System too lean error codes

[james250]

1 hour ago, johnatg said:

Your calculation of the difference in our lambda readings is not valid - there is a world of difference between 1.000 and 0.997. 1.000 indicates exactly stoichiometric combustion - 0.997 is right on the limits of spec but indicates a rich-ish mixture - ie too much fuel for the oxygen in the cylinders. (And your high-ish HC reading concurs). I know your error codes are showing too lean, but your lambda is showing too rich (if it's correct) That can be caused by a variety of things - eg blocked air filter, slight misfire - plug or coil, faulty temperature sensor (the one feeding the ecu is not the same as the one which drives your temperature gauge), etc.

I think you need some tuning diagnosis rather than relying on error codes, because they are all over the place. Maybe try a mobile tuner with an oscilloscope - that can also show the actual output from the O2 sensors.

Are you using Techstream to retrieve the codes, or a generic code reader?

The emissions are not wildly out (and indeed are within MoT specs), so the problem is likely to be minor and intermittent at the moment. I can't explain the error codes reading as they apparently do, but they can sometimes be quite misleading.

Ahhh so it seems to be getting more complicated. Because you are saying the Lambda readings indicate too rich not too lean but the error codes coming back are related to too lean. Air filter is new. Spark plugs are new only about 2k miles on them. I'm not using techstream, i'm using a wifi obd ELM312 with DashCommand App.

 

1 minute ago, normski2 said:

Sorry to read of your problems. For me it was just plug and play and off I went.

Couple of things spring to mind...but they are just thoughts...I make no claims of expertise.

• Are you sure you got all those connectors to the ecu (under the battery tray) all securely refitted? Some of them are right little toads to get at ...no bent pins anywhere on them?
• Don't think you would have ... but did you disconnect the Mass air flow meter or the piping to it and the airfilter box? If you did ... is that big circlip tightened up again? Is the plug attached securely to the MAF meter sensor. 
• May be worth giving the MAF a clean if you have the correct solvent spay available.

I suspect your O2 sensors are OK so try and double check all your re-connections if you can.

Best of luck with it all.

@normski2 thanks for your reply. all the connectors i'm sure are fine. those connectors were a pain but i made extra sure i got them in right. I didn't disconnect the MAF, but I do have a spare second-hand MAF i think sitting in the loft with a complete airbox which I bought second hand a few years ago because i was planning to do a mod but never got round to it. Might be worth trying that second-hand MAF?

[james250]

I get the feeling maybe it is the O2 sensors. They might be fake afterall. Although the seller has sold lots and has lots of good reviews. I have found one review from a buyer on his review that says it's fake and garbage. So now i'm wondering maybe that's the problem. 

[james250]

Is there a way for me to test these sensors if I remove them? I could put the originals back in and see if I get the P2197 code

[johnatg]

Is there a way for me to test these sensors if I remove them? I could put the originals back in and see if I get the P2197 code

AFAIK the only way to test the O2 sensors properly is in situ with an oscilloscope.

Sent from my PSP7551DUO using Tapatalk

[james250]

Any tips for if I am suspecting the sensors being fake and causing issues? should I put my originals back in? With the originals I only had P0174 and P0171 errors not the P2197. Or should I try my second-hand spare MAF sensor as that would be the easiest thing to replace.

[johnatg]

Any or all of those might be a good idea - and or get the new sensors tested with an oscilloscope - after all, you've implied that you think the old sensors are past their best anyway.

[james250]

If it does turn out to be the sensors, i guess it would be an expensive lesson to learn. So just incase it is I am looking for genuine Denso ones. I have the OEM part number from Lexus as 89467-30010. The original denso one on RockAuto has a denso part number of 234-9051, if you click on more info on it and it shows the equivalent OEM number which lists the number lexus gave me above (without the hyphen) http://www.rockauto.com/en/moreinfo.php?pk=1606992&cc=1432808&jsn=344

However, I found a EU Denso parts website, the website is: http://www.denso-am.co.uk/e-catalogue/ 

If you go on that website, and type the part number that Lexus gives it gives the equivalent denso part number as DOX-0263. This denso part number is completely different to the one RockAuto sells. This got me thinking, are the american denso sensors different to the europe denso sensors? If you go into the info for the DOX-0263 and click OE numbers, it gives you a list of OE numbers and the Lexus one is listed there without a hyphen. 

[james250]

11 minutes ago, johnatg said:

Any or all of those might be a good idea - and or get the new sensors tested with an oscilloscope - after all, you've implied that you think the old sensors are past their best anyway.

The old sensors have definitely done way too many miles over 120k miles 

[Shahpor]

Sorry if you have mentioned this already, but are you referring to the Air/Fuel ratio sensor or Heated oxygen sensor?

[james250]

3 minutes ago, Shahpor said:

Sorry if you have mentioned this already, but are you referring to the Air/Fuel ratio sensor or Heated oxygen sensor?

The upstream air/fuel ratio sensors

[Shahpor]

Well, I am not sure if you have already tried this, but the manual states the following:

 

Also, for testing the sensors whilst driving, doesn't the TechStream software give you live data readouts for the air/fuel ratio?

[james250]

@Shahpor thanks. I can maybe try that resistance test on the old sensors. I don't have the TechStream software but considering getting one of those off the popular internet auction site. What kind of thing should I be looking for when using techstream to view live data? will TechStream live data help determine if sensors are malfunctioning? would be a cheaper option than just buying new sensors

[Shahpor]

Well, it has been a while since I loaded mine up, so I can't be sure what parameters the live data stream shows.

I did find this screenshot which shows the voltage readouts for the sensors:

 

According to the manual, they should read about 3.3V with a fluctuation of no more than 1V:

 

Unfortunately, I don't think the Techstream software shows the actual waveform.

[james250]

Thanks @Shahpor great piece of info. So technically the perfect reading should be 3.3V? So maybe it's worth me investing in a techstream fro the internet auction site. That way I can at least read live data without buying new sensors yet and if it indicates the sensor issue then i guess I can just buy a new set.

In regards to the techstream on the auction sites, i read somewhere that only certain ones work. is there a certain version I need to look for?

[Shahpor]

Well, 3.3V seems to correlate to a 14.6:1 ratio, which would be the ideal.  There will be some variations in real time, but like it says it shouldn't exceed 1V.

As for the software, to make life easier you should get one with the 64bit drivers included.  Something like this:

http://www.ebay.co.uk/itm/16pin-Mini-VCI-Toyota-Lexus-TIS-TechStream-OBD2-Diagnostic-Interface-Scanner/112063076459?epid=714481939&hash=item1a177ad86b:g:Vw0AAOSwqfNXj555 

[james250]

That's great. Thanks @Shahpor I think the Techstream is worth the investment.

This might sound like a silly question but the air/fuel ratio sensors the air/fuel ratio after the combustion, so technically by this time it has already been too lean or too rich. So wondering, if at this point it has already done the combustion based on MAF and other things and after the combustion the air/fuel ratio sensors are determining that the combustion was too lean or too rich. Other than throwing up a code how does it help actually solve a rich or lean issue? does the ECU/ECM constantly adapt from the readings it gets?

[Shahpor]

Yep, you got it already.

The ECU has a lookup table that it uses to trim the fuelling based on the readings from the O2 sensors in the exhaust, along with data from the MAF, throttle position, temperature, etc..

It is this table that is altered when someone remap an ECU.

The ECU has a specific range to work from, so when one of the sensors starts producing data outside that range, it throws up an error (Engine management light and Check VSC in this case).

[james250]

So if the sensor sends signal back saying the combustion was too lean or too rich the ECU/ECM is supposed to make adjustments to try and make the combustion optimal and if something is to not work for example a malfunctioning sensor, or MAF problem, vacuum leak etc then the readings will continuously be wrong leading to a engine management light etc is that correct?

[Shahpor]

Pretty much.

For example, the O2 sensor is reading too lean, so the ECU adds more fuel.  The problem is that, after adding fuel the O2 sensor still reads too lean, so the ECU adds even more fuel.  Eventually, the ECU will get to the end of the lookup table, with the O2 sensor still saying it is too lean, so it will throw up an error code to signal that it doesn't know what to do anymore.

There are other things it can do as well, such as adjust the ignition timing, but that is beside the point here.

Using the O2 sensor in this way is called a closed loop engine management system.

A better description can be found here:

http://www.enginebasics.com/EFI Tuning/Closed Loop o2.html

[james250]

@Shahpor that's great info thanks. I am waiting for delivery of the techstream when that comes i will look at the live data and see what's going on. Is there anything else I should be looking at in the live data other than the sensor voltages?

[Shahpor]

Not that I can think of at the moment.

Once you get the software you will see it is quite comprehensive, so there are many thing you can check if you want to.

Post of screenshot of the live data page when you have it running and we can see if anything looks abnormal.

[james250]

Thanks @Shahpor I am still waiting for delivery of the TechStream, should arrive sometime next week. Once I get it I will post a screenshot. Should I run the live data at a certain RPM and once engine has warmed up or run it straight away even if engine is cold?

[Texas]

See this from above,

3.3v +/-1v ( so 2.3-4.3v) @2500 rpm with warm  engine  (sensors)

 

 

[james250]

2 minutes ago, Texas said:

See this from above

 

Oooops yes silly me how did I miss that LOL

[james250]

My techstream finally arrived. Somehow went missing in the post so had to be sent out another one. Going to install it this weekend and try it out. I just came across the following tech diagram that also refers to diagnosing the AF ratio sensors. But it for some reason has different instructions to the one @Shahpor posted. So it's a bit confusing.

Based on the diagram below, am I correct in understanding that the procedure is asking to start the engine and run the rpm at 2500 for 90 seconds then go to the relevant section in techstream and carry out the 3 scenario tests that it lists. Based on the standard voltage table, the reference column says that scenarios 1 and 2 which is running idle and running at 2500rpm should both be between 3.1v and 3.5V. So that's a fluctuation of 0.2V each way. and scenario 3 should apparently see the voltage increase to 3.8V or more and then drop back. Is this understanding correct?

Also, the second diagram i have attached is of a table that shows Lean and Rich voltage. It shows if voltage is 3.0V then it's rich and if 3.35 or more then it's lean. But that doesnt seem to match the first diagram. Or is the second diagram referring to if the ECM injects +25% fuel then in that condition the voltage of 3.0v or less will be Rich. And if ECM injects -12.5% and a voltage reading is 3.35 or more then it's a Lean situation? I guess we don't have a way to know how much fuel the ECM is injecting to do this comparison do we with techstream?