Xantia V6 broken exhaust (update: and rough low rpm running)

[Mandrake]

Thanks Jim, no urgent hurry, it's planning to rain the next two days anyway and if its fine on Sunday I'm thinking about swapping the fuel pressure regulator then...

[xantia_v6]

The fuel pump is pumping against maximum pressure when the engine is stopped.
Break into the return line to the tank and run it into a container to measure the flow rate per minute.

[Mandrake]

What is the minimum acceptable flow rate, is it stated anywhere ?

The pump won't run continuously with the engine stopped by the way. That would be a fire risk in the case of an accident that severed the fuel line.

[addo]

It uses 4/5 of SFA when at hot idle; you can tee off the return. From memory 15 seconds produced well into the hundreds of ml. To measure longer you can jump the tachymetric relay and just feed into a four litre can then do your simple maths.

[Mandrake]

I've found a delivery figure for the pump in the service data - 120 litres/hour at 3 bar. Or in more convenient figures for a 1 litre measuring jug, 1 litre in 30 seconds.

Even though the same service data describes the schraeder valve as a point for "delivery check", unless I'm mistaken while still useful this would not be a 100% valid test because the pump would not be providing its delivery into a 3 bar load, to test it under the full 3 bar pressure I have to break into the return path after the pressure regulator...

There's only really three places I can see to do that looking at the schematic, one is the "click-fit" connectors at the back of the timing belt cover, but I'm not sure how easy that would be to disconnect and feed to a measuring jug... another possibility is at the one way return valve which I'm assuming is somewhere under the car, or the third is to lift the return line off the top of the tank itself.

I'm wondering whether at the pump could be the easiest place ? Lift the seat up and the connector and both fuel lines are exposed, it looks like it has the same re-fittable squeeze to release connectors as the fuel filter, I should be able to disconnect the return line and divert it somehow to a measuring jug, plus I can just unplug the pump and directly apply 12 volts there so I'm not messing around with relays in the engine bay. Plan ?

Where is the tachymetric relay in case I decide to jumper the pump on in the engine bay - is it one of the relays in the ECU box or is it a standard relay in the fuse box ?

[addo]

On a four pot Xantia the tachymetric relay is in the ECU box.

You could - should you chose to go off the fuel filter outlet, which you ought be able to push a piece of 8mm fuel hose onto - just hotwire the pump with Scotchloks.

[Mandrake]

On a four pot Xantia the tachymetric relay is in the ECU box.

Ok thanks. On some cars although there is a relay that switches the positive to the fuel pump when the key is turned on, the negative is also switched by a transistor in the ECU to allow it to directly control the running of the pump so it can shut off the pump if the engine stalls. I only measured the voltage across the pump I didn't check which side was being switched so I'll check more carefully.

You could - should you chose to go off the fuel filter outlet, which you ought be able to push a piece of 8mm fuel hose onto - just hotwire the pump with Scotchloks.

It's not on the fuel filter that I'd be checking it, but the return line coming back from the pressure regulator. Otherwise the pump is not pumping into a 3 bar load. It would do for a quick test for a blocked inlet strainer I suppose, but it wouldn't be a conclusive test as the pump could have internal wear that causes a loss in delivery when pumping into a higher pressure load.

[addo]

The pumps are of a design such that wear will only affect output when either the rotor blades have digested the housing or the brushes are worn to nowt. A simple drainback of your fuel filter will reveal what sort of junk (if any) is being put out by the pump. Mostly it's fine silty dust that gets in with the filling process.

I have seen press-fit to hose barb joiners as could be used for testing your return, not common though. Don't forget to blank the vacuum reference at the reg.

[Mandrake]

Thanks. One other thing I did yesterday besides cleaning the pump electrical connector and blocking the line from the throttle body to the purge canister, was that I deliberately unplugged the power to the fuel pump to let the fuel pressure drop until the engine struggled and stalled. I did that two or three times.

Maybe not relevant but last time I bled the fuel rail at the schraeder valve through my pressure gauge (again, 2 or 3 times) the car ran really really well afterwards, for a few days in fact. Perhaps the fact that it ran a lot better yesterday was related to the fuel rail being bled down ?

I'm still struggling to see how bleeding the fuel rail down to nothing, especially with the car running could do anything that would make the car run better temporarily. Any ideas, or is it just a co-incidence ?

The only thing I could come up with is that perhaps theres a bit of grit stuck in the pressure regulator, normally the flow through the regulator would always be one way, but when bleeding the fuel rail the regulator would first shut and then back feed into the fuel rail out the schraeder valve.

Just thinking aloud, and very speculative I know...

[Hell Razor5543]

Nothing wrong with thinking out loud, sometimes it's the only way to get an intelligent conversation!

[Mandrake]

I'm sure this car is full of gremlins, when I measured the fuel pump voltage yesterday at the pump it was 12.75v while idling, the same again today, I pulled fuse 2 for the injection out put it back in and now the pump has 13.15v across it. I can only presume there was 0.4v more than usual voltage drop across the fuse contacts...

After that I decided to pull and clean (with contact cleaner) ALL the fuses and relays in the under bonnet fuse box, even the maxi-fuses, now the intermittent miss/stumble every one or two seconds while idling seems to be gone...

Performance has improved even further, a lot of the time the car is very zippy now, but it still intermittently bogs down badly under acceleration at low rpm taking several seconds to recover even though moments before it was going like a scalded cat in a higher gear.

My gut feeling is that when I check the fuel pump delivery in litres per minute it will be ok, and that I have a tiny split in the fuel pressure regulator diaphragm that is intermittently leaking fuel into the manifold causing random vacuum correlated changes in mixture from too lean to too rich or vica versa. Changes that it takes the ECU a while to adapt to, hence it cant keep the mixture within the correct range from minute to minute under changing throttle loads. The bogging down could simply be the mixture suddenly going over rich or too lean when opening the throttle and causing the manifold vacuum to change...

Even though its a pain in the neck to take the manifold out on the side of the road I'm strongly tempted to swap the fuel pressure regulator over on Sunday if the weather is as good as predicted...

Thinking more about a potential fuel leak through the vacuum diaphragm, would where the vacuum line connects to the manifold on the left hand side tend to feed some cylinders more than others, thus cause a mixture imbalance between cylinders when fuel is leaking through, causing some cylinders to run richer than others ? Because there is certainly some evidence of different mixtures on different cylinders on the plugs I took out last time, despite never being able to find any vacuum leaks on the inlet manifold gasket...cylinder 1 (front right near the battery) seemed to be running the leanest of all...

[Mandrake]

Well on Sunday I ripped the manifold out and replaced the fuel pressure regulator with the 2nd hand one from Stempy.

I tried sucking on the vacuum line to the old pressure regulator and noticed what seemed like the taste of fuel, so I though Aha, there is a leak in the diaphragm... so out it came, I didn't take the fuel rails out completely, I just unbolted the front one and lifted it enough to get the regulator out.

After fitting the new regulator I attached a hose and measuring jug to the return line off the regulator to test the delivery under pressure. The quoted delivery is 2 litres per minute, but I only measured 1.2 litres per minute. I would have thought that is still more than enough though, or is it ?

When I got everything back together I measured the fuel pressure and its now 2.9 bars (with vacuum hose off) instead of 2.7 bars, not a huge change, but I was hoping that there was a crack in the diaphragm in the old one...

I took it for a test drive and it was a bit underwhelming... it seemed better at first but after I'd taken it for a 10 minute drive there wasn't really any change in performance, if anything slightly worse.

I then didn't drive the car for a couple of days, when I drove to the pictures tonight the performance was VERY flat and gutless the first couple of minutes when cold - even half throttle in first gear it was sluggish as hell as if I was hardly pressing the throttle... on the way home later in the evening the performance was a lot better but it was still quite flat.

So basically I'm no better off with the new fuel pressure regulator even though the pressure is higher, and overall the running of the engine seems to be slightly worse. The problem is so much has to be disturbed to get the pressure regulator out - wiring, manifold gasket, injector seals etc...

I doubt that the pressure regulator is why its running worse, it will be something that was disturbed during the job, but what ?

I'm at a bit of a loss how to diagnose this any further as I've run around in circles so many times.... is it compressions ? (sticking valves ?) ignition ? fuel supply ? injectors ? At one time or another I've tested and ruled out all of them and yet its still not running right, nowhere near it! The lack of performance is so intermittent that troubleshooting it seems to be almost impossible as you can never tell whether it changed by itself or in response to something you did.

Performance seems to open up and become fairly normal at higher RPM above about 3000 rpm but is gutless and unresponsive below 3000. Sigh. The thing that frustrates me so much is that last December when I taped the hole in the spark plug wire and went to visit David the car went like an absolute bat out of hell, and continued to run flawlessly for two whole weeks, hot/cold, rain or shine it was perfect, and its never been the same since even though it has brand new spark plug leads and plugs...

Confused ? I sure am! What am I missing ?

[Mandrake]

By the way I noticed quite a bit of oil on the inlet manifold and gasket after removing the gasket, is this normal or something to be concerned about ? Patches of the gasket were wet with oil in the same places you can see on the manifold flange...could the engine be sucking in an unusual amount of oil vapor through the crank case ventilation system ?





Also here's the rear right hand spark plug well, which was always the leakiest one - the top gasket seemed to have stopped leaking since tightening the bolts, while the lower gasket is still a bit wet but doesn't seem to have leaked anything into the spark plug well since last time I had the manifold out, so theres definitely no problems with oil fouling the spark plugs:

[addo]

There's your problem - someone has stuffed rags into the intake runners. No wonder it runs like crap.

Re the fuel supply - provided your infeed pressure remained at or around 3 bar, I would not be too concerned for an idle test with the low return rate. A dyno would show you leaning out if this is the problem.

The consistent flaw in much of your experimenting, is you're testing at situations of negligible load, for problems that start to occur under load with an engine up to temperature.

[CitroJim]

It happens to us all when we reach a certain age Simon, we old 'uns have our good and off days...