Bogging down / cutting out

[dturbo]

Hi folks,

having a real bad issue with my 1.6 VTS. Just started the other day. It began with hesitation when pulling out of junctions, felt like you'd turned the key off and on again quickly. Then it got worse. Car now will start fine (hot or cold) and idle fine for about 10 seconds, then start to bog down, splutter etc and eventually die after about 30 seconds.

I can then start it up fine and the cycle repeats. Out on the road, this translates to also happening when driving. Starts to splutter, then if you try to accelerate at all it just dies when you touch the accelerator. Switch off the ignition and start up again and you have maybe 30 seconds of normality before it happens again.

I swapped the MAP sensor for a known good one, no difference. I ran the car with lexia plugged in and watched the readings, all look normal even when it is dying except the O2 sensor rarely changes at all from 1V (it changes sometimes). This made me think dodgy O2 sensor.

So I unplugged the O2 sensor and this allows the car to be driven almost normally, although there will still be a little hesitation when driving every so often (can happen at any time or speed).

I wondered if it could be coil pack as even with the O2 disconnected, it will still splutter and die after idling for 10 seconds or so. I wondered if maybe with the O2 sensor attached it is reading high due to unburned fuel and the ECU then backs off the fuelling , but with it unplugged it can continue to run 'rich' but still bogs down after a while?

I have renewed the plugs. I also tried running with stepper unplugged, coolant sensor (ECU one) unplugged, throttle pot unplugged (which was renewed fairly recently anyway) - all these produce other effects but the fault still occurs as well.

I'm gonna look at the charcoal canister and its wiring - I had a feeling it could be involved (stuck open or a leak inline somewhere) but other than that I am stumped. It throws no fault codes so no hints there!

EDIT - Just to add:

1. with O2 connected and the car barely driveable, if you go full throttle it drives perfectly. I tested this just changing up gears early and keeping it full throttle, no splutters etc at all.

2. With O2 disconnected which makes it almost normal but still occasionally misses a beat and will still splutter and die at idle, I took it out for a fast run. Car drove continuously at high throttle and was absolutely flying, so I don't suspect a fuel delivery issue as such.

I will upload a video later.

Thanks for any help
Craig

[S6SAXO]

No idea man.. And no fault codes coming up you say?

Get a video up when you get back tonight dturbo..

Sent from my CLT-L09 using Tapatalk

[MartinObviously]

the O2 sensor is working fine, it's trying to correct an issue. the fuel trim available to it is limited so eventually you'll get a fuel trim fault but it's not the o2 sensor!

If it was the coil pack, they get worse with heat when they start to fail. can be sure it's probably not that which is at fault.

Check the vacuum hose that goes from the inlet to the fuel pressure regulator. It's supposed to relieve fuel rail pressure at low vacuums. if that's failed then it'll be too rich all the time EXCEPT full throttle.

[dturbo]

Thanks Martin, you've confirmed what I was thinking - that the O2 sensor is seeing the rich condition and trying to correct it. I wasn't aware that the regulator worked in that way, sounds like a very viable possibility.

I take it the FPR is the thing on the end of the fuel rail, looks a bit like a vacuum advance unit ?

Thanks for the help, fingers crossed this is the issue. I'll let you know
Cheers
Craig

[MartinObviously]

Originally Posted by dturbo

Thanks Martin, you've confirmed what I was thinking - that the O2 sensor is seeing the rich condition and trying to correct it. I wasn't aware that the regulator worked in that way, sounds like a very viable possibility.

I take it the FPR is the thing on the end of the fuel rail, looks a bit like a vacuum advance unit ?

Thanks for the help, fingers crossed this is the issue. I'll let you know
Cheers
Craig

Yeah it's on the end of the rail

I would just replace it, they're not expensive at all, the old one will probably need a good tug to remove it.

and a new bit of hose for it too

Start with the simplest things and all that.

No worries, hope it helps.

[dturbo]

OK so checked out the vacuum hose to the FPR. Theres vacuum on it alright and I sprayed WD40 all around the hose ends and none went in so not leaking, hose looks fine too not perished etc.

I do wonder if the FPR might have failed as there was no difference at all in idle (when the car was idling OK) between hose off the FPR (and covered over to avoid a vac leak) and hose on. I would have expected there to be some sort of difference?

The weird thing is it can idle for 5 minutes sounding quite happy then just start to splutter bog down and eventually stall. Start it up and away it goes again, perfect idle.

Same with driving. I reconnected the O2 sensor and went out in it (from cold). Ran absolutely fine then suddenly just wants to splutter and die on any throttle. Off and on again and away we go for another few minutes.

I also ran the car with the canister hose off and again blocked the spigot on the intake to avoid a vac leak. Still bogged and stalled in the same way.

So should I replace the FPR first you think?

Cheers
Craig

EDIT: seems to be an array of different presures available - 2 bar, 3 bar, 3.5bar ,4 bar (this one claims to eliminate a 'flat spot' at low revs caused by lower pressure to meet Euro requirements etc. bla bla). What should I get?

[dturbo]

Just pulled the trigger on a bosch FPR from Euro using the Citroen part number as a reference, so will be whatever the original part was!

[dturbo]

UPDATE: got the part in my hand, its 3 bar. Sounds about right. Will post back later once I've fitted it . Fingers crossed!

[dturbo]

New FPR fitted, problem still there. I have tried running with every sensor disconnected (Except crank sensor and Map) - I have seen on Lexia that the Map is working fine anyway and have also tried a good spare which made no difference, so the map seems OK.

In any case, the fact that the car starts straight up again immediately and runs fine (for a bit) is very suspicious. This has also happened now with the lambda unplugged - ran fine for a good few minutes then it happened again.

When the issue occurs, the behaviour is the same every time - very repeatable. Starts with a slight misfire, then gets steadily worse. It's like pulling the leads of the plugs one at a time, or maybe pulling them and swapping them about. You can open the throttle and it makes no difference at all, just all happens at a higher rpm.

I am starting to wonder if it could be crank sensor - it's almost like the ECU is losing track of where it is and injecting / firing out of sequence. I can't help wondering if this could happen if there were dropped pulses from the crank sensor. Sounds a bit wacky but that's kind of what it is like.

Was going to get a crank sensor to keep as a spare in the car anyway, so no issues with the expense (not too dear to be honest) so I've ordered a new Bosch one.

Other than that I am dead out of ideas. I have a video now of it, so should get it posted up soon. The weird thing is it doesn't throw codes. It will throw a code for the unplugged lambda if I leave it unplugged for long enough (no EML though) but that's the only one.

[dturbo]

Realised that the car is bank to bank injection and wasted spark, so getting 'out of sequence' can't really happen. However since TDC is detected by missing teeth (i.e. missing pulses from the crank sensor) - I think it's 3 teeth missing - then I wonder if the crank sensor starts breaking down and dropping pulses, whether it could be missing enough pulses to make the ECU inject again, thinking it's at TDC.

Putting the foot down to WOT would make use of the extra fuel and you should still get ignitions occuring at the right time in addition to any additional ones (since the crank sensor will still not produce pulses at the actual missing teeth section) , so the car would still drive 'normally'. on WOT. Might bring the oscilloscope home from work and monitor the crank sensor / injectors.

Just a theory. Failing that, it still does look like overfuelling,which is killing the car at idle / low load when it happens. But having replaced the FPR and confirming that there is vacuum there to reduce pressure at idle / light load, I'm stumped as to what could cause it. Saying that, even with the new FPR in, pullling the vacuum line off the FPR makes zero difference to idling (shouldn't the pressure increase and instantly create the problem?)

[MartinObviously]

There must be something causing the ECU to think the engine needs more fuel.

What are the values of the lambda sensor? It should really be switching from high signal to low constantly when it's warm - which would produce almost a sine wave if you graphed it.

I didn't read your first post but if it's hanging about at 0.8v+ all the time it might actually be dead.

I've come across a car before, whereby unplugging the lambda sensor still made it run like a bag of shite and the ECU did NOT run open loop as you would expect, and a new lambda sensor fixed that.

ECU fuels in open loop at WOT so lambdas are ignored.

[dturbo]

Hi Martin,

Did some more fault finding today. First off, lambda is not working. I found crimped connections at the sensor end, which fell off when I touched them. So the sensor may or may not be 'good' but it's currently out of the equation. I ordered a new one anyway as it's probably pretty old. Will fit the new one next week. So this means I can disregard the apparent rich reading shown on Lexia, it's just the reading you get with the sensor O/C.

I don't think the car is overfuelling at all (at least not when running OK). Used Lexia and watched the injector times, about 2.4mS at a warm idle. When the vac tube is attached it goes up to 3.0 ms, so the pressure reduction process is working - ECU is increasing injection time to compensate for the reduced pressure. Also the car smells fine when it's running OK.

I also got a metal plate a bit bigger than the inlet at the throttle body, enough to cover the idle bleed opening too, and used it to block off the opening. The car sucked hard on the plate and stalled (I had the little vac pipe that comes off the front of the TB blocked too). So I don't have any inlet leaks.

Also pulled off the fuel return line and confirmed it's not blocked (towards the tank).

Finally getting some error codes when it happens, I get error codes for 1-4 and 2-3 on ignition (short or open). This makes me think coil pack. However I am not ruling out crank sensor dropping pulses, as despite deliberately disconnecting the crank sensor whilst the engine was running, it never threw a code for crank sensor - wonder if there is no fault code for that?

Current thoughts are either coil pack breaking down (heat?) or crank sensor breaking down (heat?) and dropping pulses, causing false TDC detections and resulting in additional fuelling when that happens.

If it isn't crank sensor or coilpack, then I'd be thinking ECU (dry joint etc).

Crank sensor here Tuesday so will fit that and take it from there. I may also bring my scope home from work and monitor the crank sensor.

[dturbo]

So, problem solved. It was the lambda sensor. Now of course this isn't 'normal' - the car should be able to run without it, however I wonder if it is due to me having a JP4S engine with the J4 ECU. Basically it looks like the ECU tries to change something (fuel, timing) and the engine at least initially, responds differently to what is expected, so the ECU continues down whatever route of adjustment it is on, thinking more of the same is required and just gradually gets all out of whack (in the absence of O2 sensor data).

There were 3 out of 4 wires at the O2 sensor 'broken' - these were crimped joins. I had replaced the crank sensor and coil pack and it was still misbehaving. I joined up the broken lambda wires and everything sorted itself out. I can hear the ECU do something about every 10s or so, just a tiny little sort of misfire / dip. Without the O2, this degrades iteratively until it stalls, with the O2, it immediately sorts itself out.

I revved the engine to really heat the O2 sensor up and after that it was a dead flat smooth idle. I will replace the sensor as I have ordered a new one anyway and the old one will be slow by now.

One to remember! also explains why it was always fine on full throttle as the O2 reading is ignored on WOT. As for it clearing after key off / start back up, I guess the ECU resets its adaptations and starts again which is why it cleared temporarily.

[S6SAXO]

Originally Posted by dturbo

So, problem solved. It was the lambda sensor. Now of course this isn't 'normal' - the car should be able to run without it, however I wonder if it is due to me having a JP4S engine with the J4 ECU. Basically it looks like the ECU tries to change something (fuel, timing) and the engine at least initially, responds differently to what is expected, so the ECU continues down whatever route of adjustment it is on and just gradually gets all out of whack (in the absence of O2 sensor data).

There were 3 out of 4 wires 'broken' - these were crimped joins. I had replaced the crank sensor and coil pack and it was still misbehaving. I joined up the broken lambda wires and everything sorted itself out. I can hear the ECU do something about every 10s or so, just a tiny little sort of misfire / dip. Without the O2, this degrades iteratively until it stalls, with the O2, it immediately sorts itself out.

I revved the engine to really heat the O2 sensor up and after that it was a dead flat smooth idle. I will replace the sensor as I have ordered a new one anyway and the old one will be slow by now.

One to remember! also explains why it was always fine on full throttle as the O2 reading is ignored on WOT.

Great news! Well done on the perseverance dturbo.

Sent from my CLT-L09 using Tapatalk

[dturbo]

Cheers man. Quite relieved!