Tech Tip 2
This month’s tip is a case study involving a 1993 Jeep Cherokee with a 4.0l engine.
Symptoms are as follows: Customer calls and says" My truck won't start and I am having it towed in for you to look at. Oh, and I need it back by the end of the day."
Those are the kinds are the kinds of phone calls I like, it's broke and I want you to fix it now! Oh well, we do our best.:)
Well the tow truck shows up and drops off the Jeep, tow truck driver (who must have been a technician in another life) says "Well it cranks over but it wont start must be a distributor problem!" I tell him thanks for the technical opinion and I will be sure to keep that in mind. :)
With work order in hand I proceed to the truck. Hop in the drivers seat, put the key in the ignition and crank it over....low and behold it fires right up. Oh great, an intermittent problem just what I need!
One of the first tests I perform on any no start, whether it be intermittent or constant, is a current ramp of the fuel pump. Now a low current probe used with a lab scope is a very useful tool and a scan of the iATN archives will let you know why. Really smart technicians are coming up with new and extremely useful tests for this very versatile tool every day.
A low current probe is especially useful when fuel pump testing as you can actually watch the analog signal of the pump motor spinning. Figure 1 is the capture I got from the Jeep with the engine running.
Fig. 1
Note how every fourth hump is missing.
Okay so here is how this works: Most pumps have 8 commutator bars and as the brushes pass over the commutator bar current flows thought the pump and causes it to spin. This is how our pattern is created every time the brush pass over a space on the commutator a drop in current is produced and thus we get a nice M shaped wave form when the brushes and commutator are all good. See figure 2.
Fig. 2
Note how the 4th and 8th commutators are not producing any current flow.
What happens to cause a no start is that the pump stops on the bad commutators and when the pump receives power there is no current flow and therefore no pump operation. When the vehicle is towed in on the tow truck it will bounce around a bit and the pump will come up on a good bar and the vehicle will start.
Okay fast forward two hours and one new fuel pump (and filter of course). Now lets take a look at figure 3.
Fig. 3
Nice pump waveform!
Hey! Now isn't this a useful test, we just diagnosed an intermittent no start in about 5 minutes. And no we didn't charge the customer for only 5 minutes of diagnostic time. When we charge for diagnostic time we are not really charging for time but for KNOWLEDGE! But that is another subject all together :)
Lets look at another towed in no start.
Now lets look at a 1993 Dodge Dakota with a 5.2l engine.
This one was definitely a no start even after the tow truck ride. Lets break out our trusty low amp probe and lab scope and take a look at the fuel pump current wave form. See figure 4.
Fig. 4
WOW! Now that is a terrible pump!
Lets look at that again in figure 5.
Fig. 5
Oh, look if you kick the tank the pump wave really looks good!
Okay that's enough of that.....oh okay one more time. Figure 6.
Fig. 6
Well...it almost started that time.
Okay now for the good pump waveform. You get the idea of what a bad one looks like. Figure 7.
Fig. 7
That is a much nicer waveform!
You can also calculate pump RPM like this: 60,000 divided by the time of 8 humps (or # of commutator bars). See figure 8. After doing a number of cars you can get a feel for what RPM a pump should run on various models.
Fig. 8
Time between cursors is about 18.5ms. So 60,000 divided by 18.5 = 3243 RPM
So with a few strategic scope captures we are able to quickly and accurately diagnose a bad fuel pump. They aren't all this easy but it is sure nice to be able to use DIAGNOSTIC STRATEGY and work smarter not harder. The key to successful diagnostics is knowing the system you are working on.
If you have any questions about this month's tip please feel free to contact us.