I have this error reading offset by around 0.5 between the UEGO and the AEMTuner using AEM EMS 2 ?!

I called AEM, they said might be grounding issues..

I tried chassis ground, still same offset...

I tried sensor ground on EMS .. the reading became totally different.

The UEGO has 4 wires..Red.Black.White and blue.. blue for serial.. not using for the moment

White for input to the O2 wire on EMS (PIN C16)
Red 12V from battery
and Black, I use to have it on battery ground.. it was offseted by 0.5 from the UEGO
on chassis ground same thing.. I tried EMS Sensor ground (Pin D9) .. the offset became like 2 ..and the reading on UEGO changed..

The UEGO should be reading correct when grounded to chassis or battery, because on idle it should 14.8 or 14.9

I might be able to calibrate it manually using the sensor tab on AEMTuner and change each value on each voltage respectively..
but that would be inaccurate ?!! on the AEMTuner there is a setup wizard where you can choose the type of O2 sensor you have and I chose 30-4100 ... that what I got.. offset by 0.5

If I choose a different model the offset becomes huge...

 

Did you get this figured out?? I know this is 10 years old but it's the exact issue I have. I have not had any luck with reading stuff online, and cannot pm you directly with my new account.

I've also just now messaged aem tech support, I'll report back if they reply.

I have a bunch of new old stock stuff I bought 10 years ago and decided to finally open the boxes and use it. I have a series 2 like you're using and the same oldschool wideband 30-4100. In the AEM series 2 wizard I have selected the correct 30-4100, I tried different ground points, and now also a new o2 sensor. The problem I have is the gauge reading and the reading on the laptop is off from .5-1.0afr, I haven't paid too close attention if it's exactly the same amount off in different afr points but it's seemingly off all over. Am I supposed to adjust "o2 sensor 1 calibration"? I've unplugged the sensor, adjusted the sensor calibration to match, but when hooked back up and running the gauge and laptop numbers are again different. Do I need to find another way to power and ground the series 2 ems and the wideband? Or do I need another wideband?

 

The issue is a ground offset, unless WB is faulty (some batches of older AEM WB (non-xSeries) had issues). My ProSport WB was .25 point off because I chose to use the chassis ground instead of a sensor ground. I was just lazy, I knew what I was doing and how to get around it. Analog signal is crap compared to SERIAL or CAN, so next time just buy X-Series gauge and use CAN (if your ECU supports it).

Your potential fix with Analog POS signal as follows:
A.

1. Ground all analog sensors to ONLY a designated/labeled 'sensor ground' PIN on the ECU. (Do not ground the WB itself to the ECU PIN, do ground the Analog signal negative (-))

B. - If the dedicated 'sensor ground' PIN did not solve the problem then:

1. Eyeball the wideband gauge and the ECU reading. Gauge is the correct reading (unless WB is faulty) and your analog signal fed into the ECU is incorrect.
2. Calculate the % discrepancy at idle, assuming it's set at 1.00LAMBDA aka 14.7AFR (How to calculate discrepancy %: Take ECU AFR reading of 14.00 / Gauge AFR reading of 14.70 = .95 LAMBDA)
3. You've now assessed that your ECU reading 5% rich/incorrect
4. Access LAMDA sensor setup in ECU, shift suggested Voltage values by 5% (+/-)
5. Cross-reference WB gauge & ECU reading again, you should be MUCH closer now

With someone's help you can shut your motor off and someone can immediately plug both tailpipes off. If wideband is good enough to read very lean values, it will be much easier for you to see the discrepancy.

Remember, you trust the gauge and not what the ECU is ingesting from your shotty analog signal. DO report back if this helps/solves the issue.

P.S. Before anyone asks why the LAMBDA/WB gauge is mostly okay with being grounded anywhere and analog signal into the ECU requires this clean-as-a-whistle 'sensor ground'. WB controllers can tolerate voltage variance (to some level) and feed the gauge mostly correct signal. The chip that is responsible for feeding the analog signal to an ECU, is the touchy one, hence why we want as clean of a ground as possible.

 

thanks for the reply.

I've been afraid to use the ecu sensor ground being a heated o2 sensor, worried if it can support the current. Do you think this is a non issue?

Before I had read through the manual it says to ground wideband to chassis or battery, I was grounded to battery, then ran another wire straight to the battery.

 

If unsure or worried about approach A. Then just play it safe and correct the voltage offset per approach B. Again, I did that with my analog WB and it worked just fine.

Edit:
I do understand your concern with the current, and agree. You do not want to overload that ECU pin. To clarify the ground I speak of is the analog signal ground and not the wideband gauge itself. For example, my AEM WB has a black ground for gauge and a brown wire for analog signal ground.

Edit #2:
If you are still getting wild swings, consider grounding your actual WB & Analog grounds to only the engine block, or your chassis. I read that negative battery (-) terminal could introduce voltage drops/fluctuation due to varying alternator load and starter loads. I would imagine even the WB controller box that can deal with some of those fluctuations would be impacted and your gauge data could be skewed, maybe?

 

Hi,

I'll try out what you said. This older aem wideband has 4 wires, power, ground, 0-5v output and an unused serial output.

I had time to unplug the sensor from the gauge, the gauge goes to 14.7 and the laptop shows the same, I did not have to adjust the o2 sensor calibration. Unfortunately the engine is now partially apart, I'll get it back together for the weekend to do some more testing. From memory, when the engine is running the voltages change and the afr will go off.

AEM also got back to me today, this was the reply...

"Do you have a multimeter you can use to test the voltage at the white wire
coming from the gauge? If so, please provide the following information:

Value displayed on gauge:
Value displayed in software:
White wire voltage:"

I'll do some testing and post back at the end of the week.

 

Nice! Glad to see AEM respond. Keep us posted.

 

With the sensor unplugged the AFR matched the laptop, voltage matches the laptop too. There is an adjustment for this but it doesn't need any adjustment with sensor unplugged.

With the sensor unplugged it's all good. It doesn't need any o2 sensor calibration adjustment.
2.392 volts at the wideband output white wire to the ems.
2.39/2.40 on the laptop.
14.7 on gauge
14.7 on laptop

with sensor plugged in and KOEO,
4.4v at white wire
4.4-4.5 volts on laptop,
gauge afr off the chart, - - -
the laptop shows 18.8 afr

When engine running things change
white wire about 2.3, afr on gauge is 13.3(this is odd)
laptop shows about 2.3v but 14.5

To check it at another voltage. If I shut off the engine, take readings immediately is says,
white wire 3.8v
15.6afr on gauge
laptop, 3.8v 17.4afr

The voltages are pretty similar, it's just with the engine on there is an afr difference.
I wasn't sure about the accuracy of my old sensor so this is a new in box aem sensor I had.

Usually I'd think to trust the gauge but the afr more closely matches what the laptop is showing. I'll wait to hear back from AEM. I'd hate to buy another wideband just to test this.

 

I don’t know AEM V2 capabilities very well but can you wire WB in using serial to avoid all this voltage nonsense? Just to check if there is still discrepancy.

 

By itself, it will only accept a 0-5v signal. I hadn't thought of this prior, but you could potentially get it on AEMnet via an expansion module and then use it from there. I have not looked into the software on how to implement that yet... At least for feedback.