This is the confusion I referred to but didn't explain. To be fair, to explain this in the classroom takes a good hour, but only after we learn about generators. This is what also makes diagnosis difficult for experienced electrical experts.
Let me start by boiling down generators to the single most important concept. To generate voltage mechanically, you need three things. You need a piece of wire, a magnetic field, and most importantly, movement between the two. In generators, we use a coil of wire, an electromagnet, and we spin that electromagnet with the belt and pulley. The spinning electromagnet is called the "rotor", or the "field winding". We use that instead of a permanent magnet because we can adjust the strength of the magnetic field by how much current is sent through it. That's the job of the voltage regulator. That current is relatively small and is easy to control, as in 0 to 3 amps. As it spins, voltage, (electrical pressure), is "induced" into the stationary, or "stator" windings. To increase the maximum possible output current of a generator, all that is needed is to add one or two more loops of wire to each stator coil. The voltages in all the loops adds up, and just like in a water pipe, the more pressure you have, the more current it will cause to flow.
At first, a starter is the exact opposite of a generator. We have stationary coils, we push current through them to develop magnetic fields, then we get movement out of them. Here's the first half of the confusion. The common starter for an older GM or Ford V-8 engine as late as the 1990s would draw close to 300 amps when it first started cranking the engine, but you'd never see that on a professional load tester unless that starter was seized or something was causing it to be locked up, meaning no movement.
Once the starter motor gets up to speed, you have a coil of wire mounted to the housing, an electromagnet developed by the armature, and it's spinning. Those are the same three things needed in a generator. In other words, a spinning motor is a generator. What it generates is called "back electromotive force", or "back EMF". There is no way to measure that back EMF, but we can measure the effects it has. For my example, we'll say it develops four volts of back EMF. That four volts opposes the 12 volts supplied by the battery. In effect, only eight volts is running the motor, and current flowing out of the battery is only 200 amps. This is why if an engine is seized and you hold the ignition switch in the "crank" position too long, the battery cables will get hot very quickly. If the starter cranks the engine just fine, but you do that for a long time, the cables do not get very hot. During normal cranking, current flow is much lower once the motor gets up to speed.
Related to this, as the battery becomes discharged after prolonged cranking, its voltage goes down a few tenths of a volt. Lower voltage means reduced current to the starter. That means a slower cranking speed. With less movement, less back EMF is developed. With less back EMF to oppose battery voltage, current flow goes up. So now starter current goes up even though battery voltage is going down. At first that seems to disagree with "Ohm's Law". Those are a set of 12 formulas that relate volts, amps, resistance, and power to each other, and that is the first part of the confusion.
The second part of the confusion stems from there are actually two inter-related, but separate starter motors built into every one. Two brushes are needed to pass the current into and out of the moving armature, but every starter motor has four brushes. Each pair is passing current through part of the armature, but at different times. This results in twice the power without doubling the size or weight of the motor. For this story, each pair of brushes are passing 100 amps, making 200 amps total leaving the battery. That is the current we're measuring with the load tester. If the battery has been found to be good, and capable of supplying sufficient current, and that starter current is measured to be within the acceptable range, we'd conclude the starter is okay, at least in that respect. With these old, high-torque starters, a lot of force was placed against the bushings that armature rides on, often resulting in those bushings wearing away on one side. That allowed the armature to drag and slow down. Here's the neat thing about starter motors, which are "series-wound". That means whatever current flows through one coil has to flow through the other coil next. Think of a river that flows through one town, then through the next town. The same water flows through both places. "Parallel-wound" motors are typical of heater fans and radiator fans where they're designed to develop constant power, but they will change speed as voltage changes. This would be like the water in a river going around an island. A drop of water flows through one path or the other, never both.
In the series-wound motor that's up to speed, 200 amps determines the strength of the electromagnetic fields that are causing the rotational movement. As that motor is loaded down, movement becomes less, so back EMF becomes less. With less back EMF opposing battery current, that current goes up; we'll say to 250 amps. 250 amps makes for stronger electromagnetic fields, and therefore a stronger motor. The important point is, ... As you load down a series-wound motor, it gets stronger. A starter might draw 200 amps on a 350 c.I. Engine, but use it on a 454 c.I. Or a high-compression engine, and it might draw 280 amps while still cranking at the normal speed.
Now for the rest of this second confusing point. Those four brushes are going to wear away from use. At some point the first one will fail to make contact with the commutator bars on the armature. No current can flow through that half of the coils, so you're left with only the other part that's still working. Remember it was drawing 100 amps earlier, but now the overall strength of the motor is half of normal, rotational speed is cut way down, back EMF is much lower, so the battery current we can measure goes way up, often as high as 200 amps. 200 amps puts a normal strain on the battery which draws its voltage down during cranking. Design specifications always say that must never drop below 9.6 volts. In fact, that's part of what we look for during a battery test. We'll say in this story it's a really good battery and cranking voltage is around 10.5 volts, which is fine. You have acceptable, (normal), voltage, and normal current of 200 amps, but the starter is cranking very slowly. This is what makes this problem so hard to identify. Voltage and current are normal, but the customer is complaining of slow cranking speed. There's no way to test for that worn brush. We can only see the results of it, but then we have to be smart enough to figure out what's happening. Because this is so uncommon, it's why even experienced mechanics become confused.
I purposely left out Chrysler starters in this story. Starting in 1960, they used a gear-reduction assembly. That caused the armature to spin about four times faster than everyone else's direct-drive starters. As such, they developed a lot more back EMF. Overall torque was higher while current flow was lower. 150 amps was typical on a big 440 c.I. Engine. They even modified them for use on high-compression GM and Ford race engines until they developed their own high-torque starters.
Beginning in 1987, GM also went to a little gear-reduction starter design. As with your Nippendenso starter, you get a lot of power out of a little package.
Now that I shared all that great and wondrous wisdom, let me go back to those solenoid contacts I showed you in my last reply. There's two coils of wire inside the starter solenoid. Solenoids are bolted onto the starter motor or in yours and many other designs, it's built in. A typical 15 to 20 amps flows through those coils, then the electromagnetic fields draw in a metal plunger. A shaft I mentioned previously on the end of that plunger pushes the starter's drive gear into mesh with a ring gear on the flywheel, flex plate, or torque converter, (something that's attached to the crankshaft). Once that drive gear is fully-engaged, the plunger continues to the end of its travel, at which time a copper disc on it makes the connection to the two contacts. That switches on the 200 or more amps that flows through the starter motor. That's more current than what is used very often with a stick-type welder, and you know how much those arc. That same arcing occurs across the contacts every time the solenoid turns off. All contacts are going to burn away over time, but that happens a real lot faster with these starters. It's so common; that's why companies have produced repair kits for them.
Every time a starter solenoid is released at the end of cranking, that copper disc is designed to rotate on the plunger a little. That brings a fresh spot around for the next time to keep wear even all the way around. The problem is though, experience has shown those discs don't wear very much while the contacts wear real fast. Because those discs do rotate is why this always starts out as an intermittent problem. You turn the ignition switch to "crank", and you hear that nice solid clunk as the plunger engages the drive gear, but the current to the motor doesn't get switched on. Release the ignition switch, the plunger retracts, the disc rotates, then it does make contact the next time, and the engine fires up.
As the contacts develop deeper and deeper worn spots, the no-crank problem acts up more often. In the case of my mother's '95 Grand Caravan, she lost count after 700 tries and a blister on her thumb, but it did eventually crank. You can be sure I heard about it that night. I had ignored the problem for well over six months.
For my final comment of value, a lot of people remove starters to have them tested at an auto parts store, but that only shows whether they are capable of operating. Bench-testing is not accurate. With no load, dragging from worn bushings won't show up. With one worn brush, the motor will still operate like normal. The bigger problem is with no load on it, an older direct-drive starter is likely to draw less than 50 amps because it will be spinning so fast. They need to be tested under real-world conditions, meaning on the engine. That's where the problem is, so it doesn't make sense to take a part away from the problem to test it.
Some car brands and models are well-known to develop corroded cables under the insulation where you can't easily see it. We find that with voltage measurements while a helper is cranking the engine. If, lets say, 8- percent of the strands of wire are corroded apart, that's like using a cable that's only a small fraction of the diameter needed. It may still be able to pass 100 amps, but if the starter needs 300 to get it up to speed, that will never happen, even though that starter tested fine on the test-bench where it only needed 50 amps. By testing on the engine, the entire system gets tested, not just one part.
SPONSORED LINKS
Thursday, March 25th, 2021 AT 7:31 PM
