wideband reads lean constantly

[2.0L_turbo]

got another one for ya guys,

 

got my wideband working in link now having an interesting issue, when I take the car for a drive the wideband reads lean constantly and sends the same signal to link with the lambda reading about 1.262. I took a small drive around the block a few times for around 10 or so mins to get a log to check my mixture map with. the mixture map then told me I was running lean ( already figured that ) so it added fuel in those areas. We made the changes and went for another spin around the block only this time the car was very hesitant and made no power. the numbers in the fuel table were originally around 34 in fuel table 1 and with the changes the mixture map made it bumped it into low 40s. when I make manual adjustments and put it back to 34 it runs a lot better. It almost feels like the fuel map percentage is backwards as I need to lower the number to even read on my wideband.

 

would this indicate that I have a failing wideband sensor, or would incorrect injector dead times cause this?

[Adamw]

To confirm the voltage calibration try unplugging the sensor, Im pretty sure the AEM controllers show 1.00 Lambda with the sensor unplugged.  If that looks ok then I would suspect the sensor or possibly an air leak somewhere.   

[2.0L_turbo]

8 hours ago, Adamw said:

To confirm the voltage calibration try unplugging the sensor, Im pretty sure the AEM controllers show 1.00 Lambda with the sensor unplugged.  If that looks ok then I would suspect the sensor or possibly an air leak somewhere.   

ok will do, I do know that there is a hole in the exhaust just a little before the sensor on a flex pipe. I may try moving the sensor to the factory o2 location and see if the issue goes away after I try unplugging the sensor. If not I will try a new sensor and report back.

[Vaughan]

On 5/24/2022 at 6:50 AM, 2.0L_turbo said:

I do know that there is a hole in the exhaust just a little before the sensor on a flex pipe

any hole in the exhaust pre sensor will introduce oxygen to the exhaust and make the sensor read leaner than the engine is running.

[2.0L_turbo]

1 minute ago, Vaughan said:

any hole in the exhaust pre sensor will introduce oxygen to the exhaust and make the sensor read leaner than the engine is running.

Ok I will move the sensor on Friday and try again

 

[2.0L_turbo]

On 5/23/2022 at 4:21 AM, Adamw said:

To confirm the voltage calibration try unplugging the sensor, Im pretty sure the AEM controllers show 1.00 Lambda with the sensor unplugged.  If that looks ok then I would suspect the sensor or possibly an air leak somewhere.   

I moved the sensor today from the where it was right before the hole in the exhaust, I put it where the factor narrow band sensor would go on the turbo elbow so it should read as accurate as possible. Unfortunately this made no change it still reads the exact same numbers. I tried unplugging the sensor as you asked and it does in fact read 1.00. I think this would confirm my suspicions that the sensor is dead but before I go out and spend 200$ on a new sensor, is there any other things I can try and is is safe to try and use sensor clean or some other product to maybe clean this thing ? when I took it out it was covered in soot so I know its running super rich.

[2.0L_turbo]

@Adamw

do you know of any ways of finding out the calibration for a VW LSU 4.9 Lambda sensor? The numbers on the sensor are 03G 906 262 A and 686 07 9 06

 

when I try this sensor it says I am running very rich which I can believe and if I remove fuel it does show up on the gauge but I need to go down to 10% or more to get any drastic change and I feel like the sensors calibration is just not correct compared to the sensor aem supplies. I will be ordering a new sensor as I feel this pretty much confirms a dead sensor but it would be cool to know if we can get these to work as well.

[Adamw]

If it has the correct plug on it then it is a lsu4.9.  The calibration is done by a resistor inside the connector, the calibration is done by Bosch so the sensors are interchangeable without any end user calibration 

[2.0L_turbo]

1 minute ago, Adamw said:

If it has the correct plug on it then it is a lsu4.9.  The calibration is done by a resistor inside the connector, the calibration is done by Bosch so the sensors are interchangeable without any end user calibration 

Interesting so then what would cause it to now be reading rich and barely changing to removing 20% fuel? It was hovering around .675 lambda, then when I removed the fuel could get it to see  .710 lambda. Just scared to remove any more incase it's not reading properly and I really lean it out. 

 

 

[Adamw]

Right now the only data you have is two sensors that read differently.  There is not enough data to tell you if one is correct, which one is correct, or if they are both wrong.  They only way you can determine that is with a known good reference.  This may be a new sensor or for example your tuners dyno wideband.  

[2.0L_turbo]

2 hours ago, Adamw said:

Right now the only data you have is two sensors that read differently.  There is not enough data to tell you if one is correct, which one is correct, or if they are both wrong.  They only way you can determine that is with a known good reference.  This may be a new sensor or for example your tuners dyno wideband.  

Sounds good Adam, I will see if I can track down a known good sensor and if not I will get one on order. 

[koracing]

As another suggestion: you can also try making enrichment adjustments at idle - you should hear the engine start to struggle if it is actually very lean or very rich at idle.  Run it all the way up to 1.2 lambda if you can and see if it pulls the same vacuum (MGP reading) as at 1.0 lambda, and as 0.8 lambda. 

[2.0L_turbo]

ok I tried a few more sensors and still the same result being too rich like my other sensor were saying. I now believe when its reading .675 - .710 it is accurate the cars exhaust also smells heavily of fuel. Here is a tune file and a short log I just took can anyone please take a look and see if you spot anything funny. I still feel like maybe the injector dead time is wrong and do not know if that could cause this, my reasoning for this is while the engine was running I changed the master fuel down. When I did this I was able to start seeing big changes in the afr, I was able to get it to 1.0 but master fuel had to be set at 5ms to achieve that. I did not save that to the ecu and then reset it back to 17 where the car liked to cold start.  I also tried to adjust the warmup enrichment table like @koracing suggested and it did not seem to do anything at all.

I appreciate the help on this everyone

https://drive.google.com/file/d/1LKwgedYwEwDHlaAg6fKSFMobinIS8lXt/view?usp=sharing

https://drive.google.com/file/d/168lZ9tgtUlGYP22V6mVuRwWZI9MTPBh8/view?usp=sharing

[koracing]

I don't think I said anything about the warmup enrichment table.  Enrichment is just another way of saying amount of fuel - I wasn't referring to a specific table.  What size are you injectors - and I'm assuming it's a 2.0L motor?  What fuel pressure are you seeing at idle?

[2.0L_turbo]

Just now, koracing said:

I don't think I said anything about the warmup enrichment table.  Enrichment is just another way of saying amount of fuel - I wasn't referring to a specific table.  What size are you injectors - and I'm assuming it's a 2.0L motor?  What fuel pressure are you seeing at idle?

My bad I figured when you said " enrichment adjustment at idle" you were referring to the warmup enrichment table, that's my mistake. But sadly as stated above I could pull nearly all the fuel out of it in the area where it idled and no change in the AFR. The injectors are fuel injector clinic 550cc low impedance with resistor packs, you are correct the motor is a 2.0L still and  base fuel pressure with the vac disconnected is 43.5psi I just put on my FPR the other night and set it.

[koracing]

The Toyota 3SGTE base map (ST205X) is a 2.0L motor with 540cc injectors from the factory (Denso low impedance with resistor pack) and it's using a 9ms master fuel number and about 28-30 VE at idle on the fuel table.  That might be a closer starting point for you.  To figure out if your dead time is way off you'll have to see if a 10% change on the VE table or AFR target actually results in a 10% change in lambda.  If it does not then the deadtimes are likely off.

[2.0L_turbo]

1 minute ago, koracing said:

The Toyota 3SGTE base map (ST205X) is a 2.0L motor with 540cc injectors from the factory (Denso low impedance with resistor pack) and it's using a 9ms master fuel number and about 28-30 VE at idle on the fuel table.  That might be a closer starting point for you.  To figure out if your dead time is way off you'll have to see if a 10% change on the VE table or AFR target actually results in a 10% change in lambda.  If it does not then the deadtimes are likely off.

do you mind if I send you a pm?

[koracing]

Feel free or message me on FB messenger or whatsapp.

[2.0L_turbo]

well the mystery issue has finally been solved it was in fact much to high of a master fuel level, the master fuel was set at 17 as that is where the car liked to start when it had the bad o2 sensor in the car. However after changing the sensor the car was now able to start with only changing the master fuel setting back down to 8. Now finally the car is responding to changes in the fuel map as it should and we have already taken the car around and began tuning its drivability but only under no heavy load, I am saving that for my tuner to handle I just need it drivable to get it to him safely. 

 

A massive thank you to everyone who has helped me on this, I love this community here and will always be a link ecu supporter and fan. I even managed to talk my buddy into getting a link for his 1990 eagle talon and we cannot wait to start getting that all set up!