MatraSport Forum

After running the car for two years without the rad cap on, it was time to admit it. Either scrap it or fix it. Well the car has the leather, three sunroofs, I tinted the windows, fitted sat nav, dvd head rests, 240v inverter, rear air bags & otherwise up to date. Also last yr did the gearbox which was an epic of a job if you remember. So as we are due to drive from the UK to Germany I decided to swap out the engine from the spare car. I really now wish I swapped my gearbox into the spare as now I have created my new nightmare!   Both cars were running before I started this repair, now all I get is three pops of backfire & then nothing other than a smell of petrol if I keep cranking. I have double checked all wiring connectors & earths. I changed the crank sensor & swopped out the ecus as I thought they may have changed the reluctor. Sounds like it wants to fire on the wrong cylinders then nothing. I have a clip, no faults found in engine or controler. Has oil pressure fuel & spark. I can see all sensors, Injectors timing, coil timing in clip. I can actuate bank a,b,c injectors etc, I thought I was a good mechanic but this has me beat! All thoughts welcome.
Oh, we go on holiday In a days time so I just added a dci to the fleet! I would still rather take the lx7 if you know where I have gone wrong.
Is it poss that the ecu has a memory where the engine stopped last time. I see the ignition timing change as it cranks but maybe the cylinders are too wet by then to start.

 I did not see any replies to this problem, has it been sorted?

Terry

No not yet, but I need to get it running before the end of the month. I have tried to turn the engine by hand until the next cylinder compression, in the hope to change the timing but with no joy. I assume this is waking up the ecu & it is reading the change in position of the crank, i'm going to try this with the tdc sensor disconnected. Do you have a similar problem, or a cure maybe? I should be so lucky, lucky, lucky, lucky......

This won't make any difference Brian if I understand what you're implying.  The TDC sensor picks up from a set of registers where there is one missing to denote TDC so it always knows where TDC is by this, whilst it reads the rpm from the rate of all the others passing it.  If it uses the flywheel teeth as the registers for rpm then there is a single register for the TDC position, plus there is often a cam sensor too, so this is used not only as a cross check, but allows for quicker response since the ECU knows which of the cycles the engine is on.

Since you have changed engines over, did you check that both flywheels were the same and fitted in the same place so the timing marks would be correct?  (I can't remember this engine and whether the flywheel only fits in one position).  It does sound from your description as if the timing is out, which is why I wonder about the flywheel registers...

Also you mention swapping ECUs so I assume you have a way to get around the immobiliser coding or did you swap all three items so they match?

Roy

Hi Roy, thanks for your input. I did indeed swop out the ecu, body controller & immo chip from the donor car. & it did the same turn - pop- die as with my original ecu set. Looking at the figures in clip I can see the timing change when cranking. I just thought it would remember where the crank stopped rather than wait for the tdc marker. This engine does not have the cam sensor. All a bit strange. I didnt think to check the flyweel position as I didnt take it off.  Looking at the wiring loom, it is quite short & is direct from the ecu to the coil pack without any loom connectors. Unlike the injector loom which breaks at the back of the engine.

I checked the wiring from the ecu to the coil pack plug today, both cars are the same. I had swapped out the coil packs incase one was faulty, I can't imagine both are duff. Both engines were runners before I started this swap, I am befuddled!

To recap Brian, you say they turn over fine, and there is fuel and spark so it suggests the timing is incorrect.

If the engine is getting fuel but is not starting, then after a short period of turning the engine over without success, the spark plugs should be wet with fuel.  Have you removed the plugs and confirmed this?  If you have, and you definitely have a spark at each plug, and it won't start then leaving aside compression (which we'll assume is O.K.) then it has to be timing.

So if the plugs are getting wet, how have you proved the plugs are definitely sparking?  This engine has coils sitting over the plugs so it is not easy to test as when we had H.T.Leads and you could simply hold the end of one near earth and watch the spark as the engine was spun over.  So you need to make sure you really do have a spark, not just an electrical supply to the coils.

Now we come back to what I still think is the most likely.  These engines were running before you swapped them.  After the swap neither will run.  Since they get their timing from a combination of flywheel, TDC sensor, and injection computer, and wiring, something has to be wrong ON BOTH vehicles.  It would be very unlikely to have two damaged or faulty components  i.e. both TDC sensors having failed; or both ECUs etc. so the one area which I feel the most likely suspect for an error is with the flywheel/pick-up.  In doing the swap something has been changed or overlooked that has thrown the timing out.  You may have to pull them apart again to inspect and compare them carefully.

I have seen engines put in before, which won't then start, and it has been because there was something different with the flywheel...

Now if each flywheel stayed on the original engines, it could be that the position of the pick-up is different, or the sensor wiring reversed.  The TDC sensor wiring is polarity conscious.  We once changed a sensor only to find the engine wouldn't run afterwards, and the fault was that although the TDC sensor plug will only fit one way, the two connections inside it had been reversed, so wrong polarity, and no start.  As soon as we reversed the connections it started fine.

You may have checked the obvious things, and even checked them many times, but as someone once said, once you have exhausted all normal possibilities, you must then consider the other things no matter how much they may not seem right.  Please let me know what you find Brian.

Roy

Hi Roy, it is nice to have you to bouce this problem off!!
I'm a mechanic of old school, I did my apprenticeship with a BL dealership. So there is nothing I dont know about old allegro, marinas, tr7, mgb etc, then MOTs eventually with the RAC. At this time we had mk3 escorts with the famous duff distributors & other cars where electronics were being brought in to engine mangement. Then I stepped out of the industry until a few years ago. I like to think my fault tracing is logical. I've read up on the system in the Espace & it's pretty basic with just the tdc sensor & a few others to keep it going. I have only swapped out one engine the faulty engine is sat under cover in the drive here. So just trying to get one car going. I was thinking that if the flywheel was fitted in a different position then the wiring would be different from ecu to the coilpack. The reason I brought the ecu over from the donor car was because I suspected this. Yes the plugs get wet, there certainly is a spark, I used easy start down the inlet holding the butterfly open with a screwdriver - the backfire almost blew my hand off!!  ;D I'm going to put my old strobe light on it to see whats going on & i'll report back. Oh, the first thing I did was to bring over the tdc sensor from the donor.
I just tried again & pop, pop, pop, this is before there is a full revolution of the flywheel. That is what makes me think that there is a memory of where the crank stopped the last time it ran.

This is an ignorant amateur's thought process -- offered with caution, to be used with even more caution. When this fool rushes in, he may trample all over the angels...

If you're getting backfiring, at least one cylinder must be sparking while its inlet valve is open. I can think of (only) three categories for a possible cause: 1. control electronics/software faulty, 2. input from sensors to electronics incorrect, or 3. output of electronic controls mis-wired. It seems that 2. has been ruled out by swapping ECU etc., and 1. is well covered by investigations so far.

That leaves incorrect connection of the ECU output. Is there a way to confirm that the spark plugs are connected correctly, so that the cylinders fire in the intended sequence? If cylinder number 1 sparks when the ECU intends to spark number 3, for example, you might get popping noises as described? Connecting the injectors in the wrong sequence would prevent the engine starting but couldn't cause a spark while the inlet valve was open.

Is there any other output from the ECU to the engine apart from spark plugs and injectors? Does it also control valve timing or throttle opening? Presumably it's not possible to connect the spark plugs to the connections intended for injectors.

(My Espace's engine's more primitive, with one coil and a mechanical distributor, so mis-wiring the spark plugs would be easy, but obvious. The family's much newer other car has a coil over each plug, but still with a lead to each, so could be connected in the wrong order.)

If my thoughts are completely wrong, I suppose they may at least give the experts a bit of a laugh.

  - Mark.

Hi Mark, you are quite right in your thinking. Very good for an amateur! I thought. Today I used a variable resistor on the temp sender & a hair dryer in the inlet hose to convince the ecu that it was a hot engine & thus not flood the plugs with fuel. So I cranked it over & got lots of popping back instead of just pop pop pop & then having to take the front plugs out to dry them. I then used some fly leads to cross the ecu to coil pack connector so I could get it to fire on different cylinders. Still the same, backfiring! didn't seem to make much difference which coil pack I connected to. So next is to scope the tdc sensor & mark up tdc no.1 on the crank pulley so I can use a strobe light to see where it is firing. Will update when I can, thanks for the conversation, all thoughts are welcome & keep my brain ticking :)

I looked at MR315Espace_JE_A_E_D_Engine.pdf from the vault -- is this right for your car? -- for  the layout. That makes it look as though this engine (Z7X) has 3 ignition coils, each producing sparks for two cylinders at the same time: 1 and 5, 2 and 6, 3 and 4. (Does this remind anyone else of the primitive 2-cylinder engine fitted to the Citroen 2CV?) The ECU controls which pair of plugs to spark based on a non-adjustable crankshaft position sensor.

The camshafts control the valve opening, entirely mechanically, so only one of each pair of cylinders should have any air-fuel mixture in it when the relevant pair of spark plugs fire. The other should be also close to TDC but between exhaust and inlet strokes.

All right, here's my next thing I would try to check: has the cam belt been disturbed during your engine transplant? If the camshaft timing is wrong for either row of cylinders, or both, then you could get backfiring. The electronics could be working perfectly.

As before, do laugh at my errors. I'm a software engineer, not a garage technician!

  - Mark.

Yes thats the right engine, just a big 2cv!!   Looking at where the slot is in the flywheel that denotes crankshaft position, it is in the same place as my other engine. I didn't even remove the cam covers on this so no chance of the cams out of timing.
I'm back to thinkng that there is a memory of where the ecu saw the crank stop the last time it ran as it pops before it gets to the marker. Or the crank has to be in a certain position after the ecu has been disconnected for a period. It would be different if there were cam sensors that would give the ecu more information. I checked the crank sensor & it appears ok on my scope.

Still in the "ask innocent questions" mode: there's a note about connecting or installing the two pinking sensors the right way round in the manual in the vault. Around page 224 of the PDF, ...

"When the sensors are removed they must be repositioned correctly so that the computer may correct the advance setting if necessary for the cylinder or cylinders concerned.

"Pinking sensor no 1, with the green connector, must be fitted to the front cylinder bank (corresponding to cylinders 1, 2 and 3).

"Pinking sensor no 2 with the blue connector must be fitted to the rear cylinder bank (corresponding to cylinders 4, 5 and 6)."

Are these one of the clues which the ECU uses to work out which cylinder it is supposed to be injecting fuel into, and when to fire the spark? Can they be connected the wrong way round, and would that confuse it enough to cause misfiring? (I'm still puzzled how the ECU can work out which injector to operate, without a signal from the camshaft, in a 4-stroke engine. Perhaps it just sprays fuel out of all of them, and relies on 5/6 of the petrol waiting around until the relevant inlet valve opens.)

I'm not familiar with the Espace ECU and the way it works, but have worked on a few others. It won't remember where the crank is when it stopped. It will just sense the gap that Roy described to "sense" TDC. What if the engine was turned without the ignition on? It has to check from scratch.
Even if you think the cam belt hasn't been disturbed, it is worth checking. It may well have slipped a few teeth and is one of the few things easy to check.
It does sound like the TDC sensor isn't detecting TDC at the right time though. Flywheel or sensor positioning issue?
No camshaft sensor is unusual as well! Glad I have a primitive diesel!! :-)

Good luck,
Graham

There is a marker in the flywheel that the sensor picks up to determine tdc but as you say this is a 4 stroke engine which means that the crank rotates @ a different speed to the camshaft. Also the marker does not appear to be tdc on cly no. 1. With so many modern engines ther are no timing markings other than to time the camshafts & this is done when all the pistons are equidistant from the cylinder head as to allow the valves to open when positioning the camshafts. To determine tdc on no.1 i have viewed the camshaft lobe through the oil filler in the cam cover.   The document you reference to is for the older engine, mine only has the one sensor, interestingly the newer vvc engine reverted to two!
If it does not remember where the crank is when stopped, then how come it fires before it does one revolution before it sees the tdc marker?  Flywheel has not been touched nor camshafts, just complete engine out of one to another. I can see the gap in the flywheel, it is in the same position as other engine, I scoped the senor & it sees the gap. Swapped out sensor, swapped out coil pack, checked wiring, no faults in ecu. Swapped ecu even. Will check ecu to coil pulses next.

As Graham says Brian and I've said before, the ECU has no memory about where the flywheel was or stops, it simply starts again from scratch each time you attempt to start the engine.  This is why the start is quicker when you have a cam sensor too.

Now if it is firing the mixture even before it has 'seen' the TDC sensor mark, that suggests a fault.  Bear in mind though that if cranking speed is 240 rpm which is quite likely, then the flywheel will rotate 4 times in a second, so it will only take a max. of 0.25 of a second to start firing, and since the timing mark may be closer it could start firing even quicker.

Since you have a scope, you really need to scope the feeds to the coils and the firing pulses if you can.  Injectors are not so critical since they don't need to be timed and very often are not. Some engines actually open multiple injectors at one time, because as they are injecting into the intake manifold, some may be injecting in at just the right time to be drawn in to a particular cylinder whilst others may be spraying in whilst the valves are closed but it will 'hang around' and get drawn in when they open which will only be fractions of a second later.  You can work out the timing of opening and closing if you wish but at even 1000 rpm you are talking in milliseconds.

If it really is firing before the TDC has picked up and 'knows' the position/timing, then that might indicate you have a coil or coils firing randomly, maybe because of a short.

However, you have to go back to the fact that it worked before you swapped it over, and immediately after it won't start.  So logic says it has something to do with the swap.  Something that has been done, or not done, or some fault that has been introduced, such as a wire cut and shorted by the engine as it was lifted in or out...  It is rare and probably very high odds that something unrelated has happened at the very same time.

Regarding the backfire of the Easystart in the inlet, be careful (in both senses!) not only of injury but in misinterpretation.  By holding the butterfly open and using Easystart you alter the mixture ratio and because of valve overlap it is always likely that some ignition flame will come back in the intake.

P.S. Isn't this the engine with three coil packs at the 'front' on one bank of the V6 with H.T. leads to the rear bank?  If it is, then you could at least get a connection in to the H.T. feeds to the rear plugs.

Roy

This has me dumfounded! I dont really know what to do with it. It would be a shame to scrap it after all the work I have put in. I could do with it on the road next weekend but think i'm going to have to hire a mpv again for the family weekend away. Fecking thing. Yes it injects fuel & the plugs get wet. As I spin it over you can see the timing change on clip but then it is too late as the plugs are wet. I put an adjustable resistor on the temp sender so it thinks its hot & not inject as much but it just backfires when cranked. Maybe the injection loom is shorted & its firing fuel all the time.

I understand your frustration, Brian. This car deserves to run.

I read your post,, and I've been thinking a bit. I thought about wires, but you say you have the CLIP on, which reports everything ok. I thought about cam sensors, which 6-cylinder engines with wasted spark like this one incidentially do not need (unlike a 4 cylinder engine with wasted spark, and the later V6 used in the Avantime, which has a coil per cylinder and therefore need to know exactly which coil to fire).

I thought about ignition and the connection to the autobox. The gearbox retards ignition when it's shifting. Could there be a problem here? It's retarding so much that the engine can't start?

And then it occurred to me, that you may have a problem with the fitting of the flywheel (which is not really a flywheel on the autobox). I think I have a picture of it somewhere... yes here it is:

(http://gallery.dinsen.net/d/12764-5/DSC_8947.JPG)

The TDC opening is on the left, barely visible in the picture (the flywheel was taken off my old engine when it was replaced with a new from a Laguna with manual box, after the cam belt came off 5 years ago).

Anyway, there's spark and fuel, but the spark must be coming at the wrong time. Have I missed something? I don't think so.

My suggestion is to get the gearbox off, and recheck the fitting of the flywheel on the engine. Something must be wrong...

Good luck!

I'll be picking up the Espace today. Really looking forward to going home to Denmark in it now :-)

Cheers,
Anders

I'm not sure I understand you here Brian.  You say it injects fuel and the plugs get wet.  So whilst that tells us there is fuel getting in, it also says it is not being fired.  If it was the plugs would not be wet.  Wet plugs indicate fuel but no ignition of that fuel.  So as I've said before it appears the timing is out and/or you are not getting a good spark.

However, you also say you can see the timing change on Clip as you spin it over.  This I don't understand.  It is quite a few years now since I actually used a Clip, but I don't remember this.  Can you explain a little more please.


As you crank what does it show for the length of the injection pulses?  This should be in milli-seconds.

However, I don't think this is a fuelling issue Brian, it is a timing issue.  Since these modern engines don't have a key anymore to lock the gear to the crank, maybe the gear has slipped on the crank and the cam timing has slipped?  You really need to start from basics and check it all carefully.

Roy

Hi Anders, glad to hear your getting yours back & heding home. I never removed the flywheel & I have checked that the reluctor is in the same place on this engine as my old one. I had brought the original ecu to this car also at one point just to see if there was a difference. There wasn't! crank, backfire, plugs then get wet & I have to remove them again to dry them. The gbox theory is now in the mix!

So Roy, it appears as if it stops sparking after the backfire. With the resistor on the temp sender it appears to be injecting less as i'm giving it a high temp reading. From memory clip displays the three coil pack advance figures. as i crank these change significantly, i cant remember the figures though. It is not easy to determine timing on an engine with no timing marks that can be seen. Neither the cambelt nor the reluctor were touched during the exchange. I need to check injector & ignition pulses, this is my next step.
to be continued.......

I remember you said earlier on Brian, you had not disturbed the flywheel and therefore the timing pick up so that can be discounted; so if the computer is 'seeing' that it should be able to calculate the correct timing - unless something else is wrong.  Since yours does not have a cam sensor, I was suggesting maybe the cam timing has slipped - not that you have done anything to it, but unknown to you it has moved.  That could account for a timing issue.

However, it is interesting that you say you think the firing stops after the initial backfire.  That needs to be investigated further.  Obviously if it does not continue to fire then that sounds like the root problem.  I think you said you have a scope, so that should be easy to check - you could even disconnect the injectors to stop them fuelling, and simply crank the engine for a period and watch the firing pulses - if they are stopping that should be easy to see.  Why might be harder to figure!  Could the knock sensor be involved here?  It should trigger a retard of the firing via the computer if it detects a knock but the backfire could be causing more of a problem.  Or if it can't detect the knock sensor, maybe it is shutting down because it won't know when there is a need to retard the firing.

These computers have failsafe modes where they shut down if they can't detect a signal they need, to save the engine from any possible damage.  Other times they have a 'limp home' mode where they allow a restricted running, but it may be yours is going into shut down?

Regarding not being able to see the timing, why not do a static timing set up and mark either the flywheel or crank pulley, wherever you can get a timing light signal on and see it, then connect a timing light and crank the engine and see if it really is firing and at the right time?

Roy

Hi all, looking at this again. From what I can tell, this is firing on coil 1, when cranking, at 40 degrees. I thought it would be a lot more but I guess this is why I get a pop back through the inlet. I need to check that the other coils are respective & that there is not a short somewhere that makes them all fire at the same time.
Why would this be the case though? After all I never had the flywheel off. One bolt hole is 45 degrees which doesnt help me. I would have thought it would start at 5 - 10 degrees. Report again tomorrow

Oh dear  :(  crank is out 80 degrees to the camshafts. The ac compressor is siezed, maybe this held the pully from turning? All very strange.
 Turning by hand I can feel what I think is piston against valve. Off to buy another car for the holiday, again!

Odd, but if the compressor had seize, surely the belt should be slipping to begin with and that shouldn't affect the crankshaft? I thought compressors are engaged by a electrical "clutch", unless something is feeding the electrics permanently.

Yes all very strange, but there is no keyway on the crank to the sprocket to my knowledge. Just the tightness of the center bolt. I had pressure cleaned the engine before I fitted it so maybe this added to the siezure. The center of the pump will rotate but the pully is locked solid.

Ah so I was correct in saying that the timing had somehow slipped.  Thanks for posting the answer Brian.  It is always good to know the answer to any puzzle like this as it adds to ones knowledge and experience.  Sorry it had to be something like this though.  Hope you find it has not done any damage when you get to repair it (assuming you will).  I did a repair on a Focus recently where the belt had snapped but only when the engine was being started, so although they are supposed to have valves hit pistons and bend them, if a belt has broken, as this had only been turning on the starter it turned out none were damaged and it just needed retiming and a new belt.  Finger crossed yours may be the same.

Roy

Yes you were right :)  I'm going to use an old belt on it if it has stripped teeth just to see if it will run. Then i'll fix it up. (with a new belt at least) I still have the old engine so I have plenty of parts. It will go back into its original car & may be sell it, it a 'Initiale' so a nice model & has to be worth something!
I have a low mileage Laguna engine (already had the cambelt changed) & box that I want to fit in this one I'm keeping. Its the vvc fly by wire. So I need a scraped laguna for the throttle, wiring, ecu etc, or an Avantine of course.