wierd cold start problem

[745 turbogreasel]

rubber shrinks when cold, and leaks more.
air wont come in around the CS shaft with the engine running, because fuel is leaking out at 40+ PSI, and hogging all the space.
If one seal is old enough to leak, the rest are too.

As far as working on  it,I think of it as a cross between a carb, and an auto tranny.  If you've got both of those to go back together, you will be fine.

[TylerDurden]

If air can get into the line anywhere near the engine when the engine is stopped, the fuel will siphon back toward the tank. Even a small leak that does not drip will allow that to happen (the small return lines being a common cause).

I verified this process for the past few days on a car here: after sitting overnight, enough air gets in to make the engine stumble... for exactly the time it takes to purge the air.

[wadem]

So you would think that the car would misfire or run rough right away. Mine starts fine even in very cold conditions but starts to run rough then dies after about a minute or so. Then it takes forever to start it if you want to kill the battery or fry your starter.

Wade

[TylerDurden]

So you would think that the car would misfire or run rough right away. Mine starts fine even in very cold conditions but starts to run rough then dies after about a minute or so. Then it takes forever to start it if you want to kill the battery or fry your starter.

I have been considering that aspect - very interesting.

One scenario is a leak admitting air further upstream (by or before the filter), but it seems the supply line has no bubbles?

Another scenario is a plugged OUT bolt - preventing flow of fuel into the IP... internal pressure would be OK until fuel is consumed by the injectors.

Another scenario is sticking vanes in the IP lift-section. Like above, remaining fuel in the IP gets used, but not replenished.


These all can be examined with bottle feeding. Particular attention to flow from the OUT bolt is advised - something like 650ml per 90sec at idle.

[8v-of-fury]

Something like 650ml per 90sec at idle.

Are these numbers the total amount of fuel the pump will pull through in 1.5 minutes? Does it include a return line going in to the same fuel container?

I swear I have idled one of these 1.6's on a 1L bottle of fuel, using not much more than like 400ml per 25-30 minutes.

[ORCoaster]

Tyler,  The numbers are correct.  The pump cycles that volume through it in that amount of time and only uses a tiny bit of it.  The rest comes back through the tiny hoses off the injectors and gets sent back to the tank.  Remember that the pump wants to be cooled as well as deliver fuel.  So they set it up to move a much greater volume than needed to make the engine run.

If you put the two lines into a bottle over time the overall level would go down like you say.  But it would get cycled many many times.

Make any sense?