Hard starting new Walbro 255 early model 2 to 1 pump conversion

[603Racing]

I wouldn't say grasping at straws. If the noid light tests pass then you know you're getting good signal from your CPS, the DME is working, and the injectors/injector harness are working.

You stated 45psi. Is that when cranking while the car is doing it's no start thing?

[rcfanatic]

Smoke test was done at the entrance to the elbow.

Noid light would be easy but might be grasping at straws. It will start first try when warm, maybe 50% of the time. What bugs me is that no one else seems to have this problem with their aftermarket (non-check) pumps.

I have a 173 DME and late model cluster, but the CEL does not illuminate because the harness is early model.

[603Racing]

Just to verify, how did they perform the test? Some shops will connect directly to the throttle body. If this is the case, you could still have a leak at the elbow/boot.

Just for fun you could try a different DME if you can find a spare one. It's something very easy to try just to rule that out. I doubt it's the DME though.

Another easy/cheap test would be to do a noid light test on the injectors. This will verify that they are pulsing while you do your "first start" cranking when your car will not start.

Curiously... Does your CEL work? Do you have any codes?

[rcfanatic]

I had a smoke test performed this morning and it came back negative. Fuel pressure is 45 psi at idle.

Any other ideas about what would cause hard starting when warm?

[dnguyen1963]

I think that after the first crank, the pistons creates enough negative pressure (vacuum) to pull the elbow in tight; thus eliminating the vacuum leak. Second crank starts normal because there was no more leak. As the car runs, there was enough vacuum to maintain a tight sealed engine. The cycles starts again after the car was stopped. This problem is more pronounced when the engine is hot.

[rcfanatic]

OK, I'll do a smoke test on Monday. Can anyone explain though the mechanism of why the leak causes it not crank *only* the first time? If it's too lean the first attempt, why does the second crank correct the mixture?

[dnguyen1963]

I had the same problem a long time ago. 1st crank...nothing, turned key off/on...2nd crank...started right up. I traced the problem to a loose elbow going into the throttle body by smoke test. Carb spray did not reveal the problem. Actually, this elbow did not always have a leak because it could move slightly in and out. Whenever it was in, there was no vacuum leak. JB weld the sucker and you will not have this problem again.

[rcfanatic]

I'd refer back to Jim's original post. You may have a vacuum leak. Smoke test.

You could also try a different DME if you have one on hand.

Possibly an electrical issue, as heat can affect resistance, but this doesn't sound like a typical electrical issue.

How could it possibly be a vacuum issue? I didn't have this problem prior to the new pump, and the fact that it starts immediately on second attempt doesn't point toward vacuum leaks.

If anything I'm starting to think it could be a bad fuel pump relay that does not pushing enough current when hot, and/or a general lack of current to the pump while the starter is turning and acting as a sink

[603Racing]

I'd refer back to Jim's original post. You may have a vacuum leak. Smoke test.

You could also try a different DME if you have one on hand.

Possibly an electrical issue, as heat can affect resistance, but this doesn't sound like a typical electrical issue.

[rcfanatic]

I've come to the conclusion that the hard starting only happens when the engine is warm. When cold it only has to turn over for a couple of seconds in order to start. Warm it will crank all day and do nothing on first attempt, then key on/start immediately after and it fires right up. Does the DME stop supplying fuel at some point during cranking to prevent a flood? What would cause this behavior?

[rcfanatic]

It should be the same pressure as your regulator.

It seems to me that the vacuum during cranking is considerably less than at idle

[603Racing]

It should be the same pressure as your regulator. And Jim is right, the pump should only have power while cranking or running.

If everything is working correctly you really shouldn't need a check valve. It may take just a bit longer to start, but only a few rotations.

[rcfanatic]

To know what is going on and thus what needs to be fixed you need to tee a gauge into the rail supply line and connect a test light across the pump power connections.

What should the pressure read while cranking?

[jlevie]

Update. Twice now I've let the pump run four about 4 seconds with the ignition in the on position and the engine does not hesitate to start afterward. I'm guessing I need a check valve

The fuel pump only runs when the DME sees timing data from a rotating engine. If the pump runs when just the ignition is on the car isn't in the stock configuration. To know what is going on and thus what needs to be fixed you need to tee a gauge into the rail supply line and connect a test light across the pump power connections.

[rcfanatic]

Does anyone have a good source for a check valve? The one used on the E32's is nla.