2001 GMC Jimmy 4.3, No Power on Acceleration, Fuel Delivery Problem

(This is not the actual vehicle we worked on. We did not get a picture of it, so we are just showing a comparable one. The one pictured is a 1999 GMC Jimmy from http://www.edmunds.com/gmc/jimmy/1999/.)

***This entry will not be so much a "how-to" as it will be a teaching point about testing a fuel pump's circuits before removing it from the tank. Therefore, there will not be as many pics in this post. We will simply discuss what we did and then focus on the test, with a couple of pics to go with it.***

This vehicle came into the shop with a customer concern of no power on acceleration. The customer described the symptoms as bogging when pressing the accelerator. The customer believed the fuel filter could have been the issue.

We drove the vehicle and verified the customer's concern. There was a definite issue with fuel delivery. We checked the readings on the Scan data and found the O2 sensors were reading lean at WOT (i.e. Wide Open Throttle). At WOT, with proper fuel pressure and volume, the O2 sensors should peg out Rich. In this case, they did the opposite, they dropped to 0mv, indicating a fuel delivery issue.

We put a fuel gauge on the vehicle and got a reading of around 50psi. This vehicle's spec is 60psi and up.

We removed the fuel filter first and sure enough, it was completely clogged. Per the customer's request, we changed the filter first and test drove the vehicle again. As we assumed, the vehicle had the same symptoms.

It was now time to drop the tank and take a look at the fuel pump. The filter was an obvious issue. It was stopped up due to a lack of proper maintenance.

The pump probably failed because of the poor shape of the filter. The pump had to pump against that resistance and simply wore out because of it.

One mistake that we have seen committed over and over, is having a faulty fuel pump and then simply jerking it out and putting a new one in the vehicle. There is a simple test that needs to be performed before removing the old pump.

That test is called a voltage drop test. Basically, with this test, the tech is testing the power side circuit and the ground side circuit to make sure everything is good. A problem with either of these circuits can cause the same symptoms a bad pump will display.

If there is a problem in one of the circuits, even though the pump probably will need to be replaced, if the tech simply puts a new pump in the vehicle, it will not be long until that pump is faulty too.

We have seen, and heard, of many such cases where a pump is replaced, only to have the new pump go out within a few weeks or months of installation. The issue, almost every time, was not a faulty new pump (though that does happen); but rather, an issue with the circuitry.

The issue is that resistance builds up in one side of the circuit or other and causes the pump to fail prematurely. If the test is done and too much resistance is found, the tech then has to divide the circuit and start searching for the problem area.

We won't give a detailed definition here of voltage drop, it can be found easily online.

Before removing the original pump, here is the test we performed to check the voltage drop. The picture below is where we back probed at the fuel pump electrical connector.

We took a DVOM meter and connected one end to the battery and  the other to the fuel pump electrical connector. We tested the power side first and then the ground side.

What we were looking for is a reading under about .5v or 500mv. As long as the reading is below half a volt, the circuit is good and the pump can be replaced with no worries. (The test must be performed with the pump running.)

First we test the power side. In the picture below you will see the reading is well below 500mv.

This test told us our power feed was good. We then moved to the ground side. As you see in the picture below, it too was below the 500mv threshold.

This test took 5-10 minutes. In that short time, we tested and found the power feed and the ground circuits to be in good electrical shape.

We were then confident to R&R the original pump, retest the pressure, take it for a test drive and ship it to the customer.

Taking the time to do the little things right, will help to prevent costly and unwanted returns. It makes the tech's life simpler and happier and the customer's too!

2000 Chevy Astro, 4.3, Crank/No Start Condition (Fuel Pump Voltage Drop Testing)

This 2000 Chevrolet Astro 4.3 came into the shop with a customer concern of a crank but no start condition. The customer had been driving the vehicle, went to start it again and it would not start.

I determined the fuel pump was the culprit. It was an intermittent issue, the fuel pump would prime and function some of the time, but mostly it would not. I knew we would need to replace the fuel pump, but I wanted to do some checks before we did.

It is important to check the circuitry of the fuel pump to be sure everything is okay there. It is a common mistake for a person or shop to simply replace the pump without checking the circuits. Often times, this causes comebacks or a recurrence of the same issue not long down the road.

One of the tests I do, is to perform a voltage drop test on the power (feed) circuit and on the ground circuit. A voltage drop is an invaluable test to confirm the integrity of the circuitry.

To perform a voltage drop test on a fuel pump, you need jumper wires and a DVOM. You also need the circuit "loaded" to do a proper test. In this case, that means you want the fuel pump engaged or priming, and checking the values when the component is under a load.

First, I checked the power side of the circuit. I backprobed in at the power wire for the pump (you will need a wiring schematic to determine which wire this is) and attached my probe to the positive terminal at the battery. With key off and the fuel pump not primed we see battery voltage on this circuit. (The voltage was a bit low, due to cranking to get the intermittent problem to occur.)

When we turned the key on and the fuel pump primed we saw what our voltage drop was.

(Sorry for the glare.) The reading on the meter is .534 volts, or basically, half a volt. The general rule for a voltage drop test of this kind is you do not want it to exceed about .5v (1/2 a volt). Here with our reading, everything is okay. We don't want to be too dogmatic about the .5v rule. As long as it is relatively close, it is fine. Typically a bad reading will be overly excessive, say 3-4 volts.

Next, I wanted to do the same test, but this time on the ground circuit. I backprobed the ground wire for the fuel pump and but my lead to the negative battery terminal. With key off, pump not primed we are reading 0v., which is what we expect for a ground.

We then turned the key on, with the fuel pump primed and read the voltage drop reading.

(Again, sorry for the glare.) You can see we are well under our .5v limit. We are below .1v, so no problem on this circuit either.

At this point, there was nothing left to do but to replace the pump. We inspected the gas tank for any contaminants and made sure the gas and tank were nice and clean. After replacement, we tested it and drove the vehicle. Everything was now fine.

We also changed the fuel filter, which was extremely clogged. It  probably contributed to the failing of the pump. We also put some good injector cleaner through as well.

1994 Ford Taurus Wagon 3.8, Cooling Fan and Fuel Pump Run Continuously With Key in On Position

It's been a while since I've been able to post.  I have several to post and I'll start with this one.  This is a 1994 Ford Taurus Wagon with the 3.8.  The customer brought it in after working on it himself since before Christmas.  Basically, after many repairs and swapping parts, the original concern was still present.  He gave me a list of all the parts he had replaced:  fuel pump, fuel pressure regulator, ECM, CCRM, Coolant Temp, Ignition Switch . . . etc.  He had tested the grounds several times and had found no problem.

The symptoms were when he would turn the key on, the cooling fan would come on immediately and the fuel pump would run continuously and not cycle after two seconds, and he could not get his code reader to communicate with the ECM.  From that description, I felt something was going on with the grounds: either with the ECM or its contacts or a ground on the vehicle.

I knew from other repairs a common problem on these was the ground connector located between the battery and the radiator.  Often times, when you unplug this connector (a two wire connector) it will be corroded.  When I unplugged this one I saw no corrosion.

However, to be sure I jumpered the two wires (harness side) straight to the negative battery post.  When I turned the key to the on position, the cooling fan did not engage and the fuel pump primed for two seconds and shut off.  Also, my diagnostic machine would now communicate.  Presto!

I tested it some twenty times or so and each time it worked as designed.

I spliced two good wires together and drilled a new ground contact on the body and put loom around it.

 I again tested it several times and each time it worked properly.

Here is a look at the orignal connector.

Apparently, the wiring on the ground side (in the picture the longer wires) was faulty and not allowing the ECM to have a good ground; hence, the running of the cooling fan and the fuel pump without cycling off after its prime.

Again, sometimes it's the small things that make a big difference.  After telling the customer of the repair, he told me he had probably unplugged and plugged in that connector numerous times.  It did look okay, but upon testing it was faulty.