Testing a fuel pump with an oscilloscope may sound like overkill when you first hear it, but if you look closer you will see why I have opted to use this method over pressure and volume testing.

Typical roller vane type automotive fuel pumps (running loaded) draw between three and six amps at 5,000 to 6,000 rpms (approximately) when they are new. They are designed to have a useful life expectancy that will usually take them just beyond the warranty period. Unfortunately, they do not just roll over and die in their final days. Generally, they start to fail in one of three ways: brush, armature or bushing wear. When the brush material wears away, the spring tension lessens, increasing the resistance at the brush contact point on the armature. The increased resistance lowers the current flow, slowing the fuel pump speed. Symptoms may include surging, low power (especially on hills), hard starting cold, hard restarting hot (usually with less than half a tank of fuel), rough idling, dying at stops, etc. Replacing the fuel filter at this point will generally give temporary relief from some of these symptoms. The fuel pump will usually pass a pressure test at this point and may or may not pass a volume test. This is perhaps one of the most misdiagnosed conditions a car may have. Figure 1 is a waveform from a fuel pump in this borderline condition.

This one came from a 1989 Chevy Astro 4.3 liter Vortex engine. The complaint was "hard starting cold." The owner thought it might have something to do with the radio drawing the battery down because he could get it to start with a jump. The battery was less than a year old. He had determined that not enough voltage was getting to the computer or module after sitting overnight. He had already tuned the engine, replaced the oxygen and MAP sensors, TPS, coolant sensor, cap, rotor, wires, plugs, fuel and air filters, ignition module and ECM. He said it had good fuel pressure (58 psi), and if I could just find out what was draining the battery overnight, he was sure that would fix it.

After not finding an electrical system drain, and test-driving the vehicle (which ran fine on the highway), I let it stand overnight. Sure enough, the next day it would crank fine but would not hit a lick. I tested the battery voltage to see if the owner's theory had any merit and found it to be cranking at 11.6 volts. I connected an ST 125 spark tester to a plug wire and cranked the engine again. It had a good spark. I connected the scan tool and found no codes and the presence of a cranking rpm signal. Being of sufficient experience not to rely totally on test results gathered by DIYers, I did my own fuel pressure test. It had 56 psi at key-on. It looked OK to me (not being familiar with this engine at the time). Since the injector is buried under the intake on this model, I couldn't get to it to observe a fuel spray. I thought I would try the owner's suggestion of jumping it just for the heck of it and it started right up. I shut it off and disconnected the cables. It restarted and ran fine. I could not get it to duplicate the hard starting condition so I thought I would try to duplicate the battery condition. I turned the headlights on and let it stand for not more than two minutes and it failed to start again. I noticed that the fuel pressure was 52 psi cranking when it would not start, and when I hooked up the cables it went up to 54 psi when it would start, then up to 56 while it was running.

Could this slight difference in fuel pressure keep this engine from starting? Surely not. I deadheaded the fuel pump and rechecked the pressure - 60 psi. Was that enough? I was not sure.

I connected my lab scope and amp probe to the fuel pump circuit and collected waveform No. 1 in Figure 1. The DC average was 168 mV, which equals 1.68 amps. I had enough experience with TBI and PFI pumps to know that this was a low reading. Assuming eight armature commutator bars at 2 ms per division, it takes nine divisions to complete eight humps on the waveform pattern (0.018 sec). Divide one by .018 = 55.55 revolutions per second. Multiply that by 60. The pump was running at 3,333 rpms.

I was convinced from my prior research that this was a slow-running pump. I ordered up a new pump and installed it in the tank. After discharging the battery with the headlights for a full 10 minutes, it started every time. The key-on pressure was 56 psi even during cranking. The deadhead pressure buried my 80 psi gauge.

The new pump waveform (waveform No. 2 in Figure 1) revealed an increase in the DC average. It went up to 4.84 amps. The operating frequency of the new pump went up to 5,454 rpms (11 divisions at 1 ms per division to go one revolution; 1/0.011 = 90.9 X 60 = 5,454). Subsequent testing of Vortex fuel pumps has shown 3,333 rpms to be the magic number for a pump exhibiting the "no-start after sitting overnight" condition. The corresponding fuel pressure seems to be 52 psi "no-start," 53 psi "start."

In the two years since I collected the waveforms from this vehicle, I have done extensive research with hundreds of fuel pumps and found about two-thirds fail from the gradual decrease in current flow and subsequent drop in rpms usually exhibiting the first driveability symptoms at just over 3,000 rpms. The other third are the ones that cause the real headaches. They are the ones with complaints of occasional no-starts. There are no particular patterns except they don't seem to ever fail to start in the presence of a technician. They also generally will pass a pressure and volume test. When one of these vehicles is towed to a shop, you can bet it will restart as soon as the wrecker driver sets it down. They cause a lot of unnecessary parts replacement. The frequency of the pump often times is not much less than that of a new one. The average amperage draw usually will be similar to that of a new pump in the early stages.

Figure 2 is a classic example of one of these types of fuel pump failures. The dead giveaway here is the sharp decrease in the current flow in one or more spots on the fuel pump armature pattern. These drops are caused by a spot on the armature where the commutator bar is worn completely away and the brush makes no contact for an instant. If the fuel pump happens to stop on the spot, the result is usually a no-start until the fuel pump cools off or is vibrated. Generally, there will not be any other driveability complaint in the early stages of this type of failure.

This particular pattern came from a 1989 Chevy Blazer owned by my cousin. He called me one Saturday from his cell phone. He was fishing at a remote lake in Indian Creek Reserve when his vehicle failed to start. He was attempting to remove the tank in a gravel parking lot based on the fact that he could not hear the fuel pump running when he cycled the key. He had the tank partially dropped when he realized that the job was a bit more than he had anticipated and he was not 100 percent sure of his diagnosis. He called me to find out if there were any tricks to dropping the tank. I quizzed him about the symptoms and told him to try to start the truck. I told him if it started, to re-secure the tank and drive it to the shop where I would test it for him. When he arrived at the shop, he informed me that it had failed to start for his wife at the grocery store one day a couple of weeks ago. He said that when he got there it fired right up and had not acted up again until today. I nonchalantly went for my scope and probe and connected them to the fuel pump circuit. I collected the lower waveform in Figure 2 while reminiscing with my cousin about the good old days. He followed me to my computer where I downloaded his fuel pump wave to my software program. I typed the words "bad pump" in the data box and dragged it down to his waveform. My cousin got quiet for a few minutes while staring intently at the screen and finally said, "What the heck are you doing?" At that point, I was happy to fill him in on the new technology which led me to my unshakable conclusion that his fuel pump was responsible for the intermittent problem he was having.

After replacing the pump, I took the new pump wave in Figure 2 and showed him the difference. I explained that the low spots were caused by a bad spot on the armature and dissected the pump to show him. Sure enough, there was an obvious place on the armature where the commutator bar had worn completely through. My cousin was spellbound and requested a copy of the waveforms, which I was happy to provide him. He called me later to let me know that although he had not noticed any loss of power prior to replacing the pump, he did notice that it ran a lot better with the new one.

This test method has proved itself time and again to be the best way to check fuel pumps - so much so that I seldom get out my fuel pressure gauge.

The current probe has enabled me to turn tough to find intermittent problems into routine maintenance (Crocodile Lemonade). I test fuel pumps like I look at brake pads. How much life is left in it? And should I recommend replacement yet?

I like to print a copy of the customer's fuel pump wave and a new one so I can show them the difference. Once they authorize the repair, I print another with their new and old pump. Of course, one of the most popular questions I get asked is, "Do you mind if I take this to show to my husband?" Figure 3.

Jeff Bach is the owner of CRT Auto Electronics, an ASA-member shop in Batavia, Ohio. For more information on this topic, contact Bach at (515) 732-3965. His e-mail address is northstarguy@zoomtown.com

RATE THIS ARTICLE What do you think of this article? Your input will help AutoInc. develop additional articles on this subject. Share your thoughts! Your name Your e-mail address

MOST ACCESSED ARTICLES Fuel Injection Service, Not Just Cleaning The Art of Extraction EGR Systems: Operation and Diagnosis Proactive Target Marketing:_Rethinking Your Business Strategy Engine Performance: HO2S Diagnostics MOST E-MAILED ARTICLES Developing Employee Potential How Critical Thinking Can Help Your Business How to Diagnose the Ford Glow Plug What to Look for When Shopping for the Right Shop Management Software Putting a Price Tag on Complaints

AutoInc. Web Site | ASA Web Site | Halftime for the 105th Congress | Proper Timing Tips to Get Under Your Belt | Assessing the Quality of Aftermarket Collision Repair Parts | Women Want Positive, Receptive Customer Service | Top 10 Automotive Repair Web Sites | Guest Editorial | Tech to Tech | Tech Tips | Shop Profile | Net Worth | Stat Corner | Chairman's Message