intermittent starting 2003 mach 1, why? good fuel pressure

Try this. Cycle the key on/pause/off several times without cranking. Now crank the motor. Does this improving cranking time? If so, suspect the problem is fuel pressure leak down as the source of the problem. Leak sources include (but not limited to), Fuel injectors, fuel pump check valve. Fuel pump flex hoses. FRPS.

IF you are positive that:

• anti-theft (PATS) has "proved out"
• That the cranking fuel pressure is good.
• That the FRPS intake vacuum reference line is connected and leak free and there's no raw gas in the line.
• That cracking the throttle improves cranking time.

THEN suspect that you have a:

• Bad or dirty IAC valve
• a vacuum leak between the MAF and throttle body.
• an inaccurate MAF. Try cleaning the MAF. Use only a product designed for the application. Be careful. It's easily damaged.
• Clogged or leaking fuel injectors.
• Inaccurate or missing temperature sensor.

1996+ Crank with no start check list

What are the values for the ECT and IAT? Are they reasonable based upon local conditions at the time? An inaccurate temperature sensor can make for a difficult start.

What is the MAF flow during cranking? Does disconnecting the MAF have any effect upon the problem?

What is the fuel pressure (FRPS) as reported by the PCM before cranking?

What are the RPM's as reported by the PCM during cranking? Do they change? Does the check engine light go out during cranking?

What have you done to rule out a fuel injector problem? Consider an injector cleaning and flow test service such as injectorRX.com. Note, it's not possible to "see" a leaking fuel injector unless it's pulled out. That's why knowing to fuel pressure before cranking can help spot a leak in the fuel system.

Does the IAC valve work? Is it clean? What is the IAC duty percentage during normal idle? Does cracking the throttle slightly open during cranking help?

What is the status of the CAM sensor? Is the electrical connector in good shape? Clean with no corrosion? A missing CAM sensor signal can "cause" long crank times.

What is the status of the base motor? Compression good? Cylinder leak down? Spark plugs in good shape? No moisture in the spark plug wells? COP boot insulators in good shape?

There's some serious "parts changing" going on there. Recommend testing more. Some of the parts changed are rather expensive and could have been ruled in/out with more tests.

Do you have an ODB2 scanner? If not, here's some information on an affordable Windows based unit. This could have paid for itself with the first part changed without a "fix".

ForScan ODB2 scanner w ELM327 USB

For example if it is known that the fuel system is maintaining fuel pressure between starts and the FRPS is accurate, THEN changing the FPDM, or the fuel pump itself couldn't possibly impact cranking time.

Bultaco51 said:

as it simply is something is not turning fuel pump on, not sure about all the other technical stuff, this car runs great

Click to expand...

I'm not sure that this problem is being thought all the way through. Let's say for the sake of argument that the fuel pump isn't being powered on as you are theorizing. IF this were true:

• This would have ZERO effect upon the motor's initial ability to initially start (run yes). There are tons of cars out there that do not turn the fuel pump on until the motor actually starts. That's why it's necessary for the fuel system to maintain pressure between starts. IF what you think is true is true then the motor would fire up and then die assuming that were the only thing wrong.
• I just don't buy that it's possible for a car with a fuel pump power problem that only happens during cranking with ZERO other symptoms.
• The PCM knows what the fuel pressure is from the FRPS. Is it possible that because the PCM knows what the fuel pressure is that the PCM is smart enough to know that it's not necessary to prime the fuel pump after multiple key cycles?

But all of the above questions could easily be answered with an advanced ODB2 scanner. There's a PID that says what the fuel pressure is. There's a PID that says what the fuel pump mode is.

Plus there are the other temperature sensors that are well known to have an effect on the PCM's fuel enrichment strategy that we have no idea IF they are accurate.

The other thing that I don't feel is being understood is just because a sensor is new does that mean that the PCM is actually "seeing" the value? How do know there's not a wiring fault to/from the sensor?

I'm a big believer in Occam's Razor theory. Occam's Razor would say that it's extremely unlikely that a PCM would work perfectly in all phases of driving except during cranking. But even if it's true there's going to be other "clues" inside the PCM as to what's going on. For example, missing RPM signal. Missing spark. Missing fuel injector pulse. All things that can be tested for without changing the PCM to find out.

Note, I feel that changing a PCM as a diagnostic aid is a bad idea. Often a new PCM will generate more problems (variables) than it solves thus trading one problem for another. Best to replace a PCM when everything else has been ruled out. Or if there's solid evidence that the PCM has lost it's mind (Fred has gone insane).


Since you don't mind changing parts my recommendation is to have the fuel injectors cleaned and flow tested. InjectorRX.com will give a report of each injector's performance before and after cleaning. They will also let you know if the injectors were leaking before cleaning.

@Bultaco51 If its not a daily driver your injectors clog much more easily. I had to change out my injectors after two years of putting new ones in just because I drove it one day a week or would skip every other week and didn't use injector cleaning regularly to keep them from gumming up. Cars were meant to be driven. Not driven once a week or every other week. Causes more harm than good on your injectors and seals.

Check the codes if it has any.

LOL! Easy fix! At least you didn't need to go repairing anything.