When you go to start your car one morning and all you get is a click instead of a crank, what's wrong? Is the battery not holding a charge? Is the starter bad? Is there something wrong with the voltage regulator? All you need to troubleshoot this with a high degree of confidence is a multimeter and a little knowledge.

The Waterworks Model
Here's an intuitive way to picture how electricity works: let's pretend that instead of electrons propogating along conductors, it's just water flowing through pipes. This is not how electricity really works -- don't try to take the analogy very far -- but it can be helpful to think about it this way.

Picture a water tower with its tank raised above the ground. There's a pipe running from the tank down to the ground, where it bends and runs horizontally. There's a valve near the end farthest from the tower, just before the pipe empties into a well. OK so far?

With the tank full of water and the valve closed, there's no water flowing through the pipe, but there is a uniform amount of pressure throughout the horizontal part of the pipe. How much pressure is determined entirely by the height of the tower, not by the volume of water in the tank or the diameter of the pipe. That pressure represents potential -- we could use that pressure to do some work if we opened the valve.

When we open the valve, water will run through our pipe into the well. The wider we open the valve, the greater the rate of flow.

All the water in the tank won't just instantly fall into the well, even if we open the valve wide, because there's a certain amount of friction in the pipe. If our pipe is narrow the current will be slow; if it's wide, there will be less opposition to the current, but there will still be some.

We now have three factors that determine the performance of our waterworks model. They are pressure (potential), rate of flow (current) and friction (resistance).

Now let's gum up the works a bit and put some sort of obstruction in the pipe; not enough to block it completely, but enough to get in the way of the flowing water. If the valve is closed and there's no current at all, there will be equal pressure on both sides of the obstruction. But when we open the valve and current flows, there will be more pressure on the tower side of the obstruction than on the valve side, yes?

The more restrictive the obstruction, the greater the pressure difference will be. Also, the greater the current, the greater the pressure difference (play you're opening the valve varying amounts, if that helps you imagine this).

Notice that the normal friction in the pipe is an obstruction in itself -- even if we do nothing to block up the pipe, there will still be some pressure difference from one end to the other whenever current flows.

Translating to electrical terms
Electrical potential, measured in volts, is equivalent to pressure in the waterworks model. The rate of flow, or current, is expressed as amperes ("amps" for short), and the measure of friction or obstruction to that current -- resistance -- is the ohm.

Whether your multimeter has a digital readout or an analog meter movement, it will have two test leads and some sort of dial that switches between measuring volts, amps or ohms. Ohms cannot be measured in an energized circuit, and automotive electrical systems operate on far too many amps for multimeters to handle, so we're going to troubleshoot everything by measuring the relative voltage at various points.

Whenever current flows, we can measure a voltage drop from one point in the system and another. A predictable drop just tells us that current is flowing through the resistance inherent in the system. If we see an unexpected drop between two points in the system that are right close together, we've found unwanted resistance -- an obstruction in the "pipe" -- usually due to dirt or corrosion. If we find less drop than predicted, not enough current is flowing.

Read the instructions that came with your meter and make sure you know where to plug the test leads in and what dial setting to use to measure 12 volts DC.

And now, back to your car
Think of the battery in your car as being equivalent to the water tank and the vertical part of the pipe in our model -- picture the positive battery terminal as being located at the bend in the pipe.

Imagine your car's starter is the machine that's going to turn potential into work. In the waterworks model, it would be situated between the valve and the well.

The well is electrical "ground." Here the water model breaks down a bit, so please take this on faith: current will not flow unless the well is there. Take away the well, and it's just like capping off the end of the pipe, OK? That's because electricity has to flow in a circuit -- there has to be a way for it to return to its source. In your car, the starter circuit is completed through the bellhousing, the engine grounding strap, the car's body and, eventually, through the negative battery terminal.

What's a solenoid, anyway?
A solenoid is a sort of electrically operated piston. When current flows through it, the piston is extended; when current is cut off, the piston retracts.

The solenoid on your starter has two purposes: 1) it pushes the starter gear out so it will mesh with the teeth on the motor's flywheel, and 2) it actuates a switch that connects battery voltage to the actual starter motor.

The electrical term load means "a demand for current." The starter motor is a large load, so a large cable is required to feed it with all the current it wants. The solenoid, on the other hand, is a small load, so regular wiring is sufficient to supply it with current.

The solenoid is controlled by the ignition switch. By using the solenoid to switch the large motor load in and out, the ignition switch can be kept to a reasonable size -- you wouldn't really want full-size battery cables hooked up to a switch the size of a soda can in your dash, would you?

Normal operation
All the following applies to 12 volt systems -- those of you with 6 volt systems, please interpolate. Values given here will normally vary a bit from car to car.

Measure the voltage between the positive battery terminal and the body of the car (ground) -- this is just like checking the pressure in our waterworks model at the pipe bend relative to the well. With the ignition off, you should read 12 volts DC. Now measure from the other end of the positive battery cable, at the big bolt on the starter, to ground. This will read exactly the same because there's no current -- just as the full pressure is felt everywhere in the pipe with the valve closed (if you like, measure from one end of the cable to the other. You should read zero -- remember, you are measuring relative voltage).

Now, back to measuring between the positive battery terminal and ground. Crank the car briefly, and observe a drop of a volt or so. Why? Because current is flowing through some internal resistance in the battery, just as the down-pipe in our water model is not entirely "friction free." Measure again at the starter terminal, and observe a slightly greater drop when cranking. Why? Because now you've added in the extra resistance of the cable. Cranking a starter takes lots of current, so even a little increase in resistance will give a noticeable voltage drop.

One more thing: measure between the positive battery terminal and ground, start the car and run it up to 2000 RPM or so. Notice that the voltage reads more than 12 VDC at first, gradually dropping to a stable 12 volts. What's going on? Your generator (or alternator) is like a pump that returns water from the well into the tank -- and it has to pump it over the top of the tank, a "height" greater than 12 volts. The voltage regulator is like a float in the tank -- as the tank fills, it gradually closes the valve on the pump. Even a few seconds of cranking lets enough "water" out of the "tank" to cause the "float" to open the "pump valve" fairly wide.

Abnormal operation
You should now be able to figure out what the problem is when the day comes that your starter cranks sluggishly or not at all. Remember that the "well" -- electrical ground -- has to be there for current to flow. What would "take the well away?" Dirty connections at the negative battery terminal or at the strap bonding the engine to the car's body, most likely. We'll proceed on the assumption that you've ruled these out.

1) Ignition off. Is there a full 12 volts between the positive battery terminal and ground?

• Yes: proceed to step 2.
• No: Battery is either not holding a charge (the "tank" is leaking, or has not been refilled for so long that the "water" has evaporated), or the charging system is not "pumping" it full. If it's possible to start the car, rev the motor a bit and see if you get 13 - 14 volts at the positive terminal. If so, the charging system is working and the battery's bad.

2) The following measurements are made while attempting to crank the starter -- do these with a helper (and don't crank for more than a few seconds at a time). Does the reading at the positive battery terminal drop about a volt (check this right on the terminal, not on the cable clamp)?

• Yes: current is flowing normally all the way through the system, so your starter is likely shot or you'd be cranking (or maybe your motor's siezed up).
• Voltage drops a lot: battery cannot handle the current and must be replaced (the "tank" is full of silt; it's 12 volts high, but the "water" can't drain freely -- batteries do "silt up" chemically).
• Voltage barely drops: low current. Now measure from the terminal to the cable clamp, then at the other end of the cable (between the cable eyelet and the terminal stud on the starter). Wherever you find measurable voltage (a "pressure difference," remember), you are observing resistance to the current. Disconnect the battery, clean those connections and try again.
• Voltage still barely drops: starter or solenoid are shot. Either the solenoid contacts are burnt and not switching voltage to the starter motor, or the motor's brushes are too far gone to conduct.
• No voltage drop at all: no current means no connection somewhere -- an open circuit. Is there 12 volts between the solenoid terminal (where the small wire goes, not the battery cable!) and ground when attempting to crank?
  • Yes: solenoid is extinct.
  • No: bad ignition switch or associated wiring.

And finally...
Power (watts) is a function of voltage and current. 12 volts does not represent very much "pressure," so, in order to get enough watts on hand to run a motor or light a bulb, everything in a car is designed to use lots of current. Because of the low voltage and high current, even a small amount of unwanted resistance can have a serious effect -- that's why it is so important to keep all connections shiny clean and as conductive as possible.

Just for the record, I can think of lots of problems in relating the waterworks model to an electrical circuit. For one thing, electrical current flows in the opposite direction. A "closed" switch equates to an "open" valve, and vice versa. Don't let this make your head hurt! I've worked in electronics for twenty-five years, and it still helps me to think of water and pipes.

I hope it will help you, too.