Quick no-start diagnosis

Order Reprints
Quick no-start diagnosis

About the author: Truglia is the owner of Car Clinic, a state-of-the-art repair facility in Mahopac, N.Y. He is ASE certified with a M.A. from Columbia University. In the automotive world he has been trained by Technicians Service Training and Automotive Technician Training Services. Car Clinic’s facility is fully equipped with factory-level equipment and services American, European and Asian vehicles, including diesels and hybrids. (All vehicles were diagnosed by G. “Jerry” Truglia, Alex Portillo and Craig Truglia.)

Don’t let those no-starts frustrate you. Unlike intermittent faults we have to wait for, the no-start is ever-present and can be isolated in about five minutes. Once we can figure out our diagnostic direction, then we can nitpick exactly what’s wrong with the vehicle.

Sparks, fuel, mechanical

Engines need three things to run. Spark ignites the fuel and if anything from the spark plug to the PCM that grounds the coil is bad, the car will not move. Fuel is needed for the controlled explosions of engine combustion to take place. So, anything from the fuel injectors to the fuel pump to the PCM may prevent fuel from getting to the engine and allowing combustion to take place. Lastly, you need a good working engine! Bad timing (jumped timing belt/chain), low compression (washed down cylinder walls), or high exhaust back pressure (clogged catalytic converter) can prevent an engine from working properly.

Now, why does my subject header list spark first, then fuel, and lastly mechanical? This is because in the real world a no-start is usually caused by the same parts that cause misfires. Sure, bad engine compression can cause a misfire, but how many more times does the correct repair involve replacing spark plugs or the ignition coil? The same applies to no-starts. Look at the most likely culprits and then work your way down the list.


Five-minute no-start diagnosis

With the information we learned up to this point, we can diagnose almost any no-start in five minutes. How? Here is the five-minute plan of attack:

• Spark: Check for spark with a spark tester. Screw the ignition tester all the way out past 30K in order to make sure there is adequate spark to start the engine.

If no spark is visible, do the following: Perform an rpm check by using the rpm PID on your scan tool. If you have rpm, that means you have primary ignition on most vehicles, and there isn’t an issue with the crank sensor.

• Fuel: Add propane to a manifold vacuum source, vacuum line, or brake booster intake hose to see if this gets the engine started. If the engine starts turning, you have a fuel delivery problem.

• Engine mechanical: Do a cranking vacuum test to see if the valves are opening and closing. Connect the vacuum gauge to a manifold vacuum source, such as the brake booster. When hooking up to the brake booster, make sure that there is not a check valve in the line. The engine’s valves are opening and closing properly if we have a bouncing needle between three to five inches. Don’t forget after disconnecting the vacuum gauge and connecting the power booster hose to pump the brake pedal so you have brakes! 

• Exhaust backpressure: Do this test to rule out a clogged cat. Connect an exhaust backpressure gauge with an old oxygen sensor or backpressure tester before the converter. Good exhaust backpressure at idle should be at 0 psi. More than 0.5 psi at idle, 1.5 psi at 2,500 rpm, or 5 psi at snap throttle tells us that there is a problem. Rising exhaust backpressure will decrease intake vacuum. So, if you can accurately measure intake vacuum by looking at the MAP PID, decreasing voltage reflects an exhaust restriction.

Case study: No-start 1997 Caravan 3.8L.

NOTE: The vehicle in this case study is old, but the essentials of diagnostics remain the same. This vehicle was diagnosed and repaired in 2012. Here we only fixed the no-start. This vehicle had an unrelated stalling problem that we fixed months later.

A customer came in with a complaint of an intermittent no-start and stalling issue. The vehicle had a system lean and a MAP sensor DTC. We replaced a vacuum line and the wiring to the MAP sensor. No more no-starts, right?

Wrong. The vehicle was no longer running lean, but the customer came back to tell us that the vehicle was still stalling. Ironically, the vehicle did not have a check engine light and it refused to start in the parking lot. Using the old “propane in the brake booster hose” trick we were able to drive the vehicle to the bay. So, we not only saved ourselves the pushing, we proved there was a fuel problem. At this point we found out from the customer that the vehicle had an aftermarket fuel pump and a new fuel filter installed about a year ago. The vehicle also started now, too. We assumed the no-start was related to the stalling issue, so this shaped our diagnostic direction. So, we went right for the fuel pump, pulling out the relay and looking for an amperage waveform.


Using iATN, we checked the waveform library and the waveform looked good. It was at about 5.5 amps and it had uniform humps.

The vehicle was starting and not stalling anymore, so we hooked up for the road. A trick we like is using painter’s tape to hold anything to glass, including a fuel pressure gauge to the windshield so the technician can observe what’s going on when driving.

Our hookup also included a labscope and scan tool. Being that intermittent no-starts and stalls are hard to catch, we like to use many channels so we can measure as many signals as possible.

Presently, the ATS EScope has the most channels, up to eight, and a deep recording function that makes it easy to catch the intermittent fault because you can set it and forget it. The drawback of PC-based tools like the EScope is that they use a clunky laptop that needs to be charged. We get around this by using a $30 DC plug that goes into the cigarette lighter/auxiliary power port. Then, run some wires through the window, plug in your OTC Genisys, and voila, you are a mobile diagnostic machine.

The vehicle was driven for a total of one hour and never once stalled. In fact, it ran great. We then realized something was wrong: We had the fuel pump relay bypassed in order to measure fuel pump amperage. Therein was the answer. The way we were hooked up prevented us from checking the integrity of the fuel pump relay.

The car was running great because everything was good— besides the relay! Take the relay out of the equation, and the car will run fine. Based upon this hunch we took the relay apart. We found the windings were damaged on one side and not the other. When the fuel pump went bad, it must have been running at a higher amperage which damaged the old relay. The Caravan no-start was fixed for $14.50.

Special intermittent battery, starter and alternator faults tips

 This article may focus on spark/fuel/engine related no-starts, but that does not mean we should forget the basics. After all, a dead battery is the most common cause of a no-start. All you need is a multimeter, amp clamp, some wire, and a light bulb to diagnose these intermittent no-starts the quick and easy way.

1. Fail-safe battery testing: The “Rich Test”

There are things you can do if you don’t have a computerized battery tester and a load tester. The “Rich Test” is my all-time favorite diagnostic trick because it always works (even on batteries with a low state of charge), it is easy to do, and all it uses is the meter I already own.

Plus, it helps me sell more batteries and prevents comebacks.

1. Put your meter on the battery on DC V and using min/max —

2. start the engine and check the minimum voltage. The voltage should not be below 9.5V.

3. Reset your meter and repeat two more times. If the voltage dips below 9V and continues to decline each consecutive test, you definitely have a bad battery.

Good batteries will simply have the same minimum voltage while bad batteries will lose their surface charge each consecutive start. An intermittently bad battery that you cannot catch due to temperature or anything else can be caught this way. Even batteries before they go bad can be caught “on their way out,” so to say.

Much thanks goes to automotive electrical engineer and diagnostician Richie “Shorts” Peterson in New Jersey, who devised this test. Hence its name, the “Rich Test.”

2. Catching an intermittently bad starter: The “Rich Starter Test”


This test is great for determining whether or not the starter has excessive internal resistance and it’s the cause of the intermittent no-start. Again, Richie “Shorts” Peterson taught us this one.

1. Put your meter with an amp clamp on power side of the starter on DC V and using min/max —

2. start the engine and check the maximum voltage.

3. Reset your meter and repeat two more times. If the voltage increases with each consecutive test dramatically, you definitely have a bad starter on your hands.

Use your common sense with this one. A few millivolts do not make a difference. We’re looking for dramatically climbing peak amperage.

Peak amperage that continues to increase dozens of amps each start indicates an internally binding starter.

3. The control/power side of the starter diagnosis trick

Big surprise: This one also owes much thanks to Richie Peterson. All you need is a wire with a light bulb in the middle. We use this one for those really tough-to-catch no-starts where we know the battery is good and the alternator is charging appropriately.

Simply connect one side of the wire to the signal wire of the ignition switch with a T-pin where the wiring outputs to the starter solenoid and then ground this wire to the chassis ground. Loop the wire so the light bulb is visible through the front window.

The light bulb will light every time the customer starts the vehicle. Now, tell the customer to notice what the light bulb is doing when the no-start occurs. Don’t tell him why, however. It might affect the diagnosis.

Now, when the vehicle fails to start and the light bulb lights, it confirms that the control side of the starter circuit is good and that the solenoid/starter is bad. If the light bulb does not turn on, then something is bad on that control side, like the ignition switch.

4. A few notes on alternators (and a special money-maker on BMW 3 and 5 Series vehicles.)

If you have a bad battery, always make sure nothing made it become bad such as excessive voltage drops on the electrical connections.

Also, make sure that charging voltage is good. Most cars charge 14 to 14.5 DC volts, with the exception of BMWs which tend to be in the mid 13 volt DC range.

If you switch to AC volts on your meter you can measure AC ripple by putting the leads on the positive and negative terminals of the battery. Most alternators will have zero volts AC ripple. A bad alternator might go as high as 500 mV, but anything above 150 mV is highly suspect. Remember, the battery will naturally absorb some of that AC ripple.

On the 2000s BMW 3 and 5 series towed in with a no-start, it is fairly typical that the glassmat battery in the trunk is bad. Test it and buy a replacement. Then, check alternator charging voltage. If it is in the 12V range, you need to change it with an alternator that comes with an updated voltage regulator, which prevents this. NOTE: Some rebuilds will not have this update.

Lastly, check the radiator fan that is right in front of that bad alternator. If you can slow it down with a shop rag, the fan clutch is bad. It needs to be replaced using special tools and watch out, because this involves a reverse thread!

I check every BMW for a fan clutch issue. It is a quick money maker and it can prevent alternators from getting too hot and going bad, preventing the comeback.   ●

Related Articles

Electric fuel pump diagnosis: Houston, we have a problem

Electric fuel pump diagnosis – Part II

ECU soft resets: Killing power to on-board modules does wonders in diagnosis and repair

You must login or register in order to post a comment.