Interesting interpretation of how the system works but -30 is not normal under any circumstance. What you need to determine is whether way too much fuel is entering the engine due to a defect, and the computer is seeing that and trying to correct it, or whether the right amount of fuel is going in but the computer is getting incorrect information and mistakenly subtracting fuel that it shouldn't be. One clue is to compare the two banks to see if both oxygen sensors are responding the same. If they are, look for something both sides of the engine have in common such as fuel pressure or a vacuum leak. If only one bank is acting up, look for something that wouldn't affect the other side of the engine. That would include an injector, spark plug, individual ignition coil, etc. You can swap two ignition coils side-to-side to see if the problem goes to the other side. Injectors take a lot of time to switch but Chrysler has extremely little trouble with theirs so that should be left as a last resort.
The throttle-by-wire system is totally independent from the emissions system. It only affects engine speed, not other performance problems. What they DO have in common is injector pulse width. The length of the "on-time" for the injectors is varied to control fuel / air mixture and for engine speed. The pulse width is varied for engine speed based on the reading from the throttle position sensor on the throttle body, and the pulse width is further modified for emissions by the readings from the upstream oxygen sensors. The oxygen sensors' readings are inaccurate when the sensors are cold as you mentioned, but the computer ignores them in open loop. O2 sensor problems should not affect engine performance when it's still in open loop. O2 sensors also have no affect on idle speed. If engine speed doesn't pick up when the accelerator pedal is pressed, that is a problem with the throttle-by-wire system, not the emissions system. This insane throttle system is the same as the one that put Toyota in the news last year. Any service work on that system should be left to the dealer. Working on it yourself, and especially modifying it in any way will make you a party to any future lawsuit.
Being in tv / vcr repair for over 30 years, I know all about using power supplies to inject voltages but we did that to test and troubleshoot circuits, never to replace what the engineers designed in. To modify anything in any way left us open to a lawsuit if that product started a house fire. It's these things mechanics and other service people have to keep in mind when working on the products they service, and it's what most do-it-yourselfers aren't aware of. I was impressed with your grasp of system operation until you mentioned the Check Engine light staying on due to the modification of the oxygen sensor heater circuit. How will you ever know if a different, potentially serious problem occurs? You could have a minor problem such as a leak in the fuel vapor recovery system which wastes fuel through evaporation, and is fairly common, or you could have a leaking purge valve which will allow way too much fuel vapor to be drawn into the engine without the computer knowing about it. That will cause running problems and very poor fuel mileage. Remember, oxygen sensors don't measure or detect unburned fuel. They only detect unburned oxygen. As long as it detects the pulses of planned, unburned oxygen, the computer will be happy and will not know about that excessive fuel.
"I wanted to make sure the PCM was causing the problem before I spent a lot of money on a new one."
You don't need a fancy power supply to run the heaters. A simple resistor in series with the switched battery voltage would do the same thing. I've never checked the circuit with an ohm meter so using your 5 ohm observation, if you're starting with just under 15 volts system voltage and you want to drop two thirds of that voltage across the resistor leaving one third for the heaters, you'd need a 10 ohm resistor. Doing the math with Ohm's Law, it would need to be at least a 10 watt resistor.
" I’ve thought about adding a resister to the dangling heater wire under the car but my wife has become so use to warning light on the dash..."
If you can make the Check Engine light turn off by inserting a resistor to mimic the heater circuit, that proves the computer is supplying the 5 volts and is monitoring current flow in that circuit. Why use the computer to run an external resistor? Use it to run the O2 sensor heaters as designed. If the computer is not supplying 5 volts, what else isn't it supplying? Or is something grounded in that circuit and the computer is shutting the supply down to protect it?
The second problem is the computer is constantly performing many self-tests during engine operation. As an example, it knows that after the engine has been off for six hours, the intake air temperature sensor and the coolant temperature sensor had better be reading the same temperature. It also knows the coolant temperature sensor's readings should increase at a given rate based on engine speed, load, and time. Once it reaches around 180 degrees, it knows it can switch to "closed loop". All that means is it is still making fuel metering decisions based on many sensor readings and other criteria, AND it includes the oxygen sensors' readings to make minute fine tuning adjustments for lowest emissions. During prolonged idle periods, the exhaust gas temperature can drop below 600 degrees at which point the O2 sensors will stop responding. That became a problem many years ago with police cars that would sit idling for a long time at crash scenes. The heaters built into the sensors solved that problem. That hardly compares to a passenger car sitting at a red light. Even popping into a grocery store for 20 minutes would not affect engine performance related to the O2 sensors. The driver should never notice any change in engine operation whether it be from a cold start-up, warm restart, or prolonged idle. An engine speed problem for five seconds has nothing to do with the emissions system and you should never HAVE to hold the gas pedal down to warm the exhaust system to make the engine run right. No one would buy a car that had that requirement. If doing that works, that may be a valuable clue to the cause of the problem but it's not a cure.
Getting back to the computer comparing sensor readings and performing self-tests, there are some tests that will not run when there are related diagnostic fault codes stored in the computer. When the computer sees road speed, for example, and a closed throttle, it knows you're coasting and manifold vacuum should be considerably higher than when the engine is idling. It measures that manifold vacuum with the MAP sensor. If there is a code set related to the throttle position sensor or vehicle speed sensor, it will never perform some tests on the MAP sensor since it doesn't know what to expect. I'm generalizing here by talking about older cars without the throttle-by-wire system, but the concept is the same. When your Check Engine light is on many self-tests related to the fault code will not be run. You could have a sensor problem or a performance problem, along with the associated running problem, and no additional fault code will be set. The only way to let the computer tell you what's wrong is to fix the first problem, in your case, the O2 heater circuit, then drive the car until the next code sets.
Some people ignore the Check Engine light for so long until a different running problem occurs, then they take the car in for service. All the mechanic knows at first is there's a fault code for one problem. Some owners are so inept at describing other symptoms or the car's history that the mechanic has to discover all the problems on his own. He may provide an estimate to remedy the fault code currently being displayed, expecting to find a defective sensor or corroded wire to that sensor. After the repairs are completed, he might be surprised to find the Check Engine light is still on and a new fault code is set seemingly unrelated to the one he just repaired. Had he known the car's history, he could have planned for finding additional problems, or at least expected there was more wrong with the car than just the one fault code, and he could have explained that to the owner and prepared them for that possibility. Instead, the second problem might finally show up, either on his test drive or the day after the customer gets the car back. Of course the uninformed customer blames the mechanic for doing something to their car or for not fixing it right the first time. They don't understand the current problem existed many months ago when they ignored the Check Engine light.
Also, the fault code the mechanic is basing his diagnosis and estimate on would likely be the one that caused the Check Engine light to turn on many months ago. Imagine his dismay when he has a fault code to tell him which circuit to check, he does that, finds the cause of the problem, corrects it, erases the code, goes on a test drive, and the light turns on again with a new code requiring a totally different diagnosis that he didn't include in his estimate. This happens all the time and is just one of the many reasons mechanics have an unfairly deserved bad reputation. Without knowing the history, (some owners intentionally keep it secret, either due to shame at ignoring the light, ignorance, or in hopes of getting two repairs for the price of one), the mechanic will typically think he did something wrong and will set out double-checking his work and possibly replacing the same parts a second time. That results in a lot of wasted time and money. More experienced mechanics trust their work so they move on to the new problem sooner but it still leads to lots of arguments with customers. That can be avoided by having known problems repaired as soon as possible before a second, different problem has time to occur.
As a side note, those unrecorded codes are a big headache with the anti-lock brake systems on GM cars and possibly other brands. GM has a real big problem with front wheel bearings developing a little play, which is normal and acceptable, but with their flimsy built-in wheel speed sensors, it's enough to cause signal dropouts, a related diagnostic fault code to set in the ABS Computer, and the system to turn off. In typical GM fashion, they like owners to have to buy big assemblies to fix little problems. The only way to get a new wheel speed sensor is to buy the expensive wheel bearing. Doesn't matter though; in this case the bearing is cause of the problem. Since this problem can occur in as little as 15,000 miles, it is very possible for the second bearing to develop the same problem before the car is finally brought in for service. Since the computer has to compare the signals from one wheel speed sensor with the other ones, when one fault code is set, it stops performing self tests. The result is you will finally take the car in for service, have a fault code for only one wheel speed sensor, have the new bearing installed, then the fault code for the second wheel speed sensor shows up for the first time. Now the mechanic has to tell you a second bearing is needed to turn off the warning light. Nothing was done improperly during the repair but wouldn't it be better to know both bearings were needed right away? Mechanics hate having to keep telling customers that they found more and more things that are needed. They would rather give you one list right up front.
I have no problem with you using a power supply to diagnose the O2 sensor heater system. The reason for my story is I don't condone leaving it in place permanently. I suspect you understand that other problems will pop up on occasion, but it's the other people reading this thread who might think that is a legitimate fix. Those are the people who don't understand the ramifications of modifications and how they affect other systems on their cars.
The biggest confusion is in your last paragraph. The injector on-time has always been controlled by the computer. It remains unusually constant. As engine speed increases, the number of pulses per second increases accordingly, one pulse per intake valve opening. The amount of air was controlled, by of all things, an unbelievably simple and reliable throttle cable physically connecting the gas pedal to the throttle body. Idle speed was controlled by an air bleed passage around the throttle blade and a valve that was opened or closed by the computer. When a higher idle speed was needed, the computer opened the air valve, (automatic idle speed motor), and increased the number of milliseconds the injectors remained open during each pulse. The oxygen sensors have absolutely nothing to do with idle speed. Where it gets confusing is there can be a problem that affects idle speed AND the O2 sensor readings but other than that common cause, the two things don't affect each other. The most common example is a vacuum leak. The extra air causes idle speed to increase, (without a corresponding increase in power), and the extra oxygen will be detected by the O2 sensors. You can't say the oxygen sensor caused the higher idle speed, only that the two clues are related.
As for open loop vs. Closed loop, the computer uses a long list of variables to calculate how much fuel should enter the engine. Once the coolant temperature is high enough, the computer adds the oxygen sensor readings to that list. You should never know when that switch takes place while driving, and there isn't really anything you can do to control it.
To add one more dimension to the story that I am less familiar with, most Engine Computers today perform some tests when a very specific set of driving conditions is met and some are done after the engine has been off for a certain period of time. Modifying the O2 sensor heater circuit will cause some of those tests to fail. I know of two tests, both on GM products because they cause so many complaints. One test causes a severe stumble at highway speed, but it occurs very infrequently. Simply being aware it's normal satisfies most owners. I can't remember which system is being tested but it was mentioned in a few high-level classes I attended. The other problem occurred on older Corvettes. Twenty minutes after stopping the engine, the computer runs a real high current through the mass air flow sensor's sensing element to burn off any contaminants. Twenty minutes is about how long it takes to change the oil and filter. Slamming the hood closed at the end of the service produced enough shock to shatter the white-hot element. The defective sensor caused a no-start condition. A new sensor was needed to solve the problem, and guess who got blamed for causing the damage?
I can't say for sure if your computer runs a similar self-cleaning test on the O2 sensor, but at least be aware those are the types of things that are going on that we don't know about. That would not take place if you leave the heater circuit disconnected from the computer. Besides the liability issues, it's why modifying anything on newer cars is not recommended. The days of common sense in car design are long gone. Today we have computers doing all of the things computers weren't needed for before. Other than the emissions system, very little of this new technology actually benefits the owner.
Things get even worse in areas with emissions testing. No car will pass with the Check Engine light on. Instead of stuffing a probe up the tail pipe, newer cars pass the test if all of the "monitors" have run in the recent past. Those are the self-tests built into the car's computer, and they only run when those very specific sets of conditions are met. A Check Engine light, a failed sensor, or any number of other conditions that don't even turn the light on will prevent those monitors from running. Many people, including mechanics, complain that no matter what they do they can't get one or more monitors to run for the car to pass emissions testing. As frustrating as that can be, imagine how much worse it will be when something is modified.
SPONSORED LINKS
Friday, November 25th, 2011 AT 1:29 AM
