91 Bravada ECM to Fuel Pump Relay Problem

The ECM will not close the fuel pump relay. With the ignition switch on, the voltage of the wire from the ECM to the fuel pump relay measures 12 volts to ground. (When I turn the ignition on the voltage stays on continuously instead of going off after two seconds.) When I connect the relay load to ground, the voltage from the ECM goes to zero instead of closing the relay contacts. With the ignition off, the continuity-to-ground of the wire coming from the ECM measures about 10,000 ohms.

Any suggestions will be appreciated. If I need to measure the continuity of the wire itself from the ECM to the relay, how's the best way to get to the ECM connector?

Thanks for the reply. Just a further question if I may?:
My wiring schematic shows the dark green with white stripe wire going to ECM-B fuse and the black with white stripe wire going to engine ground.
Is this wire routing the same as your explanation?

I obtained an Oldsmobile manual for this model and year. It shows the black/white wire to engine ground and the green/white wire to ECM and then ECM hooked to ECM-B Fuse.

With the ignition on and no load on the wire, the green/white wire measures 12v to ground. When connected to the grounded relay load, the green wire loses the voltage and has no measurable amperage on the wire (amperage measured in series with the grounded relay load). With the ignition on, the black/white wire has no voltage to ground when measured by itself and when connected to the relay load.

With the ignition off, the greenwhite wire has no voltage and measures 11M ohms to ground and the black wire has no voltage and measures 1 ohm to ground.

This wiring seems backward and the ECM performance is unusual!. Any other observations that I could make to help solve the problem? Your expertise is appreciated.

I did discover that this vehicle has a OBD-1 if that makes any difference in the wiring and diagnosis of a problem.

What is the 8th character of the VIN, (Vehicle Identification Number).

"z"

So sorry, my apologies. I thought you had a CMFI system when in fact it's TBI. Sorry I jumped the gun and didn't ask first. I will delete the earlier, incorrect post to avoid any confussion.

Unplug the fuel pump relay and probe each terminal of the harness connector:

Orange wire should have battery voltage at all times.

Black with white stripe wire should show less than 5 ohms resistance to ground. This is the ground for the relay coil.

Dark green with white stripe should show battery voltage when the ignition is in the RUN and START positions. This is the trigger, (positive) feed for the relay coil.

The grey wire goes to the fuel pump.

When the fuel pump relay is energized, the orange wire and the grey wire are connected together and the pump should run.

There may or may not be a red wire at the fuel pump relay. This is "pump prime" and is used for checking fuel pump operation quickly. It usually dead ends shortly after exiting the relay. When applying battery voltage to it, the fuel pump should run continuously. There should also be a grey pump prime wire exiting the harness; it does the same thing as the red wire. The grey wire changes to tan with a white stripe, and then to a solid tan wire, in the harness before it reaches the fuel pump.

Thanks for 'hanging-in-there' in a quest to solve this problem. The wiring is as you just described with the possible exception of the prime connection.

The fuel pump relay question came up when I began checking for a solution to driveability problems. (After the vehicle is driven for about 20 minutes, it begins bucking , jerking, idle surging, stalling suddenly, hesitating while trying to accellerate and dying when turning a corner). I had a code 34 read-out and subsequently replaced the MAP sensor and TPS sensor to no avail. I then touched the fuel pump relay and could feel it cycling on and off. I replaced the relay with no change noted.

I then discovered this problem with the wiring: With the ignition on and no load on the wire, the green/white wire measures 12v to ground. When connected to the grounded relay load, the green wire loses the voltage and has no measurable amperage on the wire (amperage measured in series with the grounded relay load). With the ignition on, the black/white wire has no voltage to ground when measured by itself and when connected to the relay load.

With the ignition off, the green/white wire has no voltage and measures 11M ohms to ground and the black wire has no voltage and measures 1 ohm to ground.

I can jumper the orange wire to the gray wire and make the fuel pump run all the time but this does nothing for the driveability problem. Replacing the relay, I find the relay now won't pull in. The fuel pressure measures 9-13 psi which is in specs for this vehicle.

I'm wondering if the problems with the positive feed from the ECM for the relay coil is causing some kind of electrical disturbance in the system?

First of all, the ECM's in these things are almost bulletproof and rarely fail.

The pump prime wire at the relay is a hit or miss thing, some vehicles have it, some don't. It's more of a "convenience" thing for diagnostics.

Before replacing any more parts, check the ground connections that exit the engine harness. Especially the one at the thermostat housing. If that ground connection is poor, loose, or nonexistant, the ECM will loose its ground and subsequently any ECM input or output that requires a ground will also be lost. This could absolutely cause the fuel pump relay to be erratic along with Diagnostic Trouble Codes, (DTC's) being set falsely.

It sounds like your voltage readings are ok. The reason the green/white wire loses voltage when connected, is because it goes to ground through the relay coil...normal.

With the relay unplugged, check the harness connector: The white/black wire should show less than 5 ohms to ground at all times. Try wiggling the wire while checking resistance, it should always stay below 5 ohms. If you see a rise in resistance, even momentarily, look for a poor/loose ground. If you can't find where the wire grounds, cut the wire a few inches back from the harness connector and solder on a piece of wire. Connect it to a good clean body ground with a sheet metal screw.

Something else to check is the connections at the fuel injectors. Make sure they are clean and tight.

Jumping the orange wire to the grey wire is a good test for the grey and orange wires and the fuel pump, but does nothing for a poor ground to the ECM. You'll need to do some investigative work to find and check the various grounds. Wiggle them where they attach to ground, and make sure they are shiny clean and tight, and no green fuzz growing on them. If you need a wiring schematic, PM me and include your email address and I'll send it to you.

I checked the fuel pump grounds and ohm readings are good.
I took off the ground at water pump outlet, shined it up real good.
I replaced every wire butt connection I could find.
The o2 sensor had never been replaced, so I did that.
I checked all the other grounds in the engine compartment.
Still the same problems exists, after the engine gets good and warm after about 20 minutes. The code 34 is on more that 50% of the time.
When driving. the problem seems greater over 5mph. When at crusing speed, the problem seems less when I have a slightly greater pressure than normal on the accelerator. Could there be a vacuum problem that's not recognizable as a hissing sound? I appreciate your expertise!

DTC 34 indicates a MAP sensor circuit error (signal is low indicating high vacuum).

1. Check the vacuum connection and wiring connector at the MAP sensor:
With engine at idle, the vacuum line should have manifold vacuum.
2. Check voltages on the purple, light green and grey wires when the engine is at idle.
3. Check voltages on the purple, light green and grey wires with the key ON and the engine off.
Post your results.

The vacuum connection at the MAP sensor has good vacuum.
I can wiggle the wiring connector and its wires while the engine is running with no change in the engine performance.

Voltages engine at idle
Purple wire -- 0 volts
Light Green wire -- 0 volts
Grey wire -- 5.04 volts


Voltages --key on, engine off
Purple wire--0 volts
Light Green Wire -- 0 volts
Grey wire -- 5.04 volts

I measured the voltages on the wires at a location just outside the MAP
wire connector after making sure the vacuum line and wiring connector were firmly attached to the sensor. Engine was warm.

Next step: Ignition off:

Unplug the MAP sensor electrical connector.
Check resistance from the harness connector to ground on the light green wire.
Check resistance from the harness connector to ground on the purple wire.

Ignition in RUN position, engine off:
Check voltage on all 3 wires.
Reconnect MAP sensor.
Post your results.

Unplugged Map sensor electrical connector

Light green harness connector to ground - 1.4 ohms
purple harness connector to ground - 1.2 ohms
Ignition on-engine off
Grey wire to ground - 5.0 volts
Lt green to ground - 0 volts
Purple to ground - 0 volts.
Reconnected Map sensor

Remeasured voltage to ground -engine now cold
Ignition on engine off
Grey to ground - 5 volts
Lt green to ground - 200 millivolts
Purple to ground - 0 volts
Ignition on - engine on
Grey to ground - 5 volts
Lt green to ground - 190 millivolts
Purple to ground - 0 volts

The MAP sensor is supplied reference voltage of approximately 5VDC from the ECM on the grey wire. The purple wire is sensor ground from the ECM. The Manifold Absolute Pressure, or MAP sensor, measures the absolute pressure inside the intake manifold. While the engine is at idle, vacuum is high, in other words, a negative pressure. When the engine is at wide open throttle, (WOT) manifold vacuum is near zero, or, atmospheric pressure. The MAP sensor measures these values and sends the data to the ECM through the light green wire. Here's where the problem is. You measured voltage on the light green wire near the MAP sensor, (connected and ignition in RUN position) and showed zero volts. DTC 34 states low voltage/high vacuum. In fact, the DTC is correct, there is no voltage signal being received from the MAP sensor at the ECM.

With the sensor unplugged you measured 1.4 ohms to ground on the llight green wire. We need to find out if the wire is shorted to ground somewhere in the harness, or if it's shorted to the purple wire, which is also ground.

Next step:
Ignition off, negative battery cable disconnected, unplug MAP sensor, unplug ECM: In the ECM connector "C", pin #11 is the light green wire. Check resistance from the ECM harness connector to the MAP sensor light green wire terminal. Have an assistant wiggle the harness while checking resistance. Must remain less than 5 ohms.

Then check resistance from the purple wire to the light green wire at the MAP sensor harness connector. Must indicate open or "infinite" resistance.

Next, check resistance from ECM connector "C" pin #11, (light green wire) to ECM connector "D" pin #2, purple wire. Must indicate open or "infinite" resistance.

Then check resistance of purple wire to ground at ECM connector and at MAP connector.

Doing the steps you suggested I found the short to ground in the light green wire where the wire exited the harness in the engine compartment.
All resisitance is now within the specs you mentioned.

After putting it back together: with the engine off, ignition off, MAP sensor unplugged, the resistance to ground on the light green wire is 52K Ohms, light green to purple wire is 52K Ohms, and purple wire to ground is <1ohm. With the ignition on grey wire to ground is 5 volts.

With the engine running, on the light green wire, voltage to ground at idle is around 1100 millivolts and as I increase the revs the voltage on the same wire decreases to around 900 millivolts. When I disconnect the vacuum line to the MAP sensor, the voltage rises to about 4+ volts and the engine sputters. Still showing code 34.

Good piece of work finding and fixing the short! Now the ECM is receiving the signal. Sputtering when unplugging the vacuum, and voltage rise is a normal result however, voltage at idle on the light green wire is too low, and that's what's setting the DTC 34 at this point. The MAP sensor may have been damaged from the short to ground in the signal wire. Some of the auto parts stores have testers for the MAP sensor, check around. You might have to install a new MAP sensor though.