Engine Diagnostics: Scopes & Multimeters Uncovering the Hidden Culprits of Driveability Woes, Larry Carley, ImportCar, May 2001

Seeing the invisible. That’s the key to diagnosis today. In order to figure out what’s causing an engine performance or emissions problem, you have to "see" what the sensors are doing electronically and how the engine control system is responding. Trouble is, you can’t directly see electrons. So you have to use various diagnostic tools to indirectly observe and measure their movements. You do that by measuring things like volts, ohms, amps and frequency.

One of the most basic diagnostic tools for doing this is a multimeter that combines the functions of a voltmeter, ohmmeter and ammeter into one. Analog meters are fine for measuring the ebb and flow of electrons in electric circuits. You can use them to check battery voltage, charging voltage, circuit voltage, relays, solenoids and a zillion other things. Even so, analog meters have certain limitations.

An analog ohmmeter can be used to measure resistance and check continuity in normal electrical circuits but care must be used when working on electronic components. An ohmmeter works by applying a small voltage through its test leads, and this voltage may be enough to damage sensitive electronic components. So for electronic testing, a high-impedance 10 megohm digital multimeter is required and, even then, caution must be used to avoid directly probing certain items such as control modules.

Another limitation of analog meters is the needle scale. Though a meter may have more than one scale to allow for more accurate readings, a needle is never as accurate as a digital reading. A digital reading can give an exact voltage, resistance or current value to tenths, hundredths or even thousandths of a volt, ohm or amp, depending on the scale you select. Exact voltage readings to the nearest tenth or hundredth of a volt may be required when adjusting a throttle position sensor, for example. But digital multimeters (DVOMs) have their limitations, too.

For one thing, it’s nearly impossible to read a rapidly changing value on a digital meter. The numbers change too quickly to be readable. An oxygen sensor’s voltage output, for example, may cycle back and forth from minimum to maximum too quickly for an ordinary DVOM. Or, the display may not update fast enough and lag behind the signal input.

A digital reading on a DVOM may also totally miss a sudden change in voltage, resistance or current that occurs too quickly for it to detect it. One way to check a throttle position sensor is to open and close the throttle linkage while observing the sensor’s output. There should be a smooth and steady change in the sensor signal as it moves from idle to wide open throttle and back. But if there’s a dead spot in the sensor, it can produce an abrupt change in the signal as it passes the dead spot. You might never see it on a digital multimeter. For this type of detective work, you need something that can display a voltage signal as a waveform.

SCOPES & GRAPHING MULTIMETERS
There are two types of diagnostic equipment that can display voltages and frequencies as waveforms: oscilloscopes and graphing multimeters (see photo). Both types can also split the screen and display two to four waveforms using different inputs depending on the capabilities of the unit.

The old-fashioned cathode ray tube (CRT)-style oscilloscopes that were used years ago for analyzing ignition patterns have been replaced by hand-held digital storage oscilloscopes (DSOs). The latter are cheaper, easier and faster to use than their laboratory counterparts.

What makes scopes and graphing multimeters so useful today is that they can take something invisible like a voltage or frequency signal, and transform it into a visible waveform on a display screen. Seeing a waveform instead of a rapidly changing number makes it much easier to tell if a sensor or circuit is functioning normally or not. Some scopes and graphing multimeters have an electronic library of known good signals for comparison. Some even include wiring diagrams and a vehicle specific database of diagnostic and test information.

You can also compare several waveforms at the same time if the scope or graphing multimeter has "dual trace" or "multiple trace" capability. This can be very useful when checking the oxygen sensor response or changes in injector duration against inputs from other sensors such as the MAP sensor, air flow sensor, throttle position sensor or coolant sensor.

One thing to keep in mind about scopes and graphing multimeters is that they are not a substitute for a good scan tool. You still need a scan tool to read fault codes and serial data. But scan tool data alone doesn’t always give you a complete picture of what’s really going on inside the system – especially when you’re dealing with an intermittent fault or a momentary glitch, or if there’s a problem that hasn’t set a code. That’s why some high-end scan tools also combine the functions of a DSO or graphing multimeter into one unit.

The drawback to using scan tool serial data to diagnose sensor problems and other faults within the vehicle’s on-board electronics is that serial data is not "real" data. It’s the computer’s interpretation or report of what it thinks it sees – which may not necessarily be what’s really going on electronically in its input and output circuits.

Let’s say a car has a hesitation problem and you suspect the throttle position sensor (TPS). You look at the output voltage of the TPS with your scan tool and watch the numbers increase then decrease as you open and close the throttle. The TPS seems to be OK, but is it? If there’s a momentary dead spot in the TPS (which typically occurs between idle and part throttle where wear is greatest), the serial data that you’re seeing may not reveal the dead spot in the TPS. Even if you’re using an analog voltmeter to directly read the TPS, the needle may not respond fast enough to detect a momentary dead spot. What’s more, sometimes a TPS will read fine when opening and closing the throttle slowly, but skips when the throttle is snapped opened quickly. But you may never see the glitch unless you have a means of viewing the TPS output signal itself.

What a scope does is translate an electronic signal into a pattern or waveform on a screen. As the waveform is traced across the screen, it creates a signature of the signal’s characteristics – including any momentary glitches that may be causing a problem.

READING WAVEFORMS
The most difficult part of reading a waveform is the initial setup of the scope itself, but the menu-driven setup screens in most scopes and automatic presets helps simplify the process. Instead of entering vehicle year, make and model or VIN number as you would with a scan tool, you tell the scope how you want the signal data displayed. This includes setting the voltage scale and time base.

A scope displays voltage on the vertical scale and time along the horizontal scale. You pick a voltage scale and time base that allows you to see the entire waveform and also makes it large enough so you can see all the important details.

Next, you have to tell the scope when to start displaying the signal unless this is done automatically (which it is on some scopes). This point is called the "trigger level" and is set to a specific voltage value. You also have to tell the scope which way to draw the pattern (up or down) when the signal voltage passes the trigger level.

By comparison, a graphing multimeter is somewhat easier to set up and use. When you select the graphing function, the unit begins to record the input signal and display it as a graph similar to a strip-chart recorder. This creates a waveform trace similar to that on a scope, with the x-axis (horizontal) displaying time vs. the y-axis (vertical) showing the changing value of the input. Changing the time base can compress or stretch the waveform trace to suit the type of test being performed. The upper and lower viewing limits can also be changed by using the cursor to set the values.

Another thing that’s different with a scope and graphing multimeter is how you connect these tools to the vehicle. Unlike a scan tool that simply plugs into a diagnostic connector, these tools require you to hook up test leads using connector backprobes or wire taps. Most vehicle manufacturers don’t like technicians poking holes in wires. Even so, if you use "Hirschmann"-style probes, you’ll make only tiny holes in the wiring, which can easily be resealed afterward with a dab of nail polish.

In addition to learning how to connect and use a scope or graphing multimeter, you’ll also have to learn how to read waveforms. This includes understanding the basic types of electronic signals (direct current, alternating current, fixed pulse width/variable frequency and pulse width modulated) and how to tell a good waveform from a bad one. This includes looking at things like signal amplitude, frequency, shape, pulse width and overall pattern.

To get the most out of these tools, you’ll also have to learn what the basic waveforms for each type of sensor and other device are supposed to look like. This is the hard part because waveforms vary a great deal depending on the vehicle application.

Different types of fuel injector drivers, for example, produce different waveform signatures. Some produce a single spike when the computer opens the ground circuit (saturated switch-type injector drivers like those used with Bosch multiport systems). Some produce a double spike while others produce an inverted spike. The height of the voltage spike as well as where it occurs on the waveform can reveal electrical problems within the injector solenoid or computer driver circuit.

A shorter than normal spike, for example, would be characteristic of a partially shorted injector solenoid. A simple resistance check with an ohmmeter might not reveal such a problem. This is just one example of the many things that you can’t see with a scan tool or multimeter but you can see with a scope.

SNAPSHOTS
Something else a digital storage scope or graphing multimeter can do is capture snapshot data when a problem occurs. One of the main limitations of a CRT lab scope is that the information it displays is live. With a DSO or graphing multimeter, the data comes in, is processed, then displayed as a slightly delayed signal. This means it can be recorded to create a histograph and allows you to slow down things and hopefully catch intermittent problems that might otherwise escape detection.

The sampling rate of a typical digital scope is normally around 25 million samples per second, which is fast enough to catch even the most momentary glitch. Depending on the scope, this can usually be increased to an even higher rate. Some scopes offer a "spike detect" mode that jumps the sampling rate up to once every billionth of a second! At this rate, the waveform contains much more detail and noise, but also reveals problems that might be overlooked in the normal sampling mode.

Most digital scopes are also hand-held units with LCD displays, which makes them easily portable. If you think a scan tool or flight recorder can help you catch a momentary glitch during a test drive, you haven’t seen anything until you’ve taken along a scope. You’ll see things you’ve never seen before, and catch problems you never would have caught before.

One of the most powerful diagnostic applications for a scope is oxygen sensor testing. A scope can tell you if the O2 sensor is capable of producing a good signal even if the sensor is reading rich or lean. The scope can also allow you to use the O2 sensor’s waveform to verify that the computer’s feedback fuel control loop is functioning properly.

When you look at an O2 sensor’s output with a scan tool, you see only a voltage value or a rich or lean indication. You can also look at cross counts to see if the sensor is flip-flopping back and forth from rich to lean at an acceptable rate. You can also check the sensor’s rich and lean response by making the fuel mixture rich (by feeding propane into the intake manifold) and then lean (by pulling off a vacuum hose) to see if the sensor responds as it should. Yet a sensor that passes all these tests may still be causing problems if its waveform is bad or full of noise. That’s where a scope comes in. It shows you everything you need to know about the sensor’s output in one simple picture.

You can see at a glance if the sensor is reading rich or lean, what the sensor’s peak and minimum voltages are, if the sensor is flip-flopping from rich to lean at a normal rate, and how it responds to changes in the fuel mixture. You can also see if the signal is clean or full of noise. If the scope you’re using has dual-trace capability, you can also display the injector driver waveforms at the same time to see if the feedback loop is changing injector duration in response to changes in the O2 sensor signal.

The O2 sensor signal is like an EKG of the entire engine because it can also reveal other problems such as vacuum leaks, ignition misfire, injector imbalance and even compression losses. Each of these conditions will produce a characteristic type of hash in the sensor waveform. Anytime that a cylinder misfires or leaks compression, unburned oxygen enters the exhaust. This shows up as a momentary dip in the O2 sensor’s output voltage. So if the O2 sensor’s waveform contains lots of little inverted spikes, it tells you the engine is misfiring or leaking compression.

A scope or graphing multimeter can also be used as a repair verification tool. If you "baseline" a vehicle before repairs are made (capture the O2 or other sensor waveform that reveals a problem), you can then compare "before" and "after" waveforms to make sure the problem has been corrected.

You can also use a scope or graphing multimeter to check the "V-ref" voltage in sensor circuits. Unlike a digital voltmeter that only gives you a number, the V-ref voltage on a scope appears as a flat horizontal line. Though not very interesting to look at, it can reveal hidden problems if the line is full of noise, has spikes or breaks up. The same technique can also be used to check battery voltage and wiring continuity. If the line breaks up or dips when you wiggle a connector, it tells you there’s a problem.

Displaying the charging output voltage as a waveform can also help you spot bad alternator diodes. The normal AC output pattern of the alternator should look like the top of a picket fence. If any tops are missing, it indicates one or more bad diodes.