Technical Fuel problem after standing still

[gbrfarkas]

Hello!
I have a 2003 1.2 Stilo (actually 2, but there is a problem with only one).
After a few hours of standing still, it's like it would not get fuel. After sucking up the gasoline with the ignition up and down procedure many times, it starts with the accelerator pedal pressed a bit (then it smokes normally for half a minute).
Unscrewing the red cap on the intake manifold and pressing the fuel valve, nothing happens before these hard starts (and not even 5 minutes after stopping), while the fuel just squirts out on the problem-free car. But once it's started, it runs without any problems.
What can be the issue?
I don't think it is the pump broke, because I suppose, there were problems while driving.

 

[bugsymike]

Hello!
I have a 2003 1.2 Stilo (actually 2, but there is a problem with only one).
After a few hours of standing still, it's like it would not get fuel. After sucking up the gasoline with the ignition up and down procedure many times, it starts with the accelerator pedal pressed a bit (then it smokes normally for half a minute).
Unscrewing the red cap on the intake manifold and pressing the fuel valve, nothing happens before these hard starts (and not even 5 minutes after stopping), while the fuel just squirts out on the problem-free car. But once it's started, it runs without any problems.
What can be the issue?
I don't think it is the pump broke, because I suppose, there were problems while driving.

Can you hear the pump in the tank when you only turn the ignition on , but do not try to start it?
It could be a lazy relay at the fuse box, or a poor connection to the trigger for the relay.
On a daughters 05 Grande Punto 1.3 multijet (I know yours is petrol) it had a rotten wire at the ECU which is used to send a signal to the relay to trigger the fuel pump.

 

[gbrfarkas]

Can you hear the pump in the tank when you only turn the ignition on , but do not try to start it?
It could be a lazy relay at the fuse box, or a poor connection to the trigger for the relay.
On a daughters 05 Grande Punto 1.3 multijet (I know yours is petrol) it had a rotten wire at the ECU which is used to send a signal to the relay to trigger the fuel pump.

Yes, the pump is definitely audible. I even changed its relay to be on the safe side.

 

[bugsymike]

As you say from your initial testing when it doesn't start, there is no line pressure to the fuel rail.
If every time you go to start you can hear the pump in the tank working and there is no external leaks to loose fuel pressure to the rail and indeed if there were it would be obvious to you and wouldn't cure it's self at other times, is it possible the pump vanes are slipping on the spindle so it doesn't build pressure, or some debris in the tank is blocking it intermittently?
When you hear the pump working does it stop each time after a few seconds, meaning it has got up to correct pressure, or do you just hear the pump spinning?
Is it possible to safely put a voltmeter to the fuel pump to see if it still pumps when engine starter cranking over?

 

[gbrfarkas]

As you say from your initial testing when it doesn't start, there is no line pressure to the fuel rail.
If every time you go to start you can hear the pump in the tank working and there is no external leaks to loose fuel pressure to the rail and indeed if there were it would be obvious to you and wouldn't cure it's self at other times, is it possible the pump vanes are slipping on the spindle so it doesn't build pressure, or some debris in the tank is blocking it intermittently?
When you hear the pump working does it stop each time after a few seconds, meaning it has got up to correct pressure, or do you just hear the pump spinning?
Is it possible to safely put a voltmeter to the fuel pump to see if it still pumps when engine starter cranking over?

Yes, as I turn the ignition on, the pump starts and stops working after 2-3 seconds. (However, it is more silent than in the other car.)
I bought the car a week ago and today was the first time I filled up the tank fully. The pointer of the fuel gauge indicates 3/4 full even when I filled 50 liters in the tank and it was at the top of the red sign at the scale before doing so. Is there a relation between the two issues?

 

[portland_bill]

I think it is suffereing from the same problem as my Panda, in that the pressure is dropping off too quickly after switch off.
The system appears to be a single feed pipe from the pump, with the pressure regulated only by the pump. This should also incorporate a non-return valve. When turned off, the non-return valve should maintain pressure in the fuel rail at the injectors, as you've found with the other car. The valve in the pump is allowing this pressure to fall, so that the pump has more work to do at startup. When the ignition is first turned on, the pump runs for about 3 seconds, but this seems to be insufficient if the pressure in the system is too low. As you've found, making the pump run several times pressurises the system properly, and allows a start.

I doubt the valve in the pump is serviceable, so a cure would be a new pump. We both need to shop around for the right price, for a quality part. Here in the UK, an OE one is over £200, but aftermarket reasonable brands are around £60.
Or we can live with it.

 

[bugsymike]

As @portland_bill suspects with similar symptoms with his vehicle it certainly points in that direction.
Is it possible to safely fit a pressure gauge to the valve on the inlet fuel rail to test and confirm suspicions?

 

[gbrfarkas]

@bugsymike @portland_bill Thank you for your suggestions. I managed to find a pump from a breaker for buttons, maybe I’ll go for the change. In the meantime, I decided to swap the pumpa between the two cars for a couple of days and we’ll see what happens.

 

[portland_bill]

As @portland_bill suspects with similar symptoms with his vehicle it certainly points in that direction.
Is it possible to safely fit a pressure gauge to the valve on the inlet fuel rail to test and confirm suspicions?

That's one of those things I've been meaning to look at. I have a Gunson pressure gauge, bought many years ago, designed then for testing mechanical pumps, so it might not be good enough for the 3.5bar pressure the fuel injection cars should run at. Then it'll need an adaptor. Or I'll have to release the pressure, remove the fuel rail valve, fit the gauge, repressurise to test, watch the pressure either hold, or fall. If it falls, repressurise, then clamp the feed hose, when it should retain pressure. Unless an injector is leaking.
Or I could just be lazy, and do nothing until it fails to run one day. In the winter. But here I am, in front of a computer, instead of getting stuff done.

This looks tempting: https://www.ebay.co.uk/itm/29526918...qgPIWxOa8l4ooJiXm6iu_oWxu7wGC2FBoCHf8QAvD_BwE

 

[bugsymike]

Don't know if it is safe to do so, but for some fuel rail valves they look like a Schrader ? type tyre valve thread?
Though finding a good tight fitting one may be hard. I know years ago I was trying to repair a classic Mini Hydrolastic suspension pump that had the same style of fitting but hard to get a good screw on fitting, though they could go to approaching 600psi I seem to recall.
Safety being the key note.
The one on eBay may do the job, though in the past cheap petrol compression testers like that have been poor.

 

[portland_bill]

Don't know if it is safe to do so, but for some fuel rail valves they look like a Schrader ? type tyre valve thread?
Though finding a good tight fitting one may be hard. I know years ago I was trying to repair a classic Mini Hydrolastic suspension pump that had the same style of fitting but hard to get a good screw on fitting, though they could go to approaching 600psi I seem to recall.
Safety being the key note.
The one on eBay may do the job, though in the past cheap petrol compression testers like that have been poor.

There's a company, AB Tools, that do mail order, and are always at major car shows, including the classic at the NEC in November. They have the identical set, but for £28.95. Or I could go to Machine Mart, they have a set for £40, which appears to have a very similar gauge.

 

[bugsymike]

Same with all tools, until tried hard to tell
As an apprentice in 1969 I bought an American pair of Proto "molegrips" for 30 bob (£1.50), all the other mechanics pointed out I could have bought a genuine pair of Mole grips at the time for only 19/6, (under £1) however I still use them over fifty years later and they grip better than many modern ones.

 

[gbrfarkas]

Hi Guys,
I have to revive this topic.
The issue has been sorted out. I replaced the entire fuel pump (bought one for buttons from a breaker). The car starts fine under every circumstances.

But the fuel gauge is a nightmare.
It shows absolutely false values.
Yesterday, I performed a dashboard test and a fuel gauge test with MES and the needle goes all the way through the states between empty and full, but it gets stuck for half a second around 3/4 state and after then it goes up to full.
Since then, it seems shows almost 50%, while my trip distance is 470 kms, which is absolutely impossible When I turn the key to OFF it rests fine in its dead position.
Has anyone performed a Value reset in MES? How to do that? I don’t want to change other gauges, only the fuel.

 

[portland_bill]

You've now got the old one sat in the shed/garage. See if you can dismantle it without breaking anything. If you can, you can do the whole job again, and put the 'new' pump with your sender unit.

 

[bugsymike]

If you haven't thrown the old tank unit away I would be tempted to swap the fuel gauge part over.
You could test it by just swapping the wiring plug to your old unit and operate it by hand with ignition light on to simulate fuel levels and if works OK then change that part over if same design .
I may be wrong but I don't see resetting the vales in MES will be the answer. (I did write this last night but forgot to press reply)

 

[gbrfarkas]

Tested and the same thing happens to the gauge with both pump units. So the problem must be somewhere else.

 

[bugsymike]

Tested and the same thing happens to the gauge with both pump units. So the problem must be somewhere else.

Earthing?

 

[gbrfarkas]

Earthing?

Would the fuel gauge test in MES work if it’s an earthing issue?