Hard to start when cold

[RMD1183]

Hello,
I have a 1995 Lumina APV with 3.8. When the engine is cold (either sitting overnight or after sitting for 2 or more hours) the engine cranks and cranks (for about 10 - 12 seconds) before it starts. This happens winter or summer. If the engine is warm or has been running within 2 hours or less, it will fire right up.
This might be a clue for somebody too: when you start it and let it run for just a few minutes or so (doesn't have to reach operating temp), and turn it off, it will fire right up again.
I have been told that it might be that the fuel is draining back into the tank, but why? I have also been told that it might be the fuel pressure regulator.
Any ideas? Suggestions?
THANKS!!!!!

[tempfixit]

Originally Posted by RMD1183

Hello,
I have a 1995 Lumina APV with 3.8. When the engine is cold (either sitting overnight or after sitting for 2 or more hours) the engine cranks and cranks (for about 10 - 12 seconds) before it starts. This happens winter or summer. If the engine is warm or has been running within 2 hours or less, it will fire right up.
This might be a clue for somebody too: when you start it and let it run for just a few minutes or so (doesn't have to reach operating temp), and turn it off, it will fire right up again.
I have been told that it might be that the fuel is draining back into the tank, but why? I have also been told that it might be the fuel pressure regulator.
Any ideas? Suggestions?
THANKS!!!!!

SOunds like the check valve in the fuel pump is faulty letting the fuel pressure release over time.

Try turning the key on without starting for 5 seconds. then turn off for about 15 seconds then try starting, if still takes to lonng try reepeating procedure a couple times then start.

[RMD1183]

Originally Posted by tempfixit

SOunds like the check valve in the fuel pump is faulty letting the fuel pressure release over time.

Try turning the key on without starting for 5 seconds. then turn off for about 15 seconds then try starting, if still takes to lonng try reepeating procedure a couple times then start.

OK, Did that and nothing changed. It still cranked and cranked before it finally started. I am wondering this: when I turn the key to the "on" position, shouldn't the fuel pump kick on and pressurize the system? If I have the key in the on position long enough for that to happen, then shouldn't it start right up?
If the fuel is draining back into the tank and I try and start the car, I can understand having to wait for the pump to push the fuel up to the injectors. But if I turn the key to the "on" position for a bit before I start, shouldn't that also get the fuel there?

Still confused.
THANKS!!!

[tempfixit]

Originally Posted by RMD1183

OK, Did that and nothing changed. It still cranked and cranked before it finally started. I am wondering this: when I turn the key to the "on" position, shouldn't the fuel pump kick on and pressurize the system? If I have the key in the on position long enough for that to happen, then shouldn't it start right up?
If the fuel is draining back into the tank and I try and start the car, I can understand having to wait for the pump to push the fuel up to the injectors. But if I turn the key to the "on" position for a bit before I start, shouldn't that also get the fuel there?

Still confused.
THANKS!!!

Yes that is correct, try turning the ignition to run position a couple of times for about 5 seconds waiting about 30 seconds between atteempts, then try starting. Also check your fuel pressure regulator to make sure it is not leaking or the vacuum line smell like cas.

[LMP]

Originally Posted by RMD1183

... when I turn the key to the "on" position, shouldn't the fuel pump kick on and pressurize the system?

At first KEY ON the fuel pump kicks in for about 2 seconds and pressurises the system. YOU SHOULD HEAR IT. At same time injectors open and prime the cylinders.
NOtice the following:
- if you turn key OFF then ON again, it will not pump again: this is programmed to avoid flooding the cylinders. This is why others have said you must wait some time before attempting KEY on again.

IMPORTANT: Initial prime and start pump operation is done through a RELAY...but once the engine is started, a contact on the oil pressure sensor engages and keeps the pump running . NOw if the relay has failed , it is POSSIBLE that the pump kicks in only after sufficient starting time has built oil pressure up. WHen engine is warm, it cranks easier and builds oil pressure up faster. SO it is very important that you confirm that you hear the initial PUMP PRIMING at first KEY ON. If it cannot be heard, this would confirm this theory.
Otherwise we will proceed to other possible causes.
Come back with your findings.

Here is the diagram for pump operation

www.avigex.ca/xport/fuelcontrol_iac.jpg
www.avigex.ca/xport/relays.jpg

[Tech II]

LMP beat me to it.......sure sounds like a relay problem.......extended crank usually means the relay has failed and until oil pressure builds, then the oil pressure switch closes and sends power to the pump....try switching relays......if that doesn't work, put hand on relay as someone turns key to on(should feel relay click)...

[RMD1183]

Originally Posted by Tech II

LMP beat me to it.......sure sounds like a relay problem.......extended crank usually means the relay has failed and until oil pressure builds, then the oil pressure switch closes and sends power to the pump....try switching relays......if that doesn't work, put hand on relay as someone turns key to on(should feel relay click)...

Thanks to you and LMP both!! I replaced the relay and it seems to still crank too long. I will do as you suggest and see if I can feel a click. The timing of this problem seems to have coincided with some other work that I did (probably should have mentioned this earlier). There is significant rust on the LF inner wheel well. I removed the support, windshield wash fluid bottle, battery, and disconnected all of the wires that run along the inner wheel well so that I could wire brush and sand the rust away. I reconnected all of the wires and put everything back together - and that is when the starting problem began. Before then I would barely bump the key and it would start. Afterward, it cranks and cranks. I have checked and everything appears to be connected, so I don't think I missed anything, and I suppose it could be coincidence.

[RMD1183]

Originally Posted by RMD1183

Thanks to you and LMP both!! I replaced the relay and it seems to still crank too long. I will do as you suggest and see if I can feel a click. The timing of this problem seems to have coincided with some other work that I did (probably should have mentioned this earlier). There is significant rust on the LF inner wheel well. I removed the support, windshield wash fluid bottle, battery, and disconnected all of the wires that run along the inner wheel well so that I could wire brush and sand the rust away. I reconnected all of the wires and put everything back together - and that is when the starting problem began. Before then I would barely bump the key and it would start. Afterward, it cranks and cranks. I have checked and everything appears to be connected, so I don't think I missed anything, and I suppose it could be coincidence.

OK. I got home and had my wife turn the key to the "on" position while I held my fingers on the relay. Though I heard clicks (from elsewhere?), I definitely did not feel anything from the fuel pump relay. And the van still cranked and cranked before it finally started. I am assuming that there may be a break in the wiring for the relay. Does any of the wiring travel across the front and then along the left side - the side where I disconnected everything to take care of the rust? Any other possibilities?
Thanks once again!!!

[tempfixit]

Originally Posted by RMD1183

Thanks to you and LMP both!! I replaced the relay and it seems to still crank too long. I will do as you suggest and see if I can feel a click. The timing of this problem seems to have coincided with some other work that I did (probably should have mentioned this earlier). There is significant rust on the LF inner wheel well. I removed the support, windshield wash fluid bottle, battery, and disconnected all of the wires that run along the inner wheel well so that I could wire brush and sand the rust away. I reconnected all of the wires and put everything back together - and that is when the starting problem began. Before then I would barely bump the key and it would start. Afterward, it cranks and cranks. I have checked and everything appears to be connected, so I don't think I missed anything, and I suppose it could be coincidence.

You may want to get some De-Oxit electrcial cleaner from Radio Shack to remove the corrosion and protect the terminals, there could be some corrision that may have been missed in hard to clean areas. This stuff has done wonders for me when I have used it. Costs about $10.00 a container, which reminds me that I need to get some again.

[RMD1183]

Originally Posted by tempfixit

You may want to get some De-Oxit electrcial cleaner from Radio Shack to remove the corrosion and protect the terminals, there could be some corrision that may have been missed in hard to clean areas. This stuff has done wonders for me when I have used it. Costs about $10.00 a container, which reminds me that I need to get some again.

Thanks. It is worth a try! I have done some research. It appears that when the ignition is turned on, there should be power to the relay at the green wire with white tracer which comes from the ECM. I will use my multi-tester to see if there is power to this wire. If not, there may be a break in this wire or the ECM may be faulty. Sound like a plan???

[Tech II]

The ECM doesn't send power to the relay, it "grounds" the relay....

[RMD1183]

Originally Posted by Tech II

The ECM doesn't send power to the relay, it "grounds" the relay....

Now I am confused. According to the Haynes repair manual, I need to disconnect the relay and connect a test light to the terminal with the dark green/white wire. When the key is turned on, the light is supposed to come of for 2 or 3 seconds and cycle off. Gotta figure this one out...

[rkvons]

I know you're probably getting close to a root cause here, but another experiment you can do to validate your current course of action is to pull the relay out and have your wife act like she's going to start the vehicle when it is cold. When she turns the key to the "on" position, short the connection that the relay would be doing, to therefore, energize the fuel pump. See it the vehicle then starts better.

[LMP]

Hey another trick is this: near the intake manifold throttle, you can probably see a red wire ending witout any connection: (you see it on the diagramas FUEL PUMP PRIME ) apply 12V to this wire, it runs the pump directly,(when relay is not energised) and you should, again, hear it from inside the vehicle. www.avigex.ca/xport/MAFArea.jpg
As suggested above though, 12V direct to contact 30 runs the pump oin any condition.

and contrary to several signals from the ECM,(which usually provides a ground) the green-white stripe wire does receive +12V from the ECM to contact 85 (see diagram above) ......you should be able to measure that.THe other end of the relay(86) is permanently grounded.....check for that ground!

[RMD1183]

Here is the latest:
1) I was finally able to check the green wire with white tracer. When I turned the key to the on position - the volts cycled up to battery voltage and then cycled back down. I understand that this is normal. I then checked the ground by putting the (-) end of the voltmeter probe into the ground terminal, and the (+) probe into the terminal with the green/w wire. I figured that if the ground was bad, I wouldn't get any voltage. But the voltage again cycled up to battery voltage and then back down. So I am assuming that both wires and the pcm are in working order.
2) Next I had my son turn the key to the "on" position, and I did feel a click in the relay.
3) I connected battery voltage to the red wire (fuel pump relay) for about 3 seconds. I heard a sort of swooshing sound which I assumed to be fuel rushing through the fuel rails to the injectors, telling me that the fuel pump was energized.

So I was assuming that the van should start fairly quickly. But alas, it still cranked for 5 to 6 seconds before it started. Not trying to be stupid, but how long should I have applied battery voltage to the fuel pump prime?

Any further ideas or suggestion would be GREATLY appreciated.

Thanks!!
Roland