1.5 DCI, poor cold starting error message on display

[triumph2.5man]

Hi folks!
For the first time in 10 years and 100,000 miles our 2003 1.5dci has failed to start. Not a bad record really. One evening this week the dreaded "check injection" appeared on the display together with SERVICE light. The next day the car would not start although the starter motor spun healthily. Checked glowplugs and found all 4 to be open circuit. They are original and I suspect have probably been dud for some time as glowplugs do not normally last that long.
There was voltage at the glowplug terminals indicating no problem with the control unit.
Confidently changed the glowplugs for a Bosch set and gave the battery a boost for good measure.
I managed to start the car after several attempts before it ran on all 4 cylinders and I attributed this to excess unburnt fuel in the cylinders. Took for a long, fast run and it ran perfectly with full power as normal. Started instantly from hot every time. However "check injection" and SERVICE did not disappear from the display.
Next morning it struggled once again to start - 1 cylinder, 2, then eventually 4. Took several turns of the starter. Most noticeable though was the huge plume of blue/grey smoke that filled the whole road! After a few fast blips of the throttle, the engine settled down smoothly and the smoke cloud disappeared. I could sense the neighbours sniggering behind curtains!
Same again this morning and this evening - loads of pollution. It appears to be over-fuelling during cold start.
Battery voltage is a full 12.2 volts reducing to 9.9v at cranking speed on a very cold day. The cloud of smoke is clearly unburnt fuel and I have seen this before with an old Mondeo diesel. Attempts to start that car before the glowplug indicator went out, always resulted in a large cloud of smoke.
I have not yet carried out OBD but hope to do so this weekend.
Observant members may remark that a cloud of blue smoke on cold start-up indicates worn valve stem oil seals but that is usually a short puff that clears quickly and gets worse over time. Our problem has come on more or less over night.
How about petrol in the fuel tank? No chance!
Blocked air filter? It's near new!
Any comments will be most welcome but I will report progress.
Mike

[tomalamix]

EGR valve produces that fault if not it the closed position, producing the effects you describe..

search around and you will have enough help how to clean it

cheers

[triumph2.5man]

EGR valve stuck open gives rough running when hot and that is certainly not the case here. I clean my EGR valve every year and last did it a few weeks ago. Removing it regularly means I don't have to struggle to get it out of its housing. I removed it again today (easy!) and all was well. Thanks anyway.

What progress so far? The diagnostic test was a failure (see my other posting) so its back to logic and experience (without wishing to sound arrogant!)
It is obvious that the problem lies with the glowplugs in some way so I removed the new ones (easy job - with deep 10mm socket...) and tested them all on the battery. They all glowed after 5 seconds. Resistance readings were consistent with 1.5ohms reducing quickly to 1.1 ohms.
I checked that voltage is reaching one plug but does it reach the others? A quick check reveals that it does (I'm thinking out loud here). Is the battery weak and despite reading over 12 volts without load, is the cranking speed drop to 9volts too low? I connected another battery in line and the voltage drop was no different. Running out of ideas now! Will the plugs glow with only 9v on tap?

With my young, patient grandson at the wheel, operating the keycard and starter button repeatedly, I came to the following conclusion. It appears that the starter operates too soon (after about 3 seconds) and does not give the glowplugs time to heat up and do what a glowplug is supposed to do - burn the mixture!
The preheating period can be as long as 15 seconds and depends on the battery voltage and coolant temperature. Clearly I am not getting my 15 seconds worth.
The ECU controls the glowplug unit and clearly it seems to be doing so but it is possible that the ECU thinks the coolant temperature is higher than it is. If my reasoning is correct, the CTS (coolant temp sensor) could be faulty. The ohmeter showed 4.1k ohms at about 5 deg C. I carried out one more little experiment by removing the plug from the CTS and trying to start the car. In this state, the response from the starter is almost instant with no preheating indicating there is a connection here somewhere.
It is late Sunday evening and I will continue the battle tomorrow. It is worth noting that within seconds of the engine starting, it runs as smoothly as silk with no hesitation or stalling.
After the preheat, 9 volts is applied during cranking. After starting, the plugs receive 13.4 volts as the alternator starts charging and this continues for about 30 seconds before reducing to 0 volts. That would appear normal.

Sorry this is starting to get boring now but if anybody has any comments to make I will be pleased to hear them.
Oh for a car with an ignition key to enable the preheating period to be longer. The keycard takes this out of the driver's hands.....

Mike the boring old gasbag!

[Liam-R]

Hi there,

Im also an owner of a 1.5 DCI which blows huge plumes of white smoke out the exhaust and takes about 10 presses of the start button before the damned thing starts.

All is perfect when up and running.

One huge problem I have is that when I purchased the car it had a snapped glowplug in the head.... so theres one HUGE problem which doesnt help.

Ive removed the glow plug relay (in the passenger front arch) and rigged up my own circuit using a 100amp relay and moment switch.
It worked for a few days, but then back to struggling to start.
I have noticed air in my fuel line (between bulb and fuel filter) so bad that the fuel line drains and is completely full of air.
Ive replaced the aftermarket fuel filter with an OEM one which worked for a few days but now empties again and back to square one.

I need to purchase a new glow plug relay and HOPE that is the cause of the problem, so this could be your problem too?

Where are you located as I REALLY REALLY could do with someone who knows a lot about these darned things lol.

Ive had this problem for over a year, and it starts fine (few attempts) in summer, but COLD starting is awful!


Hope we can get to the bottom of this,

Cheers, Liam.

[triumph2.5man]

Most interesting Liam. Relieved to hear that someone else has a similar problem.
Broken glowplug? - mechanics' nightmare! As you know, to fix this would require cyl head removal. However, in an engine that is working properly, I think it would function on 3 good glowplugs. Most engines of all makes probably have a dud glowplug or 2. My plugs were very tight after 10 years and I thought at the time, if these break I will be in trouble...
In view of the work you have done I suspect you know more about car mechanics than you give yourself credit for. Your modifications to the glowplug control unit were of interest to me.
Have you done the testing that I have outlined above ie checking resistance of glowplugs, voltage applied during cranking etc? Any error messages on the display or SERVICE warning?
Clearly you have a leak somewhere along the fuel line. Our car does not have bubbles in the plastic pipes....

I have been doing more testing today (luckily I retired long ago so time is cost-free!)
I took the Meg for a long run today and the CTS is definitely faulty. At max engine temp the resistance reading was 613 ohms, considerably more than the parameter given by Manuel Haynes. On the road, the temp gauge does not go above the first graduation (say the "quarter" mark). Even leaving the engine to idle for a long time did not raise the reading and at no time did the engine cooling fan kick in. So if I replace this sensor, will it cure the problem, bearing in mind that the cost is over £30 and I am a tight***e?
Common rail diesel engines reach maximum working temperature within a matter of seconds, unlike a petrol engine where fuel is sprayed into a cold manifold or cyl head port and do not really run well until fully warmed up.
The Delphi system as used in our cars takes temperature readings from the engine coolant, the air temperature in the inlet pipe and the fuel temp at the fuel pump. I suspect that the coolant temp is probably not as critical as say a petrol engine although I cannot say which of the 3 readings is the most important, ie has the most influence on the DCU (or ECU).
I came to this conclusion today by unclipping the lead to the CTS . The result was no difference at all to the running of the engine - nothing! Putting it back on made no difference either. What did happen of course is that the temp gauge went down to zero.
Tomorrow's task will be to check the parameters of the fuel temp and air inlet sensors.
Hang on in there mate!
I live the "back of beyond" in deepest Suffolk between Ipswich and Norwich next to the briny. If I go east I get wet feet!
Watch this space - and pass out with boredom!
Mike

[Liam-R]

Well, luckily Ive been put in touch with someone who has the tools and ability to remove the broken plug without the removal of the head but at a cost of £125!

Well the thing is, I dont have a multi-meter here with me on camp (RAF Coningsby, so not TOO far at all) and its quite hard trying to test things when your on your own.

I do have error messages when I start the car, at first I unlock the car, get in, key card in and dash just says "press clutch or brake and press start" or whatever it is... As soon as I press that start button, "INJECTION FAULT" and SERVICE appear on the dash board... thats with the relay UNPLUGGED.

Im currently uploading a YOUTUBE clip of it now, I will post as soon as its finished as to what happens when I have the relay plugged in and attempt to cold start (first start of the day).

This is becoming quite annoying it truly is. Ordered 4 new BOSCH glowplugs for the car and hopefully this guy can remove the broken one and I will replace them all so I know they are of decent quality.


HERE IS THE YOUTUBE CLIP, sorry for the short length of the video but it BEEPS, shows "INJECTION FAULT" "STOP" "Immobiliser light stays on" and Battery lamp stays illuminated.

I can get a better video of my wiring and all the crap that I have done if its needed.

https://www.youtube.com/watch?v=dpGgXuz9Ypw

[triumph2.5man]

Liam, I have trawled through U tube and seen that clip before (and others!) Without a digital meter, you are really in the dark. Why not get one through ebay and ask a colleague to operate the controls - will only take a few minutes to check that a voltage is being delivered to the glowplugs during cranking. You mentioned obtaining a new glowplug control unit but if the voltage is present, clearly the control unit will not be at fault. You could save yourself a few shekels.

Now, what have I achieved today? I bit the bullet and obtained a new CTS from my local factor which cost under £15 (my previous price of £30 was, surprisingly, from an ebay retailer!)
The storeman told me two sensors are available, one from 2002 (described as a "temperature sensor" with 3 pins) and one from 2005 with 4 pins. I was able to examine both. The former is a screw-in type and there is nothing like that on our car. The second one (a push-in type, retained by a U shaped plastic clip) is the one identical to the one in our 2003 car so where these dates come from is a mystery. Thus I bought the push-in one with 4 pins although only 2 of them are used.
I checked the resistance cold and it was spot on with the manual, around 2000ohms, compared to the old one at 6000ohms.
Having fitted it to the car (just a small loss of coolant), I waited with baited breath while I started the car. It was a struggle to start and a neighbour remarked that she thought the car was on fire! So no change there then. Disappointment all round....
I took for a test run and two things were immediately evident; 1) the temp needle reached the middle of the gauge 2) the car ran even smoother than before and ticked over like a swiss watch. It ran so well I was impressed. Having an accurate CTS does obviously affect the fuel injection.
I also checked the fuel temperature sensor which is on the back of the pump. Manuel Haynes gives the resistance as 2000ohms but does not say if this is hot or cold. The sensor in the car showed 4070ohms cold and just over 1000 hot so there appears to be a discrepancy there. To replace this might mean paying a visit to Monsewer Renault - expensive!
Yesterday I mentioned a third temp sensor - air intake/air flow combined. The manual shows a picture of it on the air filter housing. Having hunted around the engine bay I came to the conclusion that our car does NOT have an air flow/temp sensor!!! There is nothing between the air filter and the turbo except a plastic tube! This is very strange because the technical Delphi details show that air temp is one of the parameters fed to the DCU!
Ah well, if it ain''t there, no need to worry about it. By deduction therefore, only 2 sensors provide the brain with temperature details. I am ignoring the sensor in the wing mirror which feeds external temperature to the dash display as I do not think that is relevant.
More to follow tomorrow. "The continuing saga of the maddening Megane!"
Finally I had another go with the EODB scanner. Would it find the "check injection" fault or see that the MIL light is on? No chance. It declared that all was hunky-dorey. Waste of space.
Mike

[Liam-R]

Hi Mike,

Well I only uploaded that clip last night so you couldnt have seen mine, maybe another one on someone elses channel with the same fault.

Well, the thing is if there is a CONSTANT voltage to the glow plugs (with ignition off) then the glow plug relay module is a the offending item. Ive searched many forums and they all confirm that if this is the case then a new relay should do the trick.

Where is this manual you speak of? Is it the green Haynes manual? I have one in the boot of the megane, just not really got round to looking at it in that much detail. If its not the Haynes, what one are you using? I dont seem to be able to see the Voltage readings in there for the glow circuit?

I will have to have a look at my CTS... problem is though, when the car has started and is upto temp the needle sits on the middle of the dial and car runs perfectly. Im in the exact same boat when it comes to this Megane, its so frustrating with that bleeding push button start. Ive been reading that on the Clio DCI (also the 1.5 k9k engine) they have trouble with 3 relays in the engine bay, and this is the main cause for cold starting trouble, but Ive had all my circuit boards out of the engine bay and no fuses or relays have gone (that I can see)

I think I'm going to take my car and get the broken glow plug removed, replace all with decent BOSCH plugs and HOPEFULLY get it onto a CAN-CLIP diagnostic as this will give me ALL fault codes as opposed to the ZERO codes that all the readers Ive tried have given me

Seems, we have the exact same problem, but just have no idea where to look next.

I also have been told that the "crankshaft position sensor" and the "camshaft position sensor" have roles in starting the vehicle, so if these were faulty it would have trouble starting.

Could you possibly give me a "how to" on which holes in the relay plug I need to test, I'm guessing pins 1,2,6 and 7?

Hope we can get this figured out and sorted together haha!

Liam.

[tomalamix]

ok, i see you've done the EGR cleaning process very recently. by the way, the EGR valve work always, either hot or cold, its not related with the engine temperature, but lets leave that appart beucase there are no issues there.

the 9v during crank is usual, the starter is under usage so the tension will drop, is normal.

the 3 seconds for the glow plugs can be normal, several brands do that, in fact the audi's skoda mercedes etc etc sometime ago had a system that when you unlock the car the ECU checks the temperature of the engine. if necessary it will heat up the glow plugs since the moment the car unlocks and then only needs to heat up the fuel a little bit just before the starter kicks in. Again, this was sometime ago, 6 - 7 years ago, i dont know if that is still under usage, but it makes sense to me that is still under usage. Renault can use that sstrategy maybe..needs confirmation..

despite all the Service and Check injection appears on your screen so there is something for sure wrong and the ECU detects it so it must be visible on the life data. check the voltage on the glow plugs, check the coolant temp, Turbo boost and several other stuff. that info is possible to see during the starting process and maybe is possible to check if there is something missing or if there is something more than it should

good luck

[AlexB]

Hi Mike,

It would really help to see the live data plots from your engine. If you don't know how to plot these, then just post here the log in any format; I'll hack a gawk and gnuplot scripts for it. Start logging on the cold engine, preferably when it is not running yet.

To your observations. The temp sensor on the pump is actually within the range as 1000Ohm<2000Ohm<4070Ohm.

What you are describing looks like a problem with the temperature sensor. I know in some models there are two sensors in one for some reason.

Forget about EGR. There is nothing to recirculate when the engine is not running. So, it does not affect starting. The same is with the turbo boost or O2 sensors.

It can be MAP/MAF or the stop throttle...

At 2C, the preheating in your engine is supposed to last about 2 seconds, and postheating about 60 seconds. So, everything looks OK here.

Check if the glow plugs are of a correct type. My F9Q-800 uses Beru, which work well and last long.

By description it can be a problem with ramp pressure or the timing belt or slow cranking or... Just post here the OBD log!

[Liam-R]

Hi Alex,

How would I got about getting the live data plots from the engine?

Ive tried all sorts of OBD2 scanners and nothing is happening at all...

I have the exact same problem but if you take a look at my video, there it shows how my car reacts with the glow plug relay connected, but when its disconnected it goes through the start up sequence.


Cheers, Liam.

[triumph2.5man]

Hi my many followers!
You will be pleased to hear that I have solved the problem! The Megane starts instantly and the neighbours do not need to call the fire brigade!
What was the answer? It doesn't matter how much experience you have or how long you have been a "spanner monkey" (55 years in my case!) modern cars cannot be treated the same as the old uns I was brought up with.
I have detailed here all the work I have done to solve the problem and I realised that I was approaching the end of the road when it came to logic and up to about 60 quid in parts.
I failed with the OBD scanner (no use for fuel injection problems) but a better system was available to me via a friend in the trade. The Snap-On (£4000) scanner which I was able to (carefully) use, was just the job! No messing about here! Although nowhere near as comprehensive as the Renault Clip system, it is good enough for most jobs and many small garages will possess one.
The fault codes showed up at once - all 5 of them. Some were incidents I had caused myself ie, disconnecting CTS, disconnecting glow plugs etc, but the most interesting was a report of a problem with no2 injector. I cleared all the DTCs and was able to run a test on each injector whereby, presumably, the injectors are pulsed for about 10 seconds. Upon starting the car, the SERVICE light had gone out and the "check injection" message had disappeared from the display. I ran the engine for a while and no messages appeared. A long test run followed. Car ran impeccably! But the big test would come the next morning at cold start.
With thumping heart and fingers crossed I pressed that start button early this morning. Engine sprang into life instantly and quickly settled down to a smooth idle. No smoke, nothing!
I waited many hours before trying again (being a cynical individual!) but once again a perfect start. Breathe sigh of relief!
So what had happened here? For a start (pun intended), the glowplug problem was a red herring. Modern common rail,high pressure diesels are so efficient that, in practice, most of the time the glowplugs are not needed (unless maybe, you live in Siberia!) so I wasted a lot of time on these items.
With the benefit of hindsight it is apparent that injector no2 was over-fuelling which is where the clouds of smoke came in, something I suspected from the outset. Clearing the code and pulsing the injector cured the problem. It may seem unbelievable but the proof of the pudding is in the eating.
If the problem returns, I will purchase and fit a single injector. The Snap-On system enables injector codes to be entered.
Liam my advice to you is forget about paying to have that broken glowplug removed. As a priority, do have your car diagnosed on a reasonable system. It might surprise you. The fact that one glowplug is not connected will not trigger a DTC.
For me this has been a steep learning curve (wot, still learning at three score years and ten - I should be slowing down!) and I do hope that what I have written here will help somebody, somewhere, sometime. I can now get on with all the other jobs I have to do. There is life beyond Megane.....looks as if daughter will be keeping it....
Mike

[Liam-R]

Amazing news, I have sent you a PM Mike

Hopefully you can help me out.

Cheers

[AlexB]

Well done, Mike! It looks like you've invented a new method of cleaning the injectors. I new there was a mode for the leak test, but did not realise it could be used for cleaning. It makes sense: the leaking injector reduced the rail pressure while letting the fuel in at a wrong time, and then the neighbours rand the fire brigade... 1.5dci uses a different fuel system from 1.9dci, hence problems.

[AlexB]

Liam,

Clouds of smoke can be caused by a leaking injector, and the air can get into the system through the venturi tube. It may happen in the case of a fault with the return from the injectors.

A sad short beep in the video it is the system recognising that a part of it is gone.

You probably don't need the scanner any longer. It might show a low fuel pressure due to air coming through the venturi. I take the log with scanmaster connected to s laptop., then plot them with e.g. Excel.