2002 Mitsubishi Lancer Battery and Idle Problems

You can start the testing of the charging system yourself. Use an inexpensive digital voltmeter to measure the battery voltage with the engine off and with it running. Engine off, it should be at 12.6 volts. That indicates it's fully-charged. If you find around 12.2 volts, it's good but discharged.

With the engine running, you must find battery voltage between 13.75 and14.75 volts. If it is low, the battery will never become fully-charged.

The second half of the test requires a professional load tester to measure full-load current output and "ripple" voltage. If your generator has one bad diode, you'll only be able to get exactly one third of its rated maximum current, and ripple voltage will be very high. 30 amps from the common 90 amp generator is not enough to run the entire electrical system under all conditions. The battery will have to make up the difference as it slowly runs down over days or weeks.

Ripple voltage can confuse computers, and it can cause voltage spikes and surges that interfere with computer sensor signals. This is where erratic engine idle speed, rough running, and other elusive symptoms can be a secondary symptom that defies diagnosis. Start with the charging system tests first. If necessary, the idle speed can be observed on a scanner. You can see if the computer is requesting a higher idle speed or if it's trying to lower engine speed but without success.

The only way to test diodes with a digital meter is to disassemble the generator, then disconnect them to isolate them from each other, then use the meter's diode test setting. That is not practical unless the unit is already disassembled for some reason. The professional load tester measures ripple voltage which is a result of the generator's three-phase output. Whenever the voltage of one phase is near maximum, one other phase will be headed toward 0 volts and the third phase will be on its way up from 0 volts. At any given time one phase will be developing near maximum voltage. As the maximum voltage begins to go down, the next phase hasn't quite reached maximum yet, so the output voltage will be slightly lower. THAT is what the tester is looking at when it measures ripple voltage. It's the highest and lowest voltages at the output terminal.

If it helps, think of a three-piston water pump. You get three nice pulses of water pressure per each revolution of the pump, but that pressure will be pulsing. Now, if one cylinder goes dead, you'll get a pulse, a pulse, then nothing.

I've never actually looked at ripple voltage on an oscilloscope so I can't say what normal is, but for the sake of this sad story, lets say it's one volt. That one volt of ripple is easily absorbed and smoothed out by the battery, but when one diode is shorted, you lose two thirds of the current output. That's because each diode is shared by every phase, just not at the same time. Just like with the water pump, you'll get a voltage pulse, a voltage pulse, then no pulse. Since one good phase hasn't quite reached 0 volts yet, and the other good phase is on its way up from 0 volts, you might see perhaps 8.0 volts at the output terminal during that missing phase. Ripple voltage now might be 6.0 volts. THAT is what the professional load tester is measuring, and the only way it can be that high is when there's a failed diode, or, ... (PLEASE DO NOT DO THIS!), The battery is disconnected while the engine is running.

In this case you have something else going on. That is shown by the 12.5 volts with the engine running. You must have a minimum of 13.75 volts. Your generator isn't doing anything. The engine is running solely on the battery. This is not a diode issue. The typical causes are worn brushes inside the generator and a totally failed voltage regulator. Your regulator is built into the generator, and without searching for a wiring diagram, it is pretty likely there is no way to bypass it to verify everything else is working. Even if we could do that to prove only the regulator was defective, it is still a big job to replace it. You're much better off with another replacement generator, especially if this one is still under warranty.

There is just one more quick test you should do. Measure the voltage on the output terminal of the generator. With the engine off or running you must see exactly the same voltage as at the battery. If you find 0 volts with the engine off and a high voltage, say 15.0 volts or more with it running, there is a break in that wire between that terminal and the battery's positive terminal. On your car that break will be in the form of a blown very large fuse. Because of the very high current it has to pass, that fuse will almost always be bolted into the under-hood fuse box.

There's two common ways to have no output at all. The less-common is to have a defective voltage regulator. The most common, and especially at the mileage you listed, is worn brushes. Those almost always start out being intermittent. As long as they make contact once in a while, the battery will be recharged periodically. I agree you should get an indication from your warning light, but with internal voltage regulators, they are responsible for turning the light on too. There are conditions where the light won't turn on.

I hope we don't have to get too involved in deeper stuff, but you could also have an intermittent connection in the wire between the dash light and the voltage regulator. At first that is where the regulator gets its "turn-on" signal when you turn on the ignition switch. Once some output is being generated, the regulator puts full voltage back out on that wire. With full voltage from the regulator on one side of the dash bulb, and full voltage on the other side from the battery / ignition switch, the net difference is 0 volts, so the bulb goes off.

Look at the last paragraph on my previous reply. If you find the voltage at the generator's output terminal is higher than battery voltage while the engine is running, look at the fuse for that circuit. There has been a lot of intermittent problems due to the battery's smaller positive wire coming loose where it bolts to the fuse box, so it stands to reason the same thing could happen to the generator fuse. Be sure its bolts are clean and tight.

Your last question about charging voltage needs clarification. Think of a municipal water tower that has so much water it develops 100 pounds of pressure at street level. If you try to fill it more from a pump that only develops 100 pounds of pressure, there's no reason for any water to flow up into the tank. You need to develop more pressure to get the water to go up there. Same is true with electrical pressure, (voltage). A lead-acid battery develops 12.6 volts. To get current to flow through the acid and plates so electrons can get stored in the plates, you need a higher pressure to convince those electrons to flow. As it happens, we need a minimum of 13.75 volts to get an appreciable amount of current to flow to be of any value. If the water pump develops too much pressure, the tank will explode! If the generator develops too much pressure, too much current will flow through the acid and plates and cause them to overheat. That's just like forcing too much current through a wire that's too small and it overheats.

This leads to the next part of my sad story. Suppose you had a really really big water pump and a tiny storage tank. If that pump took just a couple big gulps of water, it might overwhelm that tank long before a pressure relief valve could kick in. Pressure would build up too high and too quickly. Something similar happens in a battery. As all batteries age, the lead flakes off the plates and collects in the bottom of the case. That is similar to having a water tank that's too small. To avoid overheating a battery, it only charges at around 20 to 25 amps, although that may be higher at first, but just for a minute or two. That current is a result of the difference of 12.6 volts stored in the battery's plates and roughly 14.2 volts at the battery's posts. The plates can absorb 20 amps pretty easily without overheating.

By the time a battery gets to be around three or four years old, a real good percentage of the lead has flaked off the plates. In effect, you have a puny little lawn mower battery and not a tough car battery. The capacity of the plates will be greatly reduced but they will still develop 12.6 volts. The voltage regulator still keeps system voltage at 14.2, so you'll still have roughly 20 amps going through the plates. There's our tiny water tank with a strong water pump. You're forcing full charging current through half a battery. It is less able to absorb the electrons, but instead of exploding like the water tower would, the plates get hot. That gets the acid hot and makes it give off more hydrogen gas than normal. That excessive bubbling splatters on the underside of the top cover of the battery's case. There the acid condenses and some of it seeps out between the case and the two posts. That acid causes the white or green corrosion around the posts, and any acid that spews from the vents will condense on top of the case where if enough builds up, will cause self-discharge of the battery. It has been my experience that calling that corrosion to a customer's attention, and selling them a post cleaning service or those juicy rings is of no value. That battery is going to fail within six months. It's better to explain that to the owner so they can prepare for the inevitable. Cleaning that corrosion will not save a battery or extend its life.

To finish my thought related to your question, where my water tank would explode from being too small for the size of the pump, it would explode because the pressure went too high. In the battery, the same thing happens. The normal amount of electrons will flow through what's left of the plates, and the pressure also will go too high. In this case the voltage regulator sees that and is still in control. It prevents system voltage from going too high, BUT, the battery is the key component in helping it do that. If you have a really big water pump feeding a really small tank, the pressure will go up quickly. If you have a properly-working generator feeding a battery with half the lead flaked off the plates, the voltage will tend to rise. THAT is exactly what happens as a good battery charges too, but not as dramatically. A totally dead battery will measure very close to 12.2 volts. A fully-charged battery will measure 12.6 volts. That same fully-charged battery won't accept much charging current from the generator until the voltage reaches around 14 volts or more.

When you measure 12.7 or 12.8 volts at the battery, you're measuring a "surface charge" right after the battery was being charged. The extra couple of tenths of a volt are due to free electrons in the acid that didn't make it into the plates to be stored. There's not enough of them to do any useful work. The best comparison I can think of would be the humidity in the air at the top of the water tank. Sure, it's water vapor, but it's hardly enough to do any good. To avoid confusing battery load testers, it is customary to eliminate that surface charge so it doesn't make a weak battery look good. We do that by simply turning on the head lights or some other load for a few seconds. Doing that will use up those free electrons in the acid, but it will have almost no affect on the electrons that are stored in the plates, and those are the ones we want to measure.

One more comment about the warning light. If there's a bad connection on the fuse for the generator, (typically a 100 or 120 amp fuse), you have to look on a wiring diagram to see where in the circuit the voltage regulator is sensing system voltage. On all domestic Chrysler products the voltage regulator lives inside the Engine Computer, and that's where system voltage is detected and measured. On almost all cars with the regulator built into the generator, system voltage is detected right on the output terminal or another wire that is tied directly to battery voltage all the time. Fords use a second small wire. GMs look at the output terminal. A very common problem is when people replace a generator that truly was defective, but they overlook the blown fuse it caused. The generator will still develop a voltage between our desired 13.75 and 14.75 volts, and the internal regulator will detect and measure that right there at the output terminal, but that voltage and current won't make it back to the battery because of that blown fuse. This is where you can find, lets say, 14.5 volts at the generator's output terminal but only 12.6 volts at the battery.

On Chrysler products the system voltage sensing wire is at the Engine Computer and comes from the battery. The regulator will see 12.6 volts which is too low. It will continually bump up the little field current to make a stronger and stronger electromagnet in the alternator in an attempt to get charging voltage back up to where it should be. You could find over 18 volts at the alternator's output terminal but that high voltage would never get to any of the other circuitry. No matter how hard the regulator runs the field winding, it would never see anything higher than battery voltage of 12.6 volts. In this case the voltage regulator, being inside the Engine Computer, knows the engine is running, it knows it is supplying field current, and it knows it can expect system voltage to be between 13.75 and 14.75 volts. Given that set of conditions, when it actually sees only 12.6 volts, it says, "darn the bad luck, and commands the "Battery" light to turn on.

On GM systems, output voltage is sensed right in the generator at the output terminal. Think of this as a free-standing device. The regulator does its thing without knowing when there's a break in the output wire or a blown fuse. You'll continue to see proper voltage at the output terminal, therefore, the regulator assumes all is well and it keeps the "Battery" light turned off. It has no idea there's no current going back to the battery. It just assumes the battery is fully-charged. This is the type of system found on the majority of import cars. This is why you can have no Battery light with low voltage at the battery, and it's why it's important to measure at both places while the engine is running.

Ford systems are a little different. They use a second 12 volt wire to sense system voltage, not the output terminal, and that wire has full battery voltage on it all the time. Both go back to the battery, but with a break or blown fuse in the output wire, the sensing wire will see the battery's 12.6 volts and not the 14.5 volts at the output terminal. Here the regulator will keep bumping up field current in a failed attempt to get battery voltage up to where it should be. Just as with the Chrysler system, you'd see a voltage at the output terminal that's much too high, but it wouldn't make it to the car's electrical system where it could do damage. The difference here is in how Ford determines when to turn on the "Battery" light. They look at a couple of different things. They tap off the center of the stationary "stator" winding which is where the output current is developed. Being in the middle of it, you will find exactly half the voltage that you find at the output terminal. Unlike older GM and Ford systems that PUT voltage on the dash light to turn it off, the newer systems from the late '80s on have the regulator look at that stator, ("S" terminal) voltage. 7.0 volts there means the output voltage is 14.0 volts, exactly double. The voltage regulator looks at that stator voltage, then if it's happy, it sends voltage to the dash light to turn it off. The difference with this system is it will turn the warning light on for no charging voltage, low charging voltage and HIGH charging voltage. I'm not aware of any other systems that respond to high charging voltage. That's a nice feature, but the engineers managed to put in plenty of other things to go wrong.

The point of all this is there are a variety of places in the circuit where a defect can occur, different manufacturers sense system voltage at different places, and they use different methods of causing the "Battery" light to turn on. Each system has its advantages and disadvantages. The bottom line is you can't assume the charging system is working properly just because the warning light is off. You can only know for sure something is wrong when the light is on.

Therefore and hence with, I'm not happy with 12.5 volts at the battery with the engine running.