Running Rich-Pico Scope questions-Head Scratcher

[liggity]

Hey everyone, first post here. I've really gotten a lot of help from this forum whenever I had a problem to look up. Thanks.

Got a question and I coudn't find anything close in the forum to mine.
My ej25 is running rich. My AF Ratio sensor is running at 1.5 volts which is clearly extremely rich. I ran the entire engine through every test I could think of and even some that wouldn't effect it. I have no misfires and no codes at all. Just been noticing that I was getting about a drop of 6 mpg before I decided to check out the fuel trim and O2/AF Sensors. I'm also getting backfires while at a cruising speed. Never stomping down on the throttle during the backfires. It's usually while I just barely am on the throttle and then after I take my foot off and reapply it backfires.

First off, the car only has Borla headers and axle back borla exhaust. And I have a Perrin lightweight Crank Pulley. Nothing else has been modified in the engine.

Everything I've done to the car in the last 3 months: I have run a motorvac and changed the fuel filter. Ran some Sea Foam through the vacuum line to my brake booster. (That was a whole lotta nasty smoke comin out the exhaust after that). Did a dry compression test, leakage test, and a power balance test. I was getting some knocking out of the timing belt tensioner so I replaced that along with a whole timing belt kit and water pump. I realized the car was off a tooth on the passenger side cam and thought that might have something to do with it, but alas it didn't. I did decide to throw a part at the car and went ahead and just replaced the AF ratio sensor, but that did absolutely nothing. Rear 02 is fine and the cat is fine. All vacuum tests were great too.

Hooked up the Pico and ran the trigger line on the 1st cylinder to tell which one was cyl 1. So at an idle all the waveforms looked pretty normal and the same, except for cyl 3. It had weaker "hills" and "valleys" The strange thing was that when I opened the throttle a bit up to 3000 rpm cyl 3 ran great and cyl 4 started having the abnormal waveform. Now I am thoroughly confused.

Any help would be great! This cars driving me nuts.

[mike-tracy]

Hi Liggity, wish I knew the answer. I had some of the symptoms you're having, but I didn't check the cylinders with a scope. Anyways mine was definitely caused by false knock, and I found that replacing my knock sensor (it was cracked), problem got better, then the transmission bushings (which was so bad I could twist it by hand), and that solved it.

Also we don't allow pictures in your signature, so please remove yours.

[02_Legacy]

You need to hook up a computer to your car too see what the fuel trims are. create a large vacuum leak and see if your a/f sensor goes leaner, and then run some propane or similar gas into the intake and see if the sensor reads richer, it will at least let you know if it is responding. Also when you say that your a/f ratio sensor is running at 1.5v do you mean the stock one or do you have an aftermarket one?

[liggity]

Also we don't allow pictures in your signature, so please remove yours.

Sorry bout that. Removed it.

[liggity]

You need to hook up a computer to your car too see what the fuel trims are. create a large vacuum leak and see if your a/f sensor goes leaner, and then run some propane or similar gas into the intake and see if the sensor reads richer, it will at least let you know if it is responding. Also when you say that your a/f ratio sensor is running at 1.5v do you mean the stock one or do you have an aftermarket one?

Used a Snap On Modis and the fuel trim was trying to lean it out due to the rich condition. I'll have to hook it up again to get the actual numbers.
Did that to the stock AF sensor but decided to change it anyways even though it was reacting to the test. Ran a smoke machine to check for any vacuum leaks too....nothing.
As for the 1.5 V reading, that's what I was getting with the original and the new one is a Bosch AF Sensor. The only other thing I have done was check the coils primary's resistance and the secondary also. I was able to find a testing procedure to test both. Between the 1-2 primary I got 1.18 Mega Ohms and the 3-4 I got 274 kilo Ohms. I was told they should read 0.73 Ohms +/- 10%. As for the secondary I got 14.2 Ohms for both 1-2 and 3-4. Seems to be within the specs (12.8 Ohms +/- 15%).

[02_Legacy]

Very few things will cause a constant rich condition, leaking fuel injector, and higher than normal fuel pressure are probably the most common problems. A much more common problem is a extra 02 into the exhause from one or more cylinders causing all the other to run rich, a misfire is an example of this. An ignition problem may cause the A/F sensor to read lean and fuel trims to add fuel causing a rich condition, however to my knowledge I do not think that an ignition problem could actually cause the engine to run rich.

[liggity]

Yeah, it's kind of a seperate issue I guess. When I had the Pico hooked up I noticed cyl 3 had a strange wave form at idle and then it changed to cyl 4 when at 3000 RPM.
But yeah, my final attempt at finding the problem was just going to be replacing the injectors or getting mine cleaned and checked. When I did a motorvac though I noticed the fuel pressure was perfect and held great. My guess is that one injector just isn't atomizing the fuel as well as the rest.

[02_Legacy]

To tell you the truth i don't know what you are referring to when you say pico scope, but I think you are looking into it too much, some noise or imperfections are expected with most equipment.

And for your running rich problem before you do anything too expensive I would confirm your suspicions with a true aftermarket A/F ratio sensor in the exhaust or get an actuall A/F reading off the stock sensor rather than a voltage. The best thing I ever did was download ECU tracker and buy a cheap $15 usb vagcom cable, it allows you to read all the data from the ecu using SSM protocol, it updates 100 times faster than the generic OBD protocol and allows you to read A/F ratio which you cannot do with the generic protocol. Check it out:
http://www.mympgtracker.com/ecutracker/index/
http://www.rs25.com/forums/f72/t126475-long-last-ecutracker-alpha-ready-obd2-ssm-sensor-display.html

[liggity]

A Pico Scope is an Oscilloscope that takes a "picture" up to every Pico-second. I only had it set to every 200 ms. When set to Pico-seconds, it picks up so much noise from all the electrical components in a vehicle that I can't tell what I'm looking at. Yeah, ever since I started working on cars I've lost the touch to hear the little noises. I just started off using the Pico connected to the Exhaust Pressure Transducer and went from there. I was chasing the backfire problem.

Problem is that I was getting the exact same readings from the original AF sensor as the new one. The Modis is definitely not generic. $8K scan tool that does anything and everything. I just like to read the voltages for all O2 sensors. Really the voltage reading is telling you the same thing as the ratio. The AF Ratio sensor just happens to be just the opposite of a regular O2 sensor in that the lower the voltage in an AF sensor, the richer the the mixture where as an O2 sensor with low voltage is running lean. A perfect reading in my car should be 3.5 V at idle. Anything above 4 volts is running lean. So at 1.5 V my car is running extremely rich. It also gives it in a percentage. It's just that I understand the voltage readings better than the actual ratio. Just a preference.
I do like what I've read about the ECU Tracker in your link though. I'll have to take another look. If I can just get the stinking readings to get a lil closer to 3-3.5 V in my AF sensor and want an exact fuel to air ratio I'll have to get something like that.

If I can't figure anything else out, I'm just going to have to flow test the injectors. Place over here that does it for $20 an injector.

By the way, I dig the hell out of your Legacy. You got some skills.

[mlitz]

sounds like a bad injector to me... try pulling the connector off each injector one at a time and try to narrow down the faulty one.

[Garrison]

^^^ yeah

Injectors or front o2 sensor would be my bet.

[02_Legacy]

The modis is the very definition of a generic scan tool in that it works with any and all cars. Each manufacturer has their own scan tool that will only work with their cars, these are much better than generic scan tools because they often have access to more information at a faster upload speed. An example of a few manufacturer specific scan tools are GM's tech 2, Ford's IDS, and Subaru's Subaru Select Monitor (SSM).
Also I did a little looking up on the front A/F ratio sensor for Subaru's, all the following information is from the FSM for my car (should be the same as yours). First of all there are 4 signal wires from the sensor, each of which send different parameter signals. Secondly the signal is measured in mA, not voltage, and third, the lower the signal (in mA) the leaner the mixture with a higher signal meaning richer. You see what I mean when I say you have to go off of the A/F ratio that the computer is calculating? Unless there are other signs of your car running rich, I would hold off throwing money at it until you confirm it is in fact running rich.

And thanks for the complement, just know that I am not ragging on you or saying you are wrong about anything, I'm just throwing my opinion out there on how I would go about the problem.


try pulling the connector off each injector one at a time and try to narrow down the faulty one.
This would not help because as soon as you pull the injector connector it would drive the sensor lean no matter if it is the bad cylinder or not because not that cylinder is pumping all kinds of oxygen into the exhaust.

[mlitz]

I guess you are right... now that I think about it when ever I have had a bad injector my cars runs like garbage at an idle so I just pulled each connector to see what cylinder did not make the idle worse, and bingo there is my injector.

[liggity]

I do know about the vehicle specific, I just consider completely generic scan tools to be the Innovas and different cheap scan tools bought at Schucks for $100-200. The Modis does different thinks for different makes and gets pretty specific as I'm sure you know. I know tools like the tech 2 get really specific though like you're saying. I actually especially love the way the tech 2 can turn off and on the heaters in the heated O2 sensors to make sure those are turning off and on without reaching up into ridiculously tight spots to get the pre cat O2s in some GMs. But besides the output controls I have noticed no difference between the modis and brand name specific scanners like the tech 2. The response to with lookin at the fuel trim and o2 sensor readings seem to be the same. But I have only used Fords and GMs, so I don't know about any others to comment on them.
But with the AF Ratio Sensors, it actually does read in voltage. It's extremely hard to test it in mA. I actually just finished up my course in school on Fuel Systems and we spent a good amount of time on oxygen sensors. Here's a lil bit from Wiki.

Testing
As the ECM monitors the varying current, a special circuit (also inside the PCM) converts the current into a voltage value and passes it on to the serial data stream as a scanner PID. This is why the best way to test an AFR sensor’s signal is by monitoring the voltage conversion circuitry, which the ECM sends out as an AFR-voltage PID. It is possible to monitor the actual AFR sensor varying current, but the changes are very small (in the low milliamp range) and difficult to monitor. A second drawback to a manual AFR current test is that the signal wire has to be cut or broken to connect the ammeter in series with the pump circuit. Today’s average clamp-on ammeter is not accurate enough at such a small scale. For this reason, the easiest (but not the only) way to test an AFR sensor is with the scanner.
By using a scanner to communicate with the ECM, one can view AFR sensor activity. This data is typically displayed as WRAF, A/F, or AFR sensor voltage. However, on some vehicles and scanners it will show up as "lambda" or "equivalence ratio." If the PID displays a voltage reading, it should be equal to the sensor's reference voltage when the air/fuel mixture is ideal. The reference voltage varies from car to car, but is often 3.3v or 2.6v. When the fuel mixture becomes richer (on a sudden, quick acceleration), the voltage should decrease. Under lean conditions (such as deceleration) the voltage should increase.
If the scanner PID displays a "lambda" or "equivalence ratio," the reading should be 1.0 under stoichiometric conditions. Numbers above 1.0 indicate a lean condition while numbers below 1.0 indicate rich mixtures. The ECM uses the information from the sensors to adjust the amount of fuel being injected into the engine, so corresponding changes in the short-term fuel trim PID(s) should also be seen. Lean mixture readings from the AFR sensor will prompt the ECM to add fuel, which will manifest itself as a positive (or more positive) short-term fuel trim percentage.
Some technicians will force the engine to run lean by creating a vacuum leak downstream from the mass airflow sensor, and then watch scanner PIDs for a response. The engine can be forced rich by adding a metered amount of propane to the incoming airflow. In either case, if the sensor does not respond, it likely has a problem. However, these tests do not rule out other circuitry problems or ECM issues.

So yeah, it's for sure running rich. And the O2s are ruled out for sure. One question: I did a fuel pressure test and everything was good....should I do a fuel volume test to check the amount of flow? I can't remember if thats going to tell me much for a rich condition. Usually that will just tell me if theres any obstructions for a lean condition right?




And thanks for the complement, just know that I am not ragging on you or saying you are wrong about anything, I'm just throwing my opinion out there on how I would go about the problem.



No doubt man. I'm not taking it that way at all. Just a conversation back and forth where you're trying to help me narrow it down. Hard to tell tone in my typing but I'm just trying to respond with the different things I've been able to test and trying to get everything out there to make it easier for me to get some help. I do appreciate your opinion a ton.

[liggity]

I will say one more thing....since that wiki article was probably written we have had better amp clamps come out I'm sure. The Flukes are great with that.

[02_Legacy]

I have not ever seen or heard of an injector letting too much flow causing a rich condition, like you said I think in most cases it will cause a lean condition from being restricted. The most likely cause besides fuel pressure is probably going to be a leaking injector, they should be able to test for that while flow testing though.

I still stand by my earlier statement though that you should spend the $15 and buy that vag-com cable, you will be able to see true A/F ratios, long and short fuel trims and more. And I don't know how fast your modis updated, but when I hooked a modis up to my car it updated probably every 2 or 3 seconds, which is incredibly slow. When I am running ECU tracker I get an update of all pids every 20 to 30 ms. I don't know maybe it is just our two different styles but i want to see what the computer is actually seeing, in this case in A/F ratio. The article you posted stated itself that the reference voltages can be different between different sensors. Stoich will never change.

And I'm glad you understand that I am not arguing with you, its happened to me before where the person I am helping ends up mad at me because they think I'm telling them they are doing everything wrong lol. It is hard sometimes when you are just reading what someone says without hearing their tone lol.

[liggity]

Thats what I thought with the injectors. Yeah, my guess is still an injector that isn't atomizing the fuel right in one injector. As we all know, atomized burns quicker than straight liquid.

See, I didn't see any pricing on that link you posted. I'm guessing I overlooked it. The cable is $15 and how much is the program? It works through your phone? iPhone? Cause like I said, I was just comparing Modis to the the tech 2 really. They seem to act the same, but I am extremely interested in the program you were talking about. But next I think it's flow test time.

[02_Legacy]

actually if the fuel is not atomizing correctly I think that would cause the o2 sensor to read lean, if not all the air/fuel mixture is burned it leaves extra oxygen that goes out the exhaust, at least that is how I see it.

The program itself is free, anyone can download it, the only thing is that you will need java to run it, which means it wont work with iphones. Personally I think the best way to use it is to put the program on a laptop, that is what I did.

[liggity]

Each injector can be dumping the same amount of fuel into each cylinder, but if one is not atomizing correctly it can't burn off as well. Misted fuel burns off easier and quicker than liquid fuel. Therefore getting fuel pushed out on the exhaust stroke when all of it isn't used on the compression stroke. That is my guess for whats causing my backfires. I get them quite often actually.

Guess I'll have to download that and give it a shot.

[02_Legacy]

Ya that is a possibility for your backfires but I don't think that will cause the engine to be rich unless the o2 was reading lean and bringing up the fuel trims. My car also backfires slightly under circumstances that you stated, but it has never worried me. I could not even hear them until i installed the thin walled knockoff UEL headers. And even at that, at least in my case, they are so small that I would hardly call them backfires. If you think about it when you roll off the throttle and then back on the engine is going from a deceleration to an acceleration. I think our engines cut all fuel when they are decelling and so as you get back on the throttle they add a little extra to get it going. I can watch my A/F gauge and I will usually get A/F ratios around 13.5:1 on throttle tip-in.

[liggity]

It has to cause the sensors to run rich. The extra fuel is going out on the exhaust stroke causing a rich condition to be picked up by the sensor. Yeah my backfires have startled people before. I get small ones too, but most are pretty loud.

[02_Legacy]

The 02 sensor only reads oxygen not fuel in the exhaust so extra fuel in the exhaust caused by the fuel not being vaporized correctly will not necessarily cause the sensors to read rich. If there is extra fuel in the exhaust because there actually is too much fuel than the 02 sensors will read rich because... well it is rich.
But ya if your backfires are loud enough to scare people there is probably something wrong there lol

[liggity]

Oh yeah I know, fuel going into the exhaust just causes less oxygen to be read pretty much like you said. But yeah, something's wrong and I just wish I had the time right now to find it. I need my last quarter of school to start so I can work on it and get credit for it! Haha

[02_Legacy]

Haha right and having access to a hoist doesn't hurt either lol

[liggity]

Son of a bitch it was a bad coil. The resistance was waaaaay to high on the primary. I heard another misfire like I used to get before I found out the camshaft was off a tooth on the passenger side. Checked the coil and the resistance was way to high, especially 3-4 where I was getting all that funny activity on the scope.

[Garrison]

Awesome! Good on you for being able to track down the solution.

[liggity]

Thanks!
Just don't know why I didn't check that earlier. Had the same problem at work with a single coil in an F150 truck. Then it hit me! Lil embarrassed I didn't figure that out sooner. I now remember one of my instructors at school telling me I should check the coil just to scratch it off the list of things to look at.

[liggity]

UPDATE:Well, I must admit I was wrong about one thing: The air fuel ratio sensor that I have on my outback Is actually different than what I thought it was it reads more like an oxygen sensor. So the low voltage was actually that it was running lean and I found that I had a really high oxygen count When I used the five gas. Found out I had a crack in my headers just before the front oxygen sensor, So that sensor was getting a lean reading And trying to Richen the mixture. Welded that so that's all fixed now but for some reason I'm still running really lean. And it is an intermittent problem because sometimes my car has perfect numbers on the five gas. And then out of nowhere the numbers go all the hell. My NOx jumps up to 400 ppm actually and my HC go up to 200+ ppm.
So now it just screams of an EGR issue, But my engine doesn't have an EGR, it has valve overlap. So now I'm pulling the engine And opening her up. We'll see what happens and what I find....