Are you saying the charging system is not working? Once the engine is running, battery voltage must be between 13.75 and 14.75 volts. If it stays at 12.6 or less, we need to diagnose the charging system. Computers are very intolerant of low system voltage and will set random fault codes. Also, since low system voltage adversely affects the performance of injectors, which in turn leads to increased emissions, that will turn on the Check Engine light.
The most common cause of code 203 is a spread terminal in the connector for that injector or in the connector that sits between the engine wiring harness and the one on the body. Look for corrosion between those two mating terminals too. If you think the injector itself is at fault, switch it with that from a different cylinder. Erase the fault code, then see if the code sets again for the same cylinder or the one you moved the suspect injector to.
You should also get a code 303 for cylinder three misfire if the injector stops firing, although that might not set right away. Code 203 doesn't specify a break in the electrical circuit like code 268 does. Part of the strategy is the computer watches the current flow ramp up over time. When it reaches a strong enough magnetic field to pull the valve open, there is a noticeable drop in the rate current increase. The computer can tell by that sudden change exactly when the valve opens, then it can calculate the number of milliseconds left before it turns that current off. If it doesn't see that valve opening, it may set code 203, or any of those for the other cylinders, and that weak opening can be due to low system voltage. The injector may not open at all, but more likely it will open late, resulting in too little fuel to that one cylinder. That might result in a misfire and a code 303.
I drew you a sad graph below to show the voltage and current relationship for an injector. As of today, the computers can only modify how long the injectors are held open based on the front oxygen sensor readings for the two sides of the engine, not for an individual cylinder. If cylinder three is running lean, the unburned oxygen will be detected in the left bank, and the computer will request more fuel from all four injectors on that side. It doesn't know which cylinder is responsible, just that something needs to be tweaked on that side.
Be aware too that for any fault code to set, there is always a very long list of conditions that must be met, and one of those conditions is that certain other codes can't already be set. There's two fairly common examples of this. The first is when a defect is detected for something used for reference, anything compared to that reference won't be tested, so no fault code will be set. Consider the intake air temperature sensor and the coolant temperature sensor. The Engine Computer knows that after sitting for more than six hours, those two sensors had better be reading the same temperature. If you unplug the intake air temperature sensor, the computer will detect the open circuit and set the code for "intake air temperature sensor voltage high", meaning it's stuck at the full 5.0 volts. It knows it can't rely on that obviously-incorrect reading to compare to the reading for the coolant temperature sensor, so if that one just goes out-of-range, that won't be tested or detected. You may not get a fault code for the coolant temperature sensor until the intake air temperature sensor circuit is repaired; in this case, plugged back in. Then, the continuous self-tests will resume, and if there really is some problem with the coolant temperature sensor, that is when it will finally be detected, and a new fault code will set. This happens real often on GM anti-lock brake systems when the customer waits too long to have it repaired. That gives the vehicle time to develop multiple problems, but after the first one sets a fault code, all other related tests are halted. Finally the customer has the first one repaired, then the self-tests resume, and the next problem shows up right away. That leads to lots of frustration on the part of the mechanic, and customers incorrectly assume the problem wasn't diagnosed properly.
The second time fault codes may not set is when multiple problems are causing the same symptom. This happens most commonly with single-cylinder misfires. If the screen in an injector becomes plugged, the blocked fuel flow is not detected directly, but the computer sees the resulting slow-down of the crankshaft's rotational speed for that instant. By knowing the position of the crankshaft, and when that slow-down occurs, it can tell which cylinder is responsible. It is that same slow-down that we feel when we feel a misfire, but the computer can detect much more subtle changes than we can feel. Now if you come along and short out that spark plug wire, or the spark plug gap becomes shorted with carbon or oil, the cylinder is already misfiring due to the injector, so no additional fault code will set related to the lack of spark. You might figure out from looking at the oxygen sensor voltages that the injector is causing a problem, you replace it, and still have a misfire that requires more diagnosis. This is real common on higher-mileage GM cars. I mentioned that Chrysler buys their injectors from Bosch in flow-matched sets, and they have almost no problems. In fact, to my knowledge, no other manufacturer has common, recurring injector problems, except for GM. They just grab a handful out of a big bin and toss them into an engine as it comes down the assembly line, with no regard for matching the flow rates. That rarely causes a problem at first when the vehicle is new, but by the time it hits around 100,000 miles and it's out-of-warranty, it is real common for one or two injectors to start flowing a little less fuel than the rest. This can result in setting a "lean exhaust" fault code that is almost impossible to diagnose. As it becomes worse, you can have a single-cylinder misfire fault code, and you might not feel that misfire.
The solution to that problem is to install a set of rebuilt injectors that come in flow-matched sets. Even if you've figured out the injector is the cause of the problem, if you install one new one or one used one, you really run the risk of it not flowing the same amount of fuel as the others, and it might not solve the misfire. Jim Linder used to have an injector rebuilding shop next to the Indianapolis Speedway, and he put on real high-level training classes all over the country. On his web site they described the ten most common injector-related problems, and eight of them referred to problems on GM vehicles. My reason for describing this is if you plan on replacing just one injector, you have a very good chance of introducing this new variable and a new problem that defies diagnosis.
Here's my nifty drawing showing injector current. On almost all vehicles they are supplied with 12 volts on one terminal, and it shows up on the other "control" terminal by reading through the injector. On all Chrysler products, that 12 volts comes from the ASD relay. The red graph shows that 12 volts, then, when the computer turns that driver circuit on, which is a transistorized switch, it draws that second terminal down to close to 0 volts. That puts 12 volts across the injector's electromagnetic coil.
The resulting current flow in the blue graph causes an electromagnetic field to build up inside the injector, but since it takes energy to do that, the buildup fights, or opposes that current flow. As a result, while voltage switched instantly, current builds up relatively slowly. Some of that energy goes into making the valve move. Once it reaches the full-open position, it can't move any further, it can't use up any more energy, so the current has to work harder to increase. That transition point is shown where the valve opens, and the computer sees that change. Failure to see that transition is what sets code 203, which is different from an electrical-related code.
Image (Click to make bigger)
Wednesday, February 20th, 2019 AT 5:56 PM