any wideband with a Bosch sensor, they are susceptable to lead contamination more easily than others.

They are easy to calibrate in logging software, since they automatically output the voltage equivelant of 14.8:1 a:f when the sensor is unplugged from the guage. Ground offset is the biggest problem getting steady wideband readings from another device, like the HPT. It is even more frustrating/important when you try to run closed loop off of that reading!

[Dirty Howie]

Seems like some of you guys think that resistance and good grounding can influence the the accuracy of these WB's at stoich.

I know my PLX comes with capacitors which were never installed...maybe this would make a difference.

Bill, Does yours come with any capacitors you didn't install??

Doug, Do you think this will make a difference. Did you install or experiment with the capacitors before tossing your PLX ?????


DH

[EC_Tune]

Didn't get a chance to check the cap's. LOL

[vuss383]

Been using Dynojet WideBand Commander for 3 or so yrs now . Am on my second sensor ( not that a change was needed ) & can not fault the unit at all to date . Seems very consistant on all my OLSD cars year round ! Have been thinking of purchassing another just for comparrison between the 2 but have no idea of what to buy !

[5_Liter_Eater]

If I were to buy another WB right now I think it would be the NGK AFX.

[sixvi6-camaro]

DH... I didn't update my post in the old PLX thread but I was using the Capacitors on my PLX and it didn't change anything.. I also tried multiple grounding points. I ended up going a little crazy and installed an LC1 on the drivers side. The LC1 agreed with the commanded AFR at stoich and WOT while the PLX still read its usual .5 - 1 AFR off. as I last stitch effort I even swaped the sensors from left to right and did a quick test to rule out the pass bank actually reading lean and it didn't matter the PLX still read lean. since then I completely removed the PLX and only run the LC1 now.

[Dirty Howie]

DH... I didn't update my post in the old PLX thread but I was using the Capacitors on my PLX and it didn't change anything.. I also tried multiple grounding points. I ended up going a little crazy and installed an LC1 on the drivers side. The LC1 agreed with the commanded AFR at stoich and WOT while the PLX still read its usual .5 - 1 AFR off. as I last stitch effort I even swaped the sensors from left to right and did a quick test to rule out the pass bank actually reading lean and it didn't matter the PLX still read lean. since then I completely removed the PLX and only run the LC1 now.

Thanks very much for this update. That seems pretty conclusive to me ..... same car ..... LC1 good and PlX bad .......!!!!!!!!!!!!!

Did you happen to note how they compared at WOT afr ??????

Thanks again for the update.......I won't worry about digging up my capacitors!!

I would like to get a list together of all us PLX owners and confront PLX with our data and demand they fix it for us!!!!!!


DH