Hard start, fuel pump?

dz187

Member
Sep 22, 2005
65
0
6
Arizona
Got in my car the other day and instead of starting in a few cranks, it took probably 20 seconds of cranking total to get it running. Up until now the car had always started up right away. The last thing I did before parking it was change my oil and fill up my gas tank a few days ago. The last tank of gas was probably 3 months old, but it ran perfectly fine on it, and started right up when I left the gas station.

So I went and rented a fuel pressure gauge from Autozone and hooked it up. When I prime the pump I only get about 5 psi per prime, and the car finally starts when I get around 40 psi. At idle I get 32-36 psi (low 40's when the regulator's vacuum line is disconnected) and 19 inches of vacuum. Fuel pressure drops to 30 at 2000 rpm and does the same anytime I stab the gas a bit. I shut the car off about an hour ago and the gauge still reads 36 psi.

I'm going to buy a new fuel filter since I've been wondering if filling the gas tank stirred up some crap from the bottom and clogged the filter. If the filter doesn't fix it, fuel pump?

... or something else?
 
  • Sponsors (?)


never had this in my mustang, but did in my jeep...fuel pump started to lose prime, and was going bad. it still had good pressure if i ran it, then turned it off for an hour or so, but after it sat for a few hours or over night is when it would lose the prime. so maybe you can check it like that again after it sitting over night?
 
Fuel Pump Troubleshooting for 87-90 Mustangs

Clue – listen for the fuel pump to prime when you first turn the ignition switch on.
It should run for 2-5 seconds and shut off. This on and off again cycle helps to prevent
flooding the engine when cranking. To trick the fuel pump into running, find the ECC
test connector and jump the connector in the upper RH corner to ground.
B.jpg



Underhoodpictures007-01.jpg


Underhoodpictures010.jpg


If the fuse links are OK, you will have power to the pump. Check fuel pressure –
remove the cap from the Schrader valve behind the alternator and depress the
core. Fuel should squirt out, catch it in a rag. A tire pressure gauge can also be
used if you have one - look for 37-40 PSI. Beware of fire hazard when you do this.

No fuel pressure, possible failed items in order of their probability:
A.) Tripped inertia switch – press reset button on the inertia switch. The hatch
cars hide it under the plastic trim covering the driver's side taillight. Use the
voltmeter or test light to make sure you have power to both sides of the switch

B.) Fuel pump power relay – located under the driver’s seat in most stangs built
before 92.
C.) Clogged fuel filter
D.) Failed fuel pump
E.) Blown fuse link in wiring harness.
F.) Fuel pressure regulator failed. Remove vacuum line from regulator and inspect
for fuel escaping while pump is running.

The electrical circuit for the fuel pump has two paths, a control path and a power
path.

The control path consists of the inertia switch, the computer, and the fuel pump
relay coil. It turns the fuel pump relay on or off under computer control. The
switched power (red wire) from the ECC relay goes to the inertia switch
(red/black wire) then from the inertia switch to the relay coil and then from the
relay coil to the computer (tan/ Lt green wire). The computer provides the ground
path to complete the circuit. This ground causes the relay coil to energize and
close the contacts for the power path. Keep in mind that you can have voltage
to all the right places, but the computer must provide a ground. If there is no
ground, the relay will not close the power contacts.

The power path picks up from a fuse link near the starter relay. Fuse links are like
fuses, except they are pieces of wire and are made right into the wiring harness.
The feed wire from the fuse link (orange/ light blue wire) goes to the fuel pump
relay contacts. When the contacts close because the relay energizes, the power
flows through the contacts to the fuel pump (light pink/black wire). Notice that
pin 19 on the computer is the monitor to make sure the pump has power.
The fuel pump has a black wire that supplies the ground to complete the circuit.

Remember that the computer does not source any power to actuators, relays
or injectors, but provides the ground necessary to complete the circuit. That
means one side of the circuit will always be hot, and the other side will go to
ground or below 1 volt as the computer switches on that circuit.


Diagram courtesy of AutoZone

attachment.php


Now that you have the theory of how it works, it’s time to go digging.

Look for 12 volts at the Orange/Lt. Blue wire (power source for fuel pump relay).
No voltage or low voltage, bad fuse link, bad wiring, bad ignition switch or ignition
switch wiring or connections. There is a mystery connector somewhere under the
driver’s side kick panel, between the fuel pump relay and the fuse link.

Turn on the key and jumper the fuel pump test connector to ground as previously
described. Look for 12 volts at the Light Pink/Black wire (relay controlled power
for the fuel pump). No voltage there means that the relay has failed, or there is a
broken wire in the relay control circuit.

Check the Red/black wire, it should have 12 volts. No 12 volts there, either the
inertia switch is open or has no power to it. Check both sides of the inertia
switch: there should be power on the Red wire and Red/Black wire. Power on the
Red wire and not on the Red/Black wire means the inertia switch is open.

Make sure that the power is off the circuit before making any resistance checks.
If the circuit is powered up, your resistance measurements will be inaccurate.


attachment.php


You will have to drop the tank to inspect the pump power and ground connector
and the pump wiring chassis ground.

Pump wiring: Anytime the ignition switch is in the Run position and the test
point is jumpered to ground, there should be at least 12 volts present on the
black/pink wire. You should see less than 1 Ohm between the black wire(s)
and ground. The chassis ground is up near the spare tire shell. To get some
idea of what a good reading is, short the two meter leads together and
observe the reading. It should only be slightly higher when you measure
the black wire to ground resistance.

The Tan/Lt Green wire provides a ground path for the relay power. With the test
connector jumpered to ground, there should be less than .75 volts. Use a test
lamp with one side connected to battery power and the other side to the
Tan/Lt Green wire. The test light should glow brightly. No glow and you have a
broken wire or bad connection between the test connector and the relay. To test
the wiring from the computer, remove the passenger side kick panel and
disconnect the computer connector. It has a 10 MM bolt that holds it in place.
With the test lamp connected to power, jumper pin 22 to ground and the test
lamp should glow. No glow and the wiring between the computer and the fuel
pump relay is bad.

Computer: If you got this far and everything else checked out good, the computer is suspect.
Remove the test jumper from the ECC test connector located under the hood.
Probe computer pin 22 with a safety pin and ground it to chassis. Make sure the computer
and everything else is connected. Turn the ignition switch to the Run position and observe
the fuel pressure. The pump should run at full pressure.
If it doesn't, the wiring between pin 22 on the computer and the fuel pump relay is bad.
If it does run at full pressure, the computer may have failed.

Keep in mind that the computer only runs the fuel pump for about 2-3 seconds when you turn
the key to the Run position. This can sometimes fool you into thinking the computer has died.
Connect one lead of the test light to power and the other lead to computer pin 22 with a safety pin.
With the ignition switch Off, jumper the computer into self test mode like you are going to dump
the codes. Turn the ignition switch to the Run position. The light will flicker when the computer
does the self test routine. A flickering light is a good computer. No flickering light is a bad computer.
Remove the test jumper from the ECC test connector located under the hood.

Fuel pump runs continuously: The fuel pump relay contacts are stuck together or the Tan/Lt Green wire
has shorted to ground. Remove the fuel pump relay from its socket. Then disconnect the computer and use
an ohmmeter to check out the resistance between the Tan/Lt Green wire and ground. You should see
more than 10 K Ohms (10,000 ohms) or an infinite open circuit. Be sure that the test connector isn’t
jumpered to ground.
If the wiring checks out good, then the computer is the likely culprit.

Prior to replacing the computer, check the computer power ground. The computer has its own
dedicated power ground that comes off the ground pigtail on the battery ground wire. Due to
it's proximity to the battery, it may become corroded by acid fumes from the battery. It is a
black cylinder about 2 1/2" long by 1" diameter with a black/lt green wire. You'll find it up
next to the starter solenoid where the wire goes into the wiring harness

If all of the checks have worked OK to this point, then the computer is bad. The
computers are very reliable and not prone to failure unless there has been
significant electrical trauma to the car. Things like lightning strikes and putting
the battery in backwards or connecting jumper cables backwards are about the
only thing that kills the computer.

See the following website for some help from Tmoss (diagram designer) &
Stang&2Birds (website host)

http://www.veryuseful.com/mustang/tech/engine/images/IgnitionSwitchWiring.gif

http://www.veryuseful.com/mustang/tech/engine/images/fuel-alt-links-ign-ac.gif

http://www.veryuseful.com/mustang/tech/engine/images/88-91eecPinout.gif
 
I checked codes with the engine off and running, nothing. Cylinder balance test twice, nothing. After so much testing and cranking, my battery was getting pretty worn down. I charged it overnight and when I primed the pump, my fuel pressure shot to about 35 psi! Maybe the weak prime was caused by a low battery (it still cranked fine, so I don't know), but the car still doesn't start after a second of cranking like it used to. When I'm cranking, the fuel pressure will hold steady at 40 psi. Maybe it's not the pump like I thought. Can a fuel pump put out good psi, but not flow properly?

I pulled out a spark plug immediately after trying to start the engine and it looked bone dry... Could a bad fuel pressure regulator cause this? I checked mine, and the vacuum line didn't have any gas in it. Also, sometimes the engine will sputter after I stop cranking, like it's trying to start, and usually the next five second cranking session will start it up.
 
From the Fuel Pump Troubleshooting for 87-90 Mustangs
"Keep in mind that the computer only runs the fuel pump for about 2-3 seconds when you turn
the key to the Run position. This can sometimes fool you into thinking the computer has died."

The computer will not turn the fuel pump back on until it times out or it sees more than 400-500 RPM coming from the IDM circuit that it uses to keep track of engine speed.

You might also be tricked into thinking there is something wrong with the fuel pump, the relay or some other problem. I recommend that you go back and carefully re-read the tech note to see if there is something you may have missed.
 
From the Fuel Pump Troubleshooting for 87-90 Mustangs
"Keep in mind that the computer only runs the fuel pump for about 2-3 seconds when you turn
the key to the Run position. This can sometimes fool you into thinking the computer has died."

The computer will not turn the fuel pump back on until it times out or it sees more than 400-500 RPM coming from the IDM circuit that it uses to keep track of engine speed.

You might also be tricked into thinking there is something wrong with the fuel pump, the relay or some other problem. I recommend that you go back and carefully re-read the tech note to see if there is something you may have missed.

I went through that checklist and everything looked ok. I ended up putting in a Walbro 255 and it DIDN'T fix my problem, but the original pump was on its last legs and the S shaped hose connected to it had a hole in it, so it was good to take care of that.

So in order for me to start the car I have to crank it and give it a rest about three times, usually when I stop cranking on the third attempt the car will sputter a bit like it's trying to start. The fourth crank typically gets it started. Also, the car seems to be getting worse, in addition to the hard start it's now giving me a sporadic rough idle. It will idle fine for a bit then randomly hiccup to 400rpm, then back to 650, 500rpm... etc. It stalled on me once when I was just staring at the tach in my garage. The idle seemed to get a bit better when it warmed up. I installed a new ECT sensor since a code popped up for it, but that didn't change anything either. It was probably just because the engine wasn't up to operating temp when I ran the test, but it was only $15 for a replacement/peace of mind. The car runs strong when I romp on it, but :shrug:

At this point I'm out of ideas.
 
This may not help you or be your case but I found that with one of my cars when the timing was too far advanced it had a hard time starting when cold. It would start but it would take about 20 seconds of cranking before it would fire up.
 
I last set the timing to 14 degrees, it ran fine set there for a year and hope it hasn't changed! I tried starting it this morning... it backfired a couple times and a few cylinders fired when I was cranking, but I couldn't get it to start. I noticed that a small coolant leak developed from the timing cover bolt, right under the thermostat housing. Related? Oil on the dipstick still looks good.
 
I figured it out. It was the coil. It was putting out a spark, but apparently a weak one. Swapped in a stock coil and all is well again! Plus it runs better than ever after the new fuel pump and various sensors I cleaned/replaced. The moral of the story is that MSD is crap, that coil had less than 3000 miles on it when it died.