Another dead 1999 Astro

[Dave_R]

Hi, searched this forum, but did not find a fit.

Last week, it started to randomly stall, either on the freeway, or in traffic. Just as if you turned the key off.

It stalled either in 1 minute interval, or sometimes it would run as long as 1/2 hour.

In most cases I can put it in neutral, turn the key, starts, then back in drive and go. No codes ever come on.

Sometimes, it will NOT re-start immediately.

When it does not immediately start, I noticed that the 2-3 second fuel pump prime does not happen when the key is first turned. Sometime I have to wait a few minutes, or in one case, overnight. Once I turn the key ON, and hear the pump priming, it will start.

Now, it died fully and will not re-start. The fuel pump 2-3 second prime does not happen.

I checked the relay, and I pulled the cover off, and manually tripped the contacts and the pump runs. I put a meter on the coil. and I get 0 volts with the key off, and 0.015 Volts when I first turn the key.

So, it looks like the PCM is not comanding for the pump.

I jumped the pump "ON" so that it is running, and it just turns over, not a sputter. So, I think the PCM is also keeping the ignition from firing as well.

Like I said, no lights or codes.

I was thinking if a crank sensor (or any sensor), at least the 2-3 second prime should happen.

Disconnected and re-connected the battery, thinking that this should reset the PCM, but no-go.

Any ideas????

[Dave_R]

Forgot to add..

I checked for votage on both sides of all the fuses, and they check good.

[old_master]

Remove the fuel pump relay. This will expose the 4 terminals in the relay socket. Using a volt/ohm meter, probe each terminal to identify the circuits.

The black wire should have less than 5 ohms resistance to ground. This wire is the ground for the coil in the fuel pump relay.

The dark green with a white stripe wire should show battery voltage when the ignition is initially turned to the RUN position, (engine off) for 2 to 3 seconds. If the PCM does not receive a signal from the crankshaft position sensor during the 2 to 3 seconds, the PCM shuts off the fuel pump relay, and voltage drops to zero. Also, the dark green/white stripe wire should show battery voltage continuously when the ignition is in the START position. The dark green/white stripe wire comes from the PCM and energizes the coil in the fuel pump relay.

The orange wire should show battery voltage at all times. It comes from the ECM B 20amp fuse.

The gray wire supplies voltage to the fuel pump and goes directly to the fuel pump connector on the fuel pump module. It should show battery voltage both at the fuel pump and the fuel pump relay.

When the dark green/white stripe wire has battery voltage applied to it, the contacts in the relay close and connect the orange wire to the grey wire and the fuel pump runs.

The black wire in the fuel pump module connector should have less than 5 ohms resistance to ground. This is the ground for the fuel pump.

[Dave_R]

Thanks for responding old_master.
everything checks out like you said, except the dark green with white stripe wire. The voltage goes from 0.002VDC to 0.015VDC when turning the key initially to RUN. Tried several times, and even monitored while turning the key and cranking also. No real voltage on this wire.
But, if I jump this pin with a 12VDC source, the relay trips, and the pump will run.
Even while this is jumped and the pump is running, I tried cranking it over, but I have no ignition either, so it just cranked.
Still looking.......

[old_master]

You're correct, the PCM is not commanding the fuel pump relay. There are a few things to check to find out why. However, it sounds like a poor/loose connection in a battery feed to the PCM. The reason I suspect this is because you stated it won't start when you bypass the fuel pump relay which means the ignition & injectors might not be getting battery voltage either.

Ignition in RUN position:
Using a voltmeter or test light, check both terminals of fuses ECM B & ECM 1 in the underhood fuse panel and fuses 4, 7, & 18 in the instrument panel fuse panel. Do not remove the fuses to check them, touch the test probe to the terminals in the fuse itself, they are accessible with out removing the fuse. Post your results.

[Dave_R]

Well, it is really nice and sunny today. Thought I would tackle this again.

So, First thing I did was turn the key to run, and start pulling on the wire harness. A few seconds, I hear the pump kick on and off. Shaking more, I narrowed it down to one of the four connectors going to the PCM.

Disconnected the battery, and pulled all 4 connectors, and there it was!

Looks like the bottom (Clear) connector harness was chaffing against the air Conditioner line and wore through a Pink wire.

Picture1.jpg

Picture2.jpg

Picture3.jpg

Picture4.jpg

So I cut out about 6 inches and spliced in a piece of 16ga MTW. Soldered, Heat Shrink, wrapped the harness in electrical tape, re-connected, and away I went.

Seems to run really well now!

Thanks old_master!

[old_master]

Bingo, there's your poor connection. The pink wire is a battery feed to the PCM from ECM 1 fuse. The broken wire was essentially a blown fuse to the ECM. Without battery feed, the PCM can't energize several circuits, fuel pump relay and ignition system included. Without the detailed description that you posted, the search could have taken quite a while! Good job finding and repairing it. Thanks for posting back.

[Jim2005]

Old Master,

When you check the black wire for ground, do you do it with the ignition on or off?

Off - 1.6

On - 70

I have been experiencing the same problems with it quitting and seems like no ignition. I hear the fuel pump run when I turn the key on, but wont start. I am going to take the doghouse off next and check the fuel pressure.

I have also heard that it could be the connector to the injectors, I will check that too. Any help would be greatly appreciated!

[old_master]

The injector connector can be a problem, but its not very common on 1996 and newer CSFI systems, (with 6 injectors). 1995 and older CMFI systems only have one fuel injector and it was a fairly common issue on them.

When it stalls and refuses to start, have you tried a shot of carburetor cleaner in the air intake to see if it will start? If it does, that would be an indication of a lack of fuel. If it makes no difference, the problem may be with ignition. Check fuel pressure first, the results will tell us where to look next.

When you initially turn the ignition to the RUN position, (engine off) you should hear the pump run for ~2 seconds, and then shut off. The 2 second time frame is called "pump prime" and this is when you need to check the fuel pressure, (when the pump is running). Must be 60psi to 66psi. When the pump shuts off, pressure will drop slightly, but it must remain above 55psi for at least 10 minutes. If it fails either part of the test, there is a problem with the fuel delivery system and further testing is necessary to determine what the problem is. Post your results.

[Jim2005]

I havent tried shooting starting fluid into the intake yet. I did pull one of the spark plug wires off and jumped it near ground. When I turn it over I get spark.
Fuel pressure jumps to about 67 psi when I turn the key on, then backs down to 56 psi. After 10 minutes its still at 53 psi.

I live in Green Bay, WI. Of course, it has to be the coldest days of the year when this happens. I have it in a heated garage right now. This happend a couple of weeks ago and I dumped about 6 bottles of gas line antifreeze in the tank (Isopropyl type). It started the next morining.

Just this last summer I changed the coil, distributor cap and rotor, plug wires, spark plugs and a MAP sensor.

it has got 265000 miles on it but it normally runs great. Where do I go next? By the way, its a 1997 with the CSFI injector system.

[old_master]

What brand cap & rotor & spark plugs did you install?
Did you put silicone dielectric grease on the cap terminals, (inside & out)