*one* show-stopping problem after 2 hours

[joshier]

My car.. peugeot 306 XL, 1.4 1994 will stool/choke etc after around 2 hours of driving.

I've had a guy replace all standard stuff (plugs, leads, air filter etc) didn't help.

I've had a technician put it on the diagnostic machine, no problems show.

I've had the same technician tie up some "switch" which controls the fuel line which he said should fix problem.

After driving it after he "fixed it", it still misfired.
Every now and again it 'jerked' in the lower revs like it did before.

Just an hour ago it done the same thing and totally conked out.
I had to give the engine very high revs to get it into a safe place with the clutch stinking the place out.

I give it to a technician, any one and they can't see any problem because it only happens after a while of driving and I am very, very angry.

[CitroJim]

These problems are excruciatingly hard to diagnose

Something is failing when it gets hot. My no.1 suspect would be the ignition coils.

[joshier]

These problems are excruciatingly hard to diagnose

Something is failing when it gets hot. My no.1 suspect would be the ignition coils.

I know, because it only goes wrong *outside* the garage. I've been to two different ones and they can never seen any problems.

It's only after I'm about 4 miles away that problems start arising and it's hardly convenient for them to look at it at that distance.

The guy did drive it around our neighbourhood so that's good of him, however it wasn't enough so I'm actually thinking I'll either have to:
a) Drive it around the neighbourhood myself till it goes wrong and try to roll it in the garage or b) he carries on working on it until he finds the problem.

[AndersDK]

Never seen the problem while the car is standing at tick over, and simply slowly warm up ?

I have good experiences with patient idle testings, where the engine bay is slowly rising to the hot (non-speed-ventilated) working temperature.
Any bad sensor or coil pack usually then shows up as a sudden drop in idle spped or a sudden set-in of lumpy running.
It may take an hour though ...

[joshier]

Never seen the problem while the car is standing at tick over, and simply slowly warm up ?

I have good experiences with patient idle testings, where the engine bay is slowly rising to the hot (non-speed-ventilated) working temperature.
Any bad sensor or coil pack usually then shows up as a sudden drop in idle speed or a sudden set-in of lumpy running.
It may take an hour though ...

I am not sure if the guy done this, but it doesn't *seem* to do it with just idling. I will mention that over 1 night the problem goes away again.

I've considered the idea of driving it about tomorrow around where I live and then as soon as it starts doing it - driving it into the garage for the guy to put it on the diagnostic machine.

Is this a bad idea?

Should the technician be technically able to do this patient idle testing, and if he doesn't - does that make him a bad decision to give him the job?

How to Test an Ignition Coil
This testing procedure is valid for most automotive coils. Using an ohmmeter, check the resistance between the side terminals of the coil. Do this with all of the wires to the coil disconnected. You should see 0.75 to 0.81 ohm of resistance. Then check the resistance between either side terminal and the center high tension terminal. The reading should be 10,000 to 11,000 ohms. Any significant deviation from these numbers would indicate that the coil is defective.

I have an ohmeter at work, is this a hard procedure to do?

[CitroJim]

That test will not find a fault like you have. It's only valid for a totally duff one.

The problem that befalls coils is one of shorted turns. Often they're fine cold but as they warm up, insulation between the windings breaks down leading to internal shorts. These shorts will have negligible effect on the DC resistance values but will have a massive effect on coil performance due to large eddy-current losses. A shorted turn acts like a noose around the neck of the coil. It will also cause it to run hotter than normal and that might give you a clue. Is it roasting hot when the fault occurs?

Practically, the only way to prove a suspect coil is by substitution with a known good one.

[joshier]

That test will not find a fault like you have. It's only valid for a totally duff one.

The problem that befalls coils is one of shorted turns. Often they're fine cold but as they warm up, insulation between the windings breaks down leading to internal shorts. These shorts will have negligible effect on the DC resistance values but will have a massive effect on coil performance due to large eddy-current losses. A shorted turn acts like a noose around the neck of the coil. It will also cause it to run hotter than normal and that might give you a clue. Is it roasting hot when the fault occurs?

Practically, the only way to prove a suspect coil is by substitution with a known good one.

Oh right, that's interesting.

Yes it only does it after driving about a bit..

Tomorrow I've planned with the guy at the garage to test my car again - but only after I've got it running long enough so as to cause problems - hopefully getting it in there as it starts playing up.

[joshier]

I've just been told that we *also* changed the coil pack as well on my car.

[AndersDK]

I've just been told that we *also* changed the coil pack as well on my car.

Right -

The basic idea using methodical faultfinding procedures, is to ensure yourself that anything you already have been over is working.

Its no good using the method of substitution of "suspect" parts.
You end up with a hideous large bill on new parts, and still no joy.

The number one step is ALWAYS :
Positively trace down the problem to be EITHER a missing fuel problem OR a missing ignition spark problem.
These are the main 2 systems that together makes the engine run.

In the rare event that both systems are failing, you then know for sure its the main pump/injector relay at fault, or its wiring, or its voltage feed from the engine management controller.

[joshier]

I've just been told that we *also* changed the coil pack as well on my car.

Right -

The basic idea using methodical faultfinding procedures, is to ensure yourself that anything you already have been over is working.

Its no good using the method of substitution of "suspect" parts.
You end up with a hideous large bill on new parts, and still no joy.

The number one step is ALWAYS :
Positively trace down the problem to be EITHER a missing fuel problem OR a missing ignition spark problem.
These are the main 2 systems that together makes the engine run.

In the rare event that both systems are failing, you then know for sure its the main pump/injector relay at fault, or its wiring, or its voltage feed from the engine management controller.

Well, at first instance when we were changing the parts (plugs, air filter etc) they were bought brand new - .. however, you do mention a good point and that's why I'm now taking it to a different garage.

The guy said he would not charge for a diagnostic unless it finds a problem.

I'm also going to drive it a lot first so when I pop it in - it will have the problem.

[joshier]

Just put it through a peugeot approved garage on a diagnostic.

The guy said the following:

Probably throttle body sensors and could be cleaned up to make it work again for £75, £250 or so for a new one.

A whistling noise like an old kettle when driving first for 15 min they do not know the reason for, could be cam belt.

Water pump leak or something.