Several days after the missed appointment a lady walks through the front door and reports that she is dropping a car off for us to look at. She introduced herself as only the one who called about having to jump start the car and that someone told her it needs a fuel pump. We wrote it up and I set about working it into the schedule. 

When the time came to look at it, wouldn't you know it fired right up. Well that's OK, we deal with intermittent issues all the time. I just need to get my testing set up, perform my baseline checks and then wait for the problem to occur. One of the tests that I run is to measure the fuel pump current with a low amps probe an my digital storage oscilloscope. Here is the waveform from that test.





The pump definitely has a problem, those current humps should be very similar. You can see the two humps right after the cursors, and how much different they are from the rest. That's a bad commutator to brush connection. But, they claimed they need to jump start the car when it acts up. Trying to jump start it won't fix the pumps problem. If it would happen to not want to run, jumping it wouldn't help. So either they have given me false information or there is more than one problem with this car.

I let it run for about a half an hour and there were no other issues noted. It was time to turn it off and let it sit and then try to restart it in hopes that I could catch the no-start. About an hour later the car started and ran for a few seconds and then stalled as the fuel pressure dropped to 10PSI. Here is what the fuel pump current was then.

That's a little over fifteen amps of current and notice there are no humps in the pattern. That's a pump that is frozen and not turning. The car was now a cranks but won't start. I can see if this happened to someone they might have to wait for help to come along, and maybe in the mean time have cranked, and cranked the engine until they ran the battery down. Now when they have been forced to leave the car sit for a while, and they jump it, that gets it to start but the fuel pump running again is not actually related to the jumping of the battery.                             

The low amps probe is a great tool for testing like this. Had the problem been a bad connection reducing power to the pump, there would have been less current flowing than the first capture above, and there still would have been some humps from the pump turning, although they would have been very slow. These last two captures show a steady line until the relay turned off. Now its just a matter of selling the repair.

To be continued..