Taking a deep look at starting and charging tests with a lab scope.
EDITOR'S NOTE: Motor Age is pleased to welcome Mac Vandenbrink. Long known to techs as the "guru" of interpreting secondary ignition scope patterns, Mac turns his attention now to applying the power of the scope to analyzing the health of the battery, starting and charging systems.
Over the years, you might have become accustomed to a routine by doing things based on what you learned many years ago. If it has worked for you all this time, then why change? Well, new technology has opened doors to do it faster and better. Let us look at our cranking and charging routine.
Using a lab scope provides us with more data, plus an instant footprint of the test results over time, which can be printed out for the customer.
Once we conclude that the battery as a source is good and the starter is perfect when it gets up to speed, we obviously have to look for the energy loss in another location. And that could be in the cables and connections.
The VAT 40, or equivalent, does not show time and did not see that momentary cranking rpm slowdown within a 0.5 second period (see Figure 1). In fact, you might not even hear it and the customer might have never noticed it.It is a tough job to sell the customer on service if a problem does not show up and the customer did not complain. However, a printout makes it easier to point out to the customer that it might not be a problem now, but when the cold weather sets in with temperatures near the freezing point, the car might not start.
If the customer does not want further testing at this point, you may want to comment by adding, "Do not let the tow truck guy sell you a new battery, because the problem is not in the starter or the battery." Further testing is required to pinpoint the problem.
What we see here is a slight hesitation in cranking RPM. Let us consider two possible causes:
- Pre-ignition or combustion opposing cranking movement.
- Resistance (voltage drop) in cables or connections.
To verify whether it is caused by ignition, crank the engine without spark by holding the accelerator to the floor in clear flood mode. If not equipped with this feature pull the fuse, which very likely controls ignition and fuel.
If the hump is still there, we know that the ignition did not cause it.
(We will deal with voltage drop testing later to include the charging system.)
Where Does Energy Go?
Electrical energy, in general, converts into heat, light or power (magnetic field = power), or any combination of these, that may be expressed in watts. In this case, the objective is power and anything else is a parasitic loss. If it does not turn and does not light up, it has to be in the form of heat. Any resistance in the circuit has a greater effect when current is high.
To illustrate this, suppose that the voltage drop in Figure 1 at Point A was 0.2 volt between battery negative and the engine ground. This would calculate at 30 amps to be 0.0066 ohms. That same resistance at a current of 280 amps at point B creates a 1.84 voltage drop, and that is very significant.
From these numbers we can readily see that a poor connection of only 6.6 milliohms, which is so small that it cannot be measured with standard equipment, has a drastic loss of cranking power. In other words, 280 x 1.84 = 515 watts.
A bigger battery might solve the problem temporarily by compensating for the loss, but a poor connection creates heat and is like a cancer that keeps growing. Heat and moisture cause corrosion. This needs to be explained to the customer. It is simply "fix it now or deal with the consequences later."
Slow Cranking of a Different Nature
The starter is designed to spin the engine for a split second at 150 amps.
Cranking speed is the most important objective. Even a good starter at a reduced speed demands a greater amount of current. This in turn creates greater voltage drop in the circuit and connections. That means more heat in the battery plus more discharge.
This makes the alternator work harder and longer and takes more horsepower away from where the rubber meets the road. It means more loss of energy and more wear to every part in the circuit.
Every time the starter growls when drawing 300 or more amps, it is overheating. That excessive heat at the brushes eventually melts the solder at the commutator bars. Then when the starter is up to speed, it spins the solder out and starts arcing.
In normal use, the starter does not wear out unless it is abused with a slow crank, and that might result in a major expense that could be avoided.
You don't have to be a slick salesman. Just be a good technician who knows what you're talking about and has the evidence to back it up. The customer knows the difference, and that printout will be a point of discussion at his/her workplace the next morning.
Make sure your shop's name is on the printout. Even if there is no problem, the mere fact that you care is what counts. Don't forget that the car is likely the second biggest investment this customer ever makes.
This goodwill gesture is a better investment than the biggest billboard in town. The extra few minutes is a testimony that lasts a lifetime.
Evaluating the Battery
Those of you residing in a cold climate might have experienced the following:
On a very cold morning with temperatures near zero, you turn the key and nothing happens. You try again and there is a slight growl. You try the third time and it sounds like two revolutions. You don't believe there could be any energy left in the battery, but you try again and it starts.
The reason for this is that at near zero temperature the battery capacity is reduced to 60 percent. When looking at the scope pattern, the amount of heat created in Figure 2 at Point 1 = 4.6x800 = 3,680 watts.Turning the key in start position four times brought the battery up to temperature (four times over 3,000 watts). That is a lot of energy for the alternator to replenish. Delayed cranking speed or prolonged cranking at any temperature is not only wasted energy, but is also detrimental to the starter and battery both. Heat destroys a battery faster than anything else. Let us take a few moments to explore the battery capacity.
The important functions of the battery are as follows:
To crank the engine at a sufficient speed.
To recover after a load.
To accept a recharge.
This can all be verified in graphing the test results on the scope.
Back in Figure 2 at Point 1: While most starter lock tests range near 600 amps, this starter takes 800 amps. Note that the battery voltage dropped only to 8.0 volts. This is an excellent test result verifying an excellent battery capacity.
In Figure 2, Point 3: When the starter is up to speed and current drops to about 120 amps, the battery maintains at 11.0 volts. This is also considered excellent verification of battery capacity. One commonly accepted battery spec. is "a load of 200 amps should not drop below 9.6 volts." In this case drawing a line at 200 amps reads 10.5 volts. That is well above the minimum, so the battery passed all load tests.Because resistance testing has replaced the specific gravity cell test on sealed batteries, it is interesting to understand battery behavior using Ohm's Law. When looking at Figure 2, we can calculate what happens to resistance at different points of discharge. The amount of heat generated inside the battery is voltage drop times amps, expressed in watts.
The amount of heat created at Point 1 = 4.6x800 = 3,680 watts.
This illustrates what prolonged cranking does to a battery.
The amount of heat created at Point 2 = 1.6x120 = 192 watts.
Now let us look at the internal resistance of the battery at the same points.
Point 1: Voltage drop =12.6- 8V = 4.6V. R=V/A = 4.6/800 = .00575 ohms.
Point 2: Voltage drop = 12.6-11V = 1.6V. R= 1.6/120 = .01333 ohms.
So, what did we learn from this?
Note that the internal resistance almost triples during 0.5 second cranking period. This clearly shows that a resistance check does not conclude a bad battery, but merely indicates the state of charge.
When energy is taken from the battery, the electrolyte loses the acid content and the internal resistance increases. The charging process forces the acid back from the plates into the electrolyte and conductivity is restored. When a battery is totally discharged, the electrolyte is almost pure water. That is why batteries freeze in the winter and crack the casing.
Many years ago when batteries were still serviceable, each cell was tested for specific gravity with a hydrometer. Today, the state of charge can be tested for by resistance.
The Signature of an Alternator?
Before discussing the charging system, it is beneficial to explore how the alternator works and what makes it fail.Almost all alternator stator windings consist of three coils wound 120 degrees out of phase. This three-phase winding overlap results in a nearly perfect level of DC output using three positive and three negative diodes, as seen above.
Each diode produces a rise and fall of voltage output. The dotted red area illustrates this overlap of opposing polarity, which will cancel each other out as illustrated in the next waveform. This leaves a nice clean ripple of positive polarity.A defective open diode distorts this perfect ripple pattern leaving a gap as illustrated above. The scope captures this electron behavior as a missing output at a high amplitude. So we are losing about 1/6 of the alternator output. This in itself is not the only problem.
The computer might interpret this high pulse as an extra tooth on the crank sensor, and that might cause a shift in timing, resulting in a serious misfire. The amplitude is at its peak when the battery is low and the engine starts up. So the potential misfire might only last until the regulator reduces the field current and the charging output drops off.
This could be a technician nightmare without a scope. The battery functions as a big filter (like a capacitor) reducing the ripple amplitude.When a diode shorts out, it wipes out a complete phase and the scope shows something like above. The total output with a shorted diode is not only reduced by more than 30 percent, but it might also cause some weird drivability problems like a misfire when the headlights are turned on. With more current drain from the battery (like headlights), the amplitude will increase and this might intensify feedback to the computer. You could be hunting for a short circuit between your headlight and ignition, while the problem could be in the alternator.
Again, the battery serves as a filter and absorbs most of the spike, providing there is no excessive voltage drop between the battery positive and alternator output terminal.
What Will Make Diodes Fail?
The main cause of failure is overheating. The housing is the heat-sink of the diodes. The alternator needs to cycle between load and no load to provide enough cooling periods. Adding more high-current demanding equipment, low speed driving or prolonged idling all are contributing factors. For instance, adding a powerful amplifier in the trunk with six speakers may cause frequent alternator replacement.
To allow for adequate periods of cooling, a 50 amp average load requires a 90 amp alternator.A battery with a shorted cell can ruin an alternator, because it might never reach the voltage regulator level. It just keeps producing without a cooling cycle.
Look at the capture in Figure 1 between points A and C. This time we are evaluating the charging test results.
The best location to evaluate the ripple of the diodes is between the point when current flow is maximum and that is between D and E, when the alternator is charging the battery at a steady rate. Beyond point E is the cycling of the voltage regulator making the sampling too unstable for a decent diode waveform.
The diode ripple can be observed in volts and in amps. Let's discuss both.
VOLTAGE Red waveform: It looks almost like a straight line sloping up, and that is good. When we look at the difference in amplitude between a defective diode and the amplitude of a perfect ripple, there is no way that you can miss a defect. Moving the volt lead from the battery to the alternator output terminal makes a remarkable difference. Try it sometime and see just how the battery, as a big capacitor, absorbs the ripple effect.
AMPERAGE Blue waveform: The current scope pattern is not influenced by the proximity of the battery. However the pattern might be upside down from the illustrations. We only want to know good or bad by looking for small ripples versus five times the amplitude.
But is this a valid alternator test? Definitely! There are several observations contributing to that statement.First, this capture shows that the slope energizing the field winding from point D to point E is uninterrupted. Bouncing brushes or a poor connection would show intermittent spikes. Beyond point E is the voltage regulator controlling the output, preventing the battery from overcharging.
The 50-amp charging current on the graph is what charges the battery, and that does not include the current that keeps the engine running. The fuel system, ignition system and cooling system should be considered as alternator output as well.
Second, when there is evidence that every winding of the stator is producing and all diodes conduct 100 percent, there should be no reason to believe that this test is not conclusive. We are looking for possible failures and not for capacity.
Finally, perfect cranking voltage under load supports a fully charged battery before cranking and that verifies that the charging circuit was OK at the start.
What about using a carbon pile? There is no need for an external load to verify every function of the charging system. The cranking was enough drain on the battery to make the alternator work at peak capacity.
However, if there is reason to believe that the alternator capacity is too low for the extra load beyond manufacturer specs, then by all means, use the carbon pile to verify maximum capacity. When you do so, make sure to load just below charging voltage, between 13 and 14 volts, to prevent erroneous reading by loading the battery.
In addition, make sure to place the current probe on the alternator cable and not on the battery cable.
Exploring to Learn
When using a scope for the first time, take some time off your lunch break to experiment with various settings, like changing the current to 300-amp scale and expand the time base to get the scope pattern as illustrated in Figure 5. This will verify that compression is equal on all cylinders. It also shows that the starter is up to speed within one revolution.Of course you need to keep the engine from starting to get this capture. The voltage reading on a 4-volt scale represents the voltage drop between engine and battery negative. Have a close look. This is a clear illustration of how voltage drop relates to load. Don't ask what the settings are supposed to be; just explore on your own.
Another learning experiment is moving the current probe to the alternator output cable and find out the difference in alternator output verses what is pumped into the battery. Turn on the headlights and see the increase in alternator output. Discover that when you hold the current probe close to the alternator pulley or near the back bearing, you get a high current reading, even without hooking the clamp around a wire.
Now you know that the current probe uses a Hall effect device and is very sensitive to a magnetic field, and the rotor is a big magnet. That is good to know in case you get some erroneous readings.
Voltage Drop Test Routine
Voltage drop testing with a multimeter has always been a hassle because you need to create the load while holding the test leads and memorize the reading.
If you have a four-channel scope, it takes very little effort to dedicate the two extra channels to voltage drop testing and the test results are automatically recorded while under load.
Connect one channel to the alternator output terminal and the other to the motor-block. This checks the voltage drop in the ground circuit and the difference between battery positive and alternator. Set the scale you selected for engine ground at 2 volts.
The alternator output should be set at the same scale as battery voltage. Without the engine running, those voltage levels should be right on top of each other. In fact, ideally there should be very little difference between battery voltage and alternator output, even under charging load.
All the data is collected in the graphic display of five seconds (Key ON – Start – Key OFF). You might even want to consider this as a routine hook-up, because both connections are simple and in easy reach.
What about between battery positive and starter terminal? You don't want to do that as an easy routine. If there is evidence of a voltage drop somewhere, and you have ruled out where it is not, you know where to go to fix it or how much to estimate. You know that it is not inside the starter.
Learn by Doing
How would you know what is a good starter pattern? Very simple: get the experience from doing. Hook the equipment up to the next 10 cars that come in the shop and you get a feel for what is normal. Plus, you become efficient and learn to do the hookup and test in less than three minutes.
Let us suppose that you get curious and explore what would happen when you select the 300-amp scale instead of 600-amp, and instead of battery voltage you put the clip on the motor block. Figure 5 is one example.
Look at the result: You learned a whole lot more about the starter current under compression, about valve seating at cranking speed, and about voltage drop between engine ground and battery negative. You also will notice that the starter was up to cranking speed within one engine rpm.
Note: A rough running engine complaint at idle may have many reasons, including poorly seating valves. The lower the rpm, the greater chance of leak down in the compression stroke and the more inhaled hydrocarbon is pushed out in the exhaust. What better opportunity is there than cranking speed to verify or discard this problem? You might as well make the amp-probe do double duty.
A DSO vs. Volt Amp Testers
The scope is time based and records voltage and current variations as it happens. You can see immediately how fast the starter gets up to cranking speed and it illustrates how long it takes to recharge the battery. It even records intermittent connections in split second intervals. It also allows you to lock in observation under load or no-load as a permanent record, for later analysis or to send in an e-mail to the customer.
Whether the Volt Amp Tester is a digital display or uses analog meters, the output is not time related and reads as average or as a final reading.
Case in point is this customer complaint: "My car takes a long time before the engine starts. I think it needs a tune-up." The technician notices a slower than normal cranking speed and hooks up the VAT 40. The test result is cranking amps 250. Cranking voltage is slightly more than 10 volts.
The conclusion is the battery is OK. Cranking amps is higher than normal, which could be because of slow speed. The decision then was made to check for voltage drop.
This is not bad reasoning based on the slow cranking speed symptom. Under the hood was simple, but checking the starter required two people. One man under the car and one to crank the engine.The scope shows what the real problem was. To make observing easier, this capture in Figure 6 was expanded times four. The brushes are designed to make and break contact with the commutator. To cause the current to swing from 600 to zero indicates the brush contacts are more breaking than making.
Only a scope is designed not only to tell that story but also to capture and save the evidence.
Mac Vandenbrink is owner of Dynamic Auto Test Engineering Corporation (DATEC), a company focused on designing new training concepts and teaching them to the automotive aftermarket.
