Hold on. You are going to have to be more specific of exactly what and where you are measuring. Exact wire colors, places, what you find, etc.
An ohm meter has no place in high-current circuits like those for starters and generators, so throw that away. Any readings you get are meaningless and have no value.
To try to explain this better, a standard test for a slow-running starter is the "voltage drop" test. This is the only test that has validity in a high-current circuit, and once you know how to do it in the starter circuit, the same principles apply to charging circuits.
For the first instant when a starter motor is turned on but it is not yet spinning at normal speed, it is typical for it to draw 300 amps. If you use Ohm's Law to calculate resistance, (resistance equals voltage divided by current), R = 12 / 300 = 0.04 ohms. That is way too small to measure with a standard ohm meter. The meter's leads will have a good two to five ohms themselves, so it is easy to see you cannot determine if there is excessive resistance in any connection or wire. Remember, this pertains to very high-current circuits, not all the other low-current lighting and sensor circuits.
If you were to double the resistance in the starter circuit, you would cut current flow in half. 150 amps is not enough to get the starter motor going fast enough. To make matters worse, whenever you run into a slow starter problem, in actual practice there is going to be more resistance than in my sad example, but it is still going to be way too small to measure. We cannot measure the excessive resistance, but we can measure the results of that resistance, and that is the voltage drop test.
Think, for a moment, of standing on a garden hose and partially blocking ninety nine percent of it. With the nozzle turned off and no water is flowing, you would measure the same pressure on both sides of your foot. Your foot is the resistance, but it is not having any effect.
Now open up the nozzle and try to get water to flow. This is when you will still find full pressure on one side of your foot, but almost no pressure on the other side. There is a huge drop in pressure because excessive resistance is blocking current flow. Voltage is electrical pressure, and it is that drop in voltage we can measure when resistance is too low to measure. The first drawing is of a terminal connected to a stud. This could be at the starter, the generator, a battery cable, or the fuse in your circuit. The second drawing includes the meter's probes where you might try to measure resistance. Instead, this is where you need to take a voltage reading, but just like with the garden hose, this voltage drop, (pressure drop), only shows up when current is trying to flow. That means a helper has to try to crank the engine while you take the readings, or, in the case of charging systems, the engine has to be running.
This connection between the terminal and the stud is considered to be a mechanical connection, and as such, the maximum you are allowed is 0.2 volts. The most you are allowed in any one circuit is 0.4 volts. In your charging circuit, you have a mechanical connection between the generator's output stud and that terminal, one bolt for the fuse, the second bolt for that fuse, and the positive cable clamp where it bolts to the battery's post. Every one of those four connections can have no more than 0.2 volts drop, and all four of those readings added together cannot be more than 0.4 volts.
That was all of the generator's positive half of the circuit. For starters, we also have to look at the negative half which is usually just the cable between the battery's negative post, and the engine block. We typically do not worry about that when diagnosing charging systems because a problem in the negative circuit would adversely affect the starter first.
Besides the tedious, time-consuming job of taking multiple voltage drop readings, you can also put one probe on the battery's positive post, (not its cable clamp), and the other one right on the generator's output stud, (not the copper terminal attached to it). If you find less than 0.4 volts with the engine running and the generator trying to supply as much current as possible, there is no defect in that part of the circuit. It is only when you find more than 0.4 volts that you need to test each connection individually to find out which one is responsible.
In this story, at first it sounds like you would find 8.7 volts, the difference between 12.5 volts and 3.8 volts, which is really too high, but I am pretty sure we do not have the right readings at the right time. If the generator is working and there is a high-resistance connection, or blown fuse, going back to the battery, you would find higher voltage at the generator's output stud. The way you listed it at first would equate to finding 12.5 pounds of pressure in the municipal water tower, and 3.8 pounds of pressure leaving the water pump that is filling that tower.
The first thing you should do is measure the voltage on the generator's output stud with the engine off. It must be exactly the same as what you find across the battery posts. Next, measure those voltages again with the engine running. If the fuse is blown or there is excessive resistance in that circuit, you will find the voltage at the output stud is significantly higher than the battery's voltage. Let me know what you find.
Images (Click to enlarge)
Sunday, December 9th, 2018 AT 7:00 PM