A/F sensor, O2 sensor, trouble code strangeness - MR2 Owners Club Message Board
 
LinkBack Thread Tools
post #1 of 14 (permalink) Old March 26th, 2018, 19:30 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
A/F sensor, O2 sensor, trouble code strangeness

I am in the process of testing some new, larger injectors in my 2GR-FE (RAV4 with stock ECU). The testing seems to have brought to the surface some other issue(s) which need to be addressed.

Shortly after having replaced the injectors, I monitored the fuel trims, and was surprised to see that the Bank 1 long term / total trims were lesser by about 3% (~ -12% vs ~ -15% for Bank 2). Very soon, a pending trouble code emerged: P0136 (Bank 1 postcat O2 sensor problem). This code was followed by the code P0138 Pending (Bank 1 postcat O2 sensor reading rich). This code was followed by the code P2195 (Bank 1 precat A/F sensor stuck lean). I checked the data regarding the lean run, and it was in a 0-load situation, so I wasn't worried. However, it struck me as odd that in combination with the lesser trim, the postcat O2 sensor (showing rich, which makes sense since that bank is presumably using more fuel and running relatively more rich than Bank 2) was giving contradictory readings to the precat A/F sensor.

As a troubleshooting step, to make sure there was not a problem with one or more of the injectors, I pulled and inspected the injectors, then swapped them for both sides of the engine. I cleared the codes and the ECU memory, and started over. Same result, same codes, except I am not seeing the P0136 this time... so, presumably, not a problem with the injectors. I ran a test using the OBD-II device, and commanded the injectors to run rich, then lean, and didn't notice anything particularly unusual, other than slightly different readings between the sensors of the two banks, and that the O2 sensor which initially reported the problem did seem to get stuck reading rich at one point... other than that, it seemed to be working okay. My thought, based strictly on that observation, would be to replace that sensor, however I'm suspicious of the readings of the precat sensor as well, because it is presumably causing that bank to run lesser trims than the other bank, and thus command more fuel use. It's almost like the two sensors are fighting, with one reading rich, the other one reading lean, and I don't entirely know how the ECU uses them to define the trims, or know which one is "right"; Does it use both of them? If not, which one takes precedence? I would think the precat A/F sensor takes precedence, especially in open-loop use. So, is the precat A/F sensor reading leaner than it should be, thus causing that bank to run rich, that causing the postcat O2 sensor to read rich and fight with it in closed-loop?

3% seems like a big difference between banks. Installation of a 2-bank wideband has been held up, but these stock sensors should be sufficient for now, if I can solve this, since the precats read from 12:1 to 19:1. I don't just want to replace both sensors, unless of course they both happen to be defective, which seems unlikely.

It seems relevant that the P0138 remained pending, since that suggests the ECU was still using it for AFR, meaning the postcat sensor could be offering data which conflicts with the precat sensor (rich vs lean).

Thoughts?

Last edited by biyanpian; March 26th, 2018 at 19:38.
biyanpian is offline  
Sponsored Links
Advertisement
 
post #2 of 14 (permalink) Old March 26th, 2018, 19:57
Gold Level
 
Join Date: Sep 2005
Location: Fort Wayne, IN
Age: 38
Posts: 3,446
OldTrader Rating: (14)
I don't think you're going to find many if any people here running the secondary O2 sensors because the cats simply do not fit in the MR2 application so it'll be hard to give you feedback on this stuff. but as far as i know the secondary O2 sensors are not factored into the normal engine operation.

I wonder if these injectors have a different opening time. this could cause them to run lean at one end and rich at the other depending on the amount of fuel expected and the difference in opening times. the difference would be most pronounced at idle.
Gouky is offline  
post #3 of 14 (permalink) Old March 26th, 2018, 20:25 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
Quote:
Originally Posted by Gouky View Post
I don't think you're going to find many if any people here running the secondary O2 sensors because the cats simply do not fit in the MR2 application so it'll be hard to give you feedback on this stuff. but as far as i know the secondary O2 sensors are not factored into the normal engine operation.

I wonder if these injectors have a different opening time. this could cause them to run lean at one end and rich at the other depending on the amount of fuel expected and the difference in opening times. the difference would be most pronounced at idle.
Thanks for the idea. I figured the ECU might learn battery offsets, but that is probably wishful thinking. On the other hand, I don't see definite evidence of this, yet, and the problem seems to be with one bank only, which differs from the other even after swapping the injectors. The fact that it is only affecting one side, and is apparently producing consistently lesser trims, makes me suspicious of the sensors. The testing I just did was at and near idle, and the trims seemed consistent. Definitely something to keep in mind, though, and it might explain the "stuck lean" situation if it pops up on both sides and isn't related to the downstream sensor.

The point about the downstreams not being used for trims would surely simplify things if it is the case. My reading says that they are used, though...supposedly Toyota has been doing it for a long time. In between posting the problem and now I did some quick searching, and found this (anecdotal) info describing the same thing, at acurazine.com:

"The post cat O2S is a fuel control input. Toyota and Saab have used the
post-cat O2 to control the fuel trim since 1988; Before OBD-II. This sensor is used to fine-tune the air-fuel ratio to maximize catalyst efficiency. It can also adjust the air fuel ratio to compensate for a degraded Catalyst. If you were to measure the actual amount of time the post-cat O2 is used during the running of the catalyst monitor (maybe once per trip) and compare that to the amount of time the post-cat O2 is used fine tune the air-fuel ratio (almost always in closed loop), you will see that this sensors major function is fuel control. Every vehicle manufacture today uses the rear O2S for fuel correction; even if it is undocumented.

a.. How much control over fuel trim does the rear O2S have?
a.. GM material says less than 1.0%
b.. Ford says 0.5%
c.. Toyota says 2.0%
d.. Under normal conditions. We have seen up to 30% correction in fuel trim from the rear O2S in abnormal conditions"

I was thinking it would make sense if one was reading very differently from the other, they would fight, specifically in closed-loop feedback. If the downstream is reading very rich, it is likely to affect things, especially in emissions-centric operation as things are now.
biyanpian is offline  
Sponsored Links
Advertisement
 
post #4 of 14 (permalink) Old March 26th, 2018, 20:36
Gold Level
 
Join Date: Sep 2005
Location: Fort Wayne, IN
Age: 38
Posts: 3,446
OldTrader Rating: (14)
It's certainly possible, while i do have a decade of experience hacking around with this engine i have never worked for Toyota or have had any access to their internal documentation or code.

At the end of the day the rear O2 sensors are significantly lower resolution than the front wideband AFR sensors so there's not a ton that can come forward. I really think this type of documentation is a bit outdated from the times when cars would run front and rear narrow band O2 sensors but i have no good way to prove it.

as for the ECU learning opening offsets it learns enough to compensate for fuel trims and this will cancel things out enough.

but please do more testing, they is way more to be learned on this stuff and i appreciate you sharing what you are seeing.
Gouky is offline  
post #5 of 14 (permalink) Old March 26th, 2018, 20:58 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
The thing that first struck me as odd was the difference in trims. I have been keeping an eye on it, and it's been consistent.. So, even without the codes, it's a 3-4% difference in fuel overall, on the same bank, both before and after swapping sides with the injectors, and through multiple ECU resets. I could have done something wrong in seating the injectors, but it's a mistake I'd have to have repeated, and I tried to be careful.

Codes aside, does that suggest anything to you, in conjunction with the downstream sensor (the side presumably using more fuel) reading very rich?
biyanpian is offline  
post #6 of 14 (permalink) Old March 26th, 2018, 21:06
Gold Level
 
Join Date: Sep 2005
Location: Fort Wayne, IN
Age: 38
Posts: 3,446
OldTrader Rating: (14)
Differences in trim between the banks is pretty normal operation, especially with the stock exhaust. a 3-4% in fuel trims between the banks is nothing to be concerned about.

There are things in the ECU that survive a reset, i'm not sure what the complete list is but there appear to be some stuff related to the DBW learning and likely other things also. If you're willing to send me your ECU i can reset that part of the memory. I could also clone your ECU and only reset the clone in case you're concerned about keeping the factory one.

This would be a potential learning opportunity for everyone if you're willing.
Gouky is offline  
post #7 of 14 (permalink) Old March 26th, 2018, 22:03 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
That is a kind offer. I'll see if I can find a cheap-o on Ebay, and perhaps have it shipped to you? I have a second car, so I could drive that for a bit. Do you have a quick list of ones I could look for which would serve the purpose?
biyanpian is offline  
post #8 of 14 (permalink) Old March 26th, 2018, 22:05
Gold Level
 
Join Date: Sep 2005
Location: Fort Wayne, IN
Age: 38
Posts: 3,446
OldTrader Rating: (14)
You'll want to match whatever part number you currently have on your car. Unfortunately because of the position in the chassis the rav4 ECUs get wrecked very easily in a light front end crash so an ECU with an intact connector generally runs $150-200
Gouky is offline  
post #9 of 14 (permalink) Old March 27th, 2018, 13:58
Registered User
 
Join Date: Feb 2016
Location: Maui, HI
Posts: 1,148
OldTrader Rating: (0)
The difference between banks is within the usual range. I monitor both banks and see a difference of a few percent. Recently looked at logs from Gavin's car and saw a 2% difference.

I don't think anybody really knows the function of the postcat O2 sensor beyond monitoring the cat function. As you know on most of our hacked ECU's the postcat sensor codes are masked along with the transmission codes. Definitely the ECU can maintain closed loop control with the postcat sensors disconnected and this clearly means they are not essential for fueling control. As an experiment just disconnect them and compare logs before and after. On every other modified car that I've had (and many stock vehicles that I service) I use a postcat O2 spacer to inhibit the postcat codes P0420 P0421 without any detectable effect on runnability.

I have checked the official 2GR-FE repair manual for diagnosing and troubleshooting P0136 and P0138. According to these materials (which are often very different from generic online OBD2 reference) these codes most likely refer to sensor and wiring issues NOT fueling issues.

You might take a look at the more detailed troubleshooting sections of the BGB for these codes, and reassess your interpretation of what they are telling you.

If you need the relevant sections of BGB I can provide them for you and also you should be able to download them from other sources.
merryfrankster is offline  
post #10 of 14 (permalink) Old March 30th, 2018, 14:02 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
I was using the troubleshooting manuals for the upstream and downstream codes using Techstream, run the fueling override modes at +25% and -12.5%, monitoring sensor output voltage, which still left me scratching my head... That's what led me to posting, since the only real anomaly I noticed was that the Bank 1 downstream sensor seemed to be stuck reading rich at one point, for an extended period (.8V) when by all appearances it should not have been (I had moved the commanded fuel back to normal from rich), unless the upstream sensor had been giving falsely lean feedback, causing the bank to run very rich, and the downstream sensor was correctly sensing unburned fuel which was passing through the cat or something similar (unlikely?...since it only seems to be about 4% different from Bank 2, in general). I then cycled commanded fuel again and it went back to reading what looked to be normal. This is what has me confused... It seems like one of the sensors is faulty, but either one could be causing the problem. I posted the issue on TN, too, because the more educated feedback the better in this case, and the only response was with suspicion of the upstream sensor. I suppose I could replace them one at a time, starting with the less expensive downstream one.

I am also tempted to take Gouky up on the ECU cloning with data wipe on one, especially considering the info about retained DBW info (I thought it was just transmission data) and the possibility of adding to general knowledge.

I am leaning more and more toward one aftermarket wideband sensor for each bank, rather than a single one for both. Thoughts on this?
biyanpian is offline  
post #11 of 14 (permalink) Old June 20th, 2018, 11:45
Registered User
 
Join Date: Mar 2004
Location: Victorville, CA
Age: 36
Posts: 4,625
OldTrader Rating: (4)
Did you put the AFR sensors on their own relay and fuse? Mine tend to draw 5.5A each in normal operation, and that 11A is not safe to have on a single 15A EFI circuit.

Though the newer planar AFR sensors do draw less power, it's still a good idea to allow them to draw as much as they need, esp. on cold start. 2GRs enter closed loop within 15-20s, while my 1MZ with 1st gen AFRs takes 30-40s.

As for bank-to-bank fuel trim differences, that's usually okay because each cylinder will produce a variation of combustion between each other due to small tolerance differences in the cylinder walls/pistons/head/ports. Without flow matched fuel injectors too, it's hard to put that down to anything other than that bank producing a tiny bit more power than the other, so it needs slightly more fuel due to slight underfueling/leanness. 3% is pretty good, actually.

Post-cat O2s can be used for fuel adjustment. It's a small amount since the MAF determines initial fuel enrichment based on measured airflow, which AFRs adjust to compensate for any under/overshoot of MAF reading and associated fuel injection. When the cat is lit-off, the post cat O2s should read pretty rich between 0.700v and 0.875v and should NOT mirror any AFR sensor adjustments. The reason it reads rich is because the catalysts store any oxygen in the exhaust, so the post-cat O2 sensors read that as rich (low O2).

If you log the post-cat sensors, they should have a steady voltage output with good cats. Once a converter degrades, it becomes erratic until it mirrors pre-cat AFR adjustments (O2 sensors output inverse voltage of AFR sensors), which indicates a blown cat.

Now on deceleration, all AFR sensors will read full lean, and post-cat O2 sensors may have no output or full rich (0.9-1.0v). I've logged my AFRs and they do go to 5V (pure air). Here's the tricky part: the ECU will time the AFRs to detect any deterioration in the sensors. If a sensor doesn't switch back on within an allotted time after fuel injection resumes, you get a code. Stuck lean does indicate a sensor issue. Typically, I only use Denso AFR sensors, as those are OE for Toyota. If it's brand new, have it replaced under warranty if it's still covered. If it's used, replace it or swap the bank's sensors around to be absolutely sure.

It's also a good idea to have high quality ground connections from engine to chassis. AFR sensors are ECU grounded, so ensure the ECU main ground and ECU sensor ground are in good shape too. That's always a good idea.
Jason.MZW20 is offline  
post #12 of 14 (permalink) Old July 3rd, 2019, 13:01 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
Apologies for not having followed up on this...making me guilty of one of the larger forum annoyances (not posting potential solutions to problems).

I did resolve the problem. I suspected the upstream AFR sensor because of the contradictory sensor readings, so I replaced it with a new sensor (bought one for a Camry, which was much less expensive, for what it's worth). No more codes.

Granted, I am running larger-than-stock injectors (~440cc), but for the engine to be relying so heavily (for fueling feedback) on a part, it should either have redundancy or speedy detection if it is running out of specification. I have installed an aftermarket wideband AFR sensor on each exhaust manifold for peace of mind.
biyanpian is offline  
post #13 of 14 (permalink) Old July 3rd, 2019, 14:53
Registered User
 
Join Date: Feb 2016
Location: Maui, HI
Posts: 1,148
OldTrader Rating: (0)
Quote:
Originally Posted by biyanpian View Post
but for the engine to be relying so heavily (for fueling feedback) on a part, it should either have redundancy or speedy detection if it is running out of specification. I have installed an aftermarket wideband AFR sensor on each exhaust manifold for peace of mind.
What are you talking about? The list of AFR sensor related DTC's is about a mile long. A code gets set if you so much as look at a sensor wrong. You can also detect potential problems by just looking at the trends and patterns in your AFR's and FT's on a scantool, even if a code is not set. Because of emissions and fuel efficiency requirements, the fueling system is at an amazing level of refinement, far beyond what you can achieve with aftermarket components. Maybe we need to send you back to school on this.
merryfrankster is offline  
post #14 of 14 (permalink) Old July 7th, 2019, 16:54 Thread Starter
Registered User
 
Join Date: Apr 2015
Posts: 22
OldTrader Rating: (0)
Quote:
Originally Posted by merryfrankster View Post
What are you talking about? The list of AFR sensor related DTC's is about a mile long. A code gets set if you so much as look at a sensor wrong. You can also detect potential problems by just looking at the trends and patterns in your AFR's and FT's on a scantool, even if a code is not set. Because of emissions and fuel efficiency requirements, the fueling system is at an amazing level of refinement, far beyond what you can achieve with aftermarket components. Maybe we need to send you back to school on this.
If that wasn't a rhetorical question, I'm talking about redundancy so as to more easily track down (or identify outright) the defective part in a case like this, without throwing a lot of time and/or $ at it. Refined or not, the system did not clearly indicate the bad part, nor did the provided data reveal it (to me, at least).

I did spend a lot of time using a scan tool and looking at fueling trends and related real-time and log data (and I did state that in this thread), which didn't reveal anything conclusive to me. In the end, after having eliminated injectors as the problem (which took a long time), I reasoned out the most likely cause of the problem and it happened to be correct... but, this was after having spent a long time in troubleshooting and study.

If I read your comments as they were intended, I see a comparison between the respective capabilities of OE and aftermarket solutions. I am also impressed by the OE capabilities, even in a boosted application for which it was presumably not designed, and I don't intend to replace it outright, or to any significant extent, by aftermarket equipment. The aftermarket stuff fills different roles: ability to look at AFRs and manifold pressure for practical reasons and fun, and for providing data to a piggyback system which is intended to compensate for having larger-than-stock injectors.

Edit: I should also mention that when I found the AFR sensor (Denso 234-9008), through Googling it I found a post by you also referring to it as a possible replacement part, and that contributed toward the decision to give it a shot. So far, it seems to be working fine. When I compared it to the stock one, as i recall, the stock one stuck out a little further into the exhaust manifold (the tip stuck out further from the threads).

Last edited by biyanpian; July 7th, 2019 at 17:06.
biyanpian is offline  
Reply

Quick Reply
Message:
Options

Register Now



In order to be able to post messages on the MR2 Owners Club Message Board forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.

User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
Please enter a valid email address for yourself.

Email Address:
OR

Log-in











Currently Active Users Viewing This Thread: (0 members)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page



Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

 
For the best viewing experience please update your browser to Google Chrome