I have a 2001 Ford F350 with a Triton V10 engine. The engine starts and performs perfectly all the time except for cold weather cold starting. After 1-2 failed attempts to cold start in cold weather, the engine needs to be started with the accelerator pressed to the floor. Once started, the engine recovers and fast idles as it normally does. What is the problem? Why does it flood itself? Any advice would be greatly appreciated, thanks.
When an engine is started cold, there are four primary inputs into the computer.
These are the crankshaft position sensor, throttle position sensor, MAP/MAF sensor, and the coolant temperature sensor.
I’d be looking at the coolant temp sensor.
Tester
I’m missing something in that train of thought, can you elaborate? Enrichment is determined by temp sensor. This one is flooding cold. If the temp sensor fails in a way that indicates cold engine, no harm as it is actually cold so computer provides normal cold enrichment. If it fails in a way that falsely indicates it is warm, the computer will not provide enrichment. Aren’t problems with the temp sensor usually manifested in warm restart/operation? What am I missing?
Could measure the CTS resistance, check fuel rail pressure under cold cranking, consider leaky injectors- maybe pull plugs after cranking to see how many are wet and how wet to identify if there are one or two problem childs versus all of them soaked…
Hello all. I should add that I am not a mechanic, just a guy trying to determine if the problem is something I can tackle or if I need a pro. A faulty coolant temperature sensor seems to make sense to me. What if the sensor is failed in the warm engine mode. Then, when I try to cold start the engine in cold weather it tries to start but quits because the enrichment is not being provided (too lean). BUT, after a couple of failed attempts to start the engine there is now too much fuel in the cylinders hence the need to start the engine as if it is flooded. Does this seem like valid thinking?
It would start once the required amount of fuel was in the cylinder but before it flooded. Just extended cranking perhaps. I’ve had this failure mode on Chevy trucks and the result was fine cold starting but overly rich operation when warm. But I’m certainly open to other possibilities so that’s why I asked. It’s certainly easy enough to measure the sensor to see if it is failed in some way…
If the coolant temp sensor is informing the computer that the coolant temperature is at -40 degrees, the computer will command that more fuel be injected into the engine than required, flooding the engine.
Tester
What you are saying makes sense. If the mixture is too lean why wouldn’t the engine just start after a few attempts as soon as it reached the appropriate air/fuel level rather than going to a flooded-like condition.
That makes sense and would explain why the engine floods so quickly (after 1 or 2 attempts to start). From my very brief internet research it seems that coolant temperature sensors are relatively inexpensive and fairly easy to change. This is something I can probably fix so I think I will look into this a bit more. Thanks.
I think I see the difference. Both Chevy and Ford use separate sensors for the ECM/PCM vs temp gauge. The Chevy’s I’ve had used a temperature switch for the computer input and a resistance sensor for the gauge operation. So there, it was either cold or not. If your truck has a resistance sensor and uses an enrichment curve, it could be significantly off if it was railed to one side. However, you’d think it would set a diagnostic code if it was that far off as the rest of the system warmed up as expected but the temp sensor was still offset significantly to one side. Is the check engine light on?
No, the check engine light is not on. There are no other noticeable indicators that anything is wrong. As mentioned originally, once the engine is finally started it recovers quite quickly and runs perfectly. The problem became evident as soon as the weather got cold (currently –11C). The truck ran beautifully all summer. I still get the same mileage now at 169K as I did when it was new.
Tester, if you unplug the sensor so there’s no input from it to the pcm, what happens? Does the computer assume a default temperature? Curious what that temperature would be and if unplugging the sensor and trying to start it would offer any clues.
Could be the CTS, but that part isn’t a very common failure item is the problem. It’s easy enough to test, if you can find the resistance spec vs temperature, and you have an ohm meter. I’m thinking however you may have something more simple going on. A faulty fuel pump check valve. Cold starts are “cold” b/c the engine hasn’t been running in some time. And that time allows the fuel (which should stay in the fuel rail) to instead drain back into the tank if the check valve isn’t working. My theory says that’s the reason it starts the third attempt, as by then the fuel pressure has built back up again from the prior cranking. A fuel pressure test is how to prove or disprove this idea.
Another clue would be to remove a spark plug when this happens. My theory says the problem isn’t flooding. If the plug isn’t wet, that would be consistent w/the check valve.