Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[al brown]

Hi everyone, sorry but I've not been on here much lately, the estate is say on the drive waiting for sunny weather so I can weld the sills up for the mot.
Anyway, the wife's corsa c 1.4 twinport has suddenly decided to be a bugger to start when is cold. It tries to fire but doesn't catch fire the first 8 or so tries, when it does start it takes a few seconds to fire on all counts bit then runs spot on and can be stopped and restarted fine until it's been sat for an hour or so. There is pressure in the fuel rail, coolant is showing as 8*, coil has been temporarily swapped with a known good one, plugs are new, compression in all cylinders is good and no fault codes are showing.
Anyone have any ideas? It's frustrating as its not all the time so it's hard to track down and I don't want to go buying parts just cause.

Cheers, Al

[X30XE]

Got a scope? Check CPS signal.

[al brown]

No, don't have a scope, computer reads rpm when I crank it, it's the fact that after it finally fires up it can then be stopped and started fine, or is this how the crank sensor on these fails, I'm used to the mig where it fails when warned up.

[zirk]

Also worth checking Cam Sensor but that normally through a fault Code.

[al brown]

I'll see if I can find a scope to check both the crank and can sensor.
Why would either of these sometimes cause poor cold starting, but then as soon as its started the first time it will start first go with the next hour or so?

[X30XE]

I'll see if I can find a scope to check both the crank and can sensor.
Why would either of these sometimes cause poor cold starting, but then as soon as its started the first time it will start first go with the next hour or so?

Because it could be a high resistance connection that gets better when warm.  I'm not saying it definitely is this... just that without an accurate RPM signal you're goin nowhere and seemingly they don't set fault codes readily.  Had a renault trafic van that liked playing the same game when the CPS was on the way out the door.

Check for an injection pulse with a DMM? If the ECU doesn't get a comprehesible CPS signal it most likely won't pulse the injectors. / can you smell fuel when it's cranking / does it splutter as if it might want to start?

[cam.in.head]

When was the fuel filter last changed ?

[al brown]

I'll see if I can find a scope to check both the crank and can sensor.
Why would either of these sometimes cause poor cold starting, but then as soon as its started the first time it will start first go with the next hour or so?

Because it could be a high resistance connection that gets better when warm.  I'm not saying it definitely is this... just that without an accurate RPM signal you're goin nowhere and seemingly they don't set fault codes readily.  Had a renault trafic van that liked playing the same game when the CPS was on the way out the door.

Check for an injection pulse with a DMM? If the ECU doesn't get a comprehesible CPS signal it most likely won't pulse the injectors. / can you smell fuel when it's cranking / does it splutter as if it might want to start?

Not checked the injection pulse, but there is fuel when I cranked it with the plugs removed, but then again it doesn't always fail to start so maybe I checked it on a good start, it started fine this morning but then failed again 3 hours later. It coughs like it wants to start and after 4 or 5 goes it does start, then it feels like its adding cylinders one at a time for the next couple of seconds, then it idles and runs fine. If I turn it off as soon as it starts it will start straight away again fine every time unless I leave it for an hour or so.
I don't like swapatronics but maybe I should try a crank sensor.

[al brown]

When was the fuel filter last changed ?

Not sure, we've not had it long but once it starts it runs fine.

[X30XE]

Check resistance from engine block to batt neg. Might not be sparking properly? *

(and if you're going to check for spark by removing plugs and grounding them be careful not to electrocute yourself and remove injector conn. for that cylinder so you don't blow yourself sky high )

If you've got fuel coming through you've most likely got some form of CPS signal.



Edited to add : * although one would assume this would record a fault code 

[X30XE]

I'll see if I can find a scope to check both the crank and can sensor.
Why would either of these sometimes cause poor cold starting, but then as soon as its started the first time it will start first go with the next hour or so?

Because it could be a high resistance connection that gets better when warm.  I'm not saying it definitely is this... just that without an accurate RPM signal you're goin nowhere and seemingly they don't set fault codes readily.  Had a renault trafic van that liked playing the same game when the CPS was on the way out the door.

Check for an injection pulse with a DMM? If the ECU doesn't get a comprehesible CPS signal it most likely won't pulse the injectors. / can you smell fuel when it's cranking / does it splutter as if it might want to start?

Not checked the injection pulse, but there is fuel when I cranked it with the plugs removed, but then again it doesn't always fail to start so maybe I checked it on a good start, it started fine this morning but then failed again 3 hours later. It coughs like it wants to start and after 4 or 5 goes it does start, then it feels like its adding cylinders one at a time for the next couple of seconds, then it idles and runs fine. If I turn it off as soon as it starts it will start straight away again fine every time unless I leave it for an hour or so.
I don't like swapatronics but maybe I should try a crank sensor.

^ possibly as much to do with fact it's flooded with fuel by that point as anything else.

[Kevin Wood]

In addition to the above, I'd be checking the coolant temperature sensor.

[al brown]

In addition to the above, I'd be checking the coolant temperature sensor.

My code reader showed it at 8 degrees, that was my first thought. I know there is more than one sender, and I'm assuming the reader is showing the one the ecu sees.

[Kevin Wood]

In addition to the above, I'd be checking the coolant temperature sensor.

My code reader showed it at 8 degrees, that was my first thought. I know there is more than one sender, and I'm assuming the reader is showing the one the ecu sees.

Sorry. I'd missed that.

I have an oscilloscope (or two ) by the way, but if the ECU is reporting stable RPM while cranking I think it knows all it needs to. I wonder if the fuel pump is taking a while to prime. You say there's pressure at the rail, but maybe not enough?

[Andy H]

In addition to the above, I'd be checking the coolant temperature sensor.

My code reader showed it at 8 degrees, that was my first thought. I know there is more than one sender, and I'm assuming the reader is showing the one the ecu sees.

Does your code reader show you the air inlet temperature?