Am I being thick? VR4 will not start.

[Nick Mann]

I'm having trouble getting a VR4 to start.

It is a prefacelift legnum manual conversion running a map2ecu.

It was running fine last year (mostly fine, it had an occasional misfire on cruise) but then was stood around doing nothing for a few months. The engine turns like a goodun and just occasionally catches but never for long - so far not long enough to keep spinning while I let go of the key in the ignition!

Things I have done:
Added fuel (10 litres) just in case the fuel level was too low.
Swapped coil packs.
Changed cam sensor.
Changed IFS.
Checked stock ECU using evoscan - no issues that I can see.
Checked MAP 2 using MAPCAL software, again no issues that I can see.
Swapped ecu.
Swapped MAP 2 ECU.
Checked wiring continuity to coil packs.
Checked other wiring to MAP 2. (Not that I expected anything - stock ECU via evoscan is reporting barometric pressure, inlet air temp and air flow, so wiring looked okay from that perspective too.)
Checked all fuses, cabin & engine bay.
Changed stock FPR for another stock FPR.
Changed fuel pump to another stock fuel pump.
Cranked engine with fuel pump pumping into a bucket, there was plenty of flow!
Completely removed Clifford alarm/immobiliser.
Changed ignition switch.

I am out of sensible ideas at the moment. I have used a spare HT lead with a plug in the end to check for a spark and I can see one, although the ignition switch is a fair distance from the plug and it is bright outside which makes it difficult to tell if it is a strong spark. Nothing I do seems to make any difference. Sometimes it sounds like it is going to fire but more often it is just cranking. When it does fire it is only for a few revolutions.

Anyone got any ideas? Have I missed something obvious?

[adaxo]

Only one thing left is the crank sensor?? i know its not an easy to swap for a gudun but from your list this what s left not done?

[Nick Mann]

Yeah, I've not done that yet. Would Evoscan show up a lack of measurement anywhere if that was the issue?

[adaxo]

Yeah, I've not done that yet. Would Evoscan show up a lack of measurement anywhere if that was the issue?

Dont know about evoscan and how to check this working either, may galant PDF manual have some procedure written?

[fassi1]

Timing belt jumped? Is coolant temp. correct?

[Nick Mann]

Coolant temp is good in Evoscan. I need to take more cam covers off to check all the markings. The front cams agree with each other but that's all I can say at the moment.

It does look like I am going to have to take it apart more. (sigh)

[adaxo]

Galant manual 13A-17 there is few steps how to check it

[adaxo]

66303

66304

[Davezj]

Nick have you full pushed home the cam and cranck sensor.
i plugged mine in once and did not push one of them home until they clicked this stoped the car starting, but as it was the only thing nemoved to stop the engine from firing when doing a compression test, so it was an easy find and fix for me.
just a thought

[Nick Mann]

Cheers, Adam, I've found that too. Much appreciated. It gives me a check to do without taking even more of the car apart.

Yes, Dave, they are fully pushed home. I sometimes disable my car that way if I am worried about where I am leaving it.

[TAR]

As its nearly firing, it may be just a poor battery. Try a jump start at the same time. I've known cars refuse to start after sitting for a while and all it needed was a kick from an extra battery.

Also, try disconnecting the MAP2ECU as this is controlling the sparks and may be messing the timing slightly - I hope it has a removable harness

good luck! :happy:

[exevoowner]

Air filter over fueling ?

[Nick Mann]

Air filter is neither here nor there, Nigel. The MAP ecu is controlling air flow readings to the ECU and I have tried starting it with the intake pipework removed.

It could be over fuelling, but it shouldn't be - evoscan is reporting injector duty cylces of just over 4% on cranking, which is about what it should be. Fuel pump, injectors and fpr are all stock.

[Nick Mann]

Tim - it is being jumped all the time, I am trying things often enough that a trickle charger is not able to keep up! It is turning over plenty fast enough - evoscan & the map 2 report rpms of 185-200 which is as good as it ever gets with the starter motor.

The map 2 isn't wired in with a removable harness but the coil pack signals are easy enough to remove from the map 2 - they are the only thing on that multiplug. A simple short to reconnect the signal wires from the stock ecu shouldn't take long.

Thanks for all the ideas guys - tomorrow I will be doing the crank/cam/coil checks from the manual and wiring out the coil packs from the map 2.

[adaxo]

I would say (in standard car without mapecu) seems like you got spark and fuel (after cranking you can smell petrol on spark plug?) so it's 99% something which managing to run engine in right order is dead ie maf, crank, camshaft sensors.


good luck

[Nick Mann]

I can smell fuel after cranking. I can see a spark although I think it may be weak. It may be that there is an issue with the coil packs not sparking strongly enough - just occasionally the spark is enough to ignite. I've not had a good look at the earthing either today, that is another thing to check out, general earthing and also earthing of the various sensors/controls.

The last time I had one as bad as this it turned out to be the aftermarket FPR had split its diaphragm so was not holding pressure - the car would fire intermittantly but straight away die. That's why I have changed so much on the fuelling side!

[Trotty]

I would change the plugs. My bet is they are fouled up causing weak spark.

[Nick Mann]

You could be right, Daniel. It's something that had already gone on the list in my mind whilst I was eating dinner. It's certainly one of the few things I haven't done. Compression test is in my head too, so I can do that at the same time.

[Davezj]

I had a thought, it is more of a running issue than a starting issue, but thought I would chuck it out there as it is a really easy thing to test.
I had a issue with the MAF at one point, the car would try and start but would not fire up properly, I disconnected the MAF and it started right up. But as soon as I plugged the MAF in the car would stall.
As I say it is more a running issue, or top put a more appropriate name to it, continuing to run issue, but it is a very easy thing to test.

[Nick Mann]

MAF is unplugged at the moment. The car has a MAP2 on it running in MAP mode (MAF elimination). However the kvf values the stock ecu is seeing are spot on for cranking scenarios, the barometric pressure is fine and the inlet air temperature is fine. I have also swapped the MAP2 ECU for my own. Same values at the stock ecu and no difference to the non-starting.

Keep the ideas coming though - it's all good!

[hbkuk1]

Ignition failure sensor? this was a common thing on ftos if the battery went totally dead and was jumped they popped sometimes I would also force firmware update and then reinstall map on the map2 I've had a similar issue in the past where it just stopped and after much back on forth on mapecu forum I reflagged firmware and all was fine

[Nick Mann]

I've already changed the IFS. It's such a quick job it was about the first thing I did! ;)

My last car ran a MAP2 and I still have it, so I have swapped them over to double check. I'm going to go with cam/crank/coil pack resistance measurements next, including earth path resistance, then assuming that all comes back good I will change the spark plugs. TBH if I don't get it running today I will be stripping it for parts.

[WOODY72]

If you don't get it running and strip it, it will plague you for years. Bit like being married really!

[Adam.Findlay]

with the car sitting for a while. possibly the map2ecu has lost its calibration. Kind of like a old gameboy cartrage internal battery dying and deleting your saved games.
bit of a far fetched Idea but seems you have covered all your other basis without scoping the injector pulse and coil driver signal.

but then again the MAP ecu's are piggy back, Correct?
so the stock ecu should still fire the car even if the mapecu was not altering any of the signals

[Nick Mann]

Sorted.

Cam/crank sensors and coil pack signals were all good and checked out as far as I could with one set of hands and no ability to check with the car running. So I moved on to the plugs.

The plugs were black as the ace of spades. They look quite new as well other than that. I put new plugs in and started the car - it did start but it still didn't want to run. It became apparent quite quickly that the car was running very rich. I plugged the laptop in to evoscan and couldn't see anything obvious, then plugged the laptop in to the map2 which seemed to think that the cars vacuum was running at -1.8inhg. I figured that the map2's map sensor was on the fritz - the plenum was in place and who cares with a map sensor on the plenum if there is a leak before the throttle body? I swapped the map2 again. No difference. Exactly the same lack of vacuum. It turns out that the dump valve was not sealing properly to the reference line (I've never seen a dump valve that could leak there!) so I removed the dump valve line from the plenum. The vacuum is now just about -20inhg and the car is running much better. I have solved the dump valve issue by dismantling, cleaning and regreasing the O rings. It is now running properly with the dump valve plumbed in again. It has taken a while to start idling smoothly, which I guess is the car de-coking itself.

So it seems that the dump valve leaking in a strange way was destroying the vacuum in the plenum - the MAP2 was then over fuelling and the car was coking up. I'm also wondering if this was the cause of the "misfire" that the car was displaying occasionally whilst cruising last year.

Thanks again to everyone for their suggestions.

[TAR]

Yay, well done, glad you got it sorted.

:happy:

[adaxo]

Result!!!/GJ

[MarkSanne]

Great you've found it!
It's funny (well not really actually) that this problem (again) would've been solved with a vacuum/boost line check. It's crazy how often errors occur when there is something wrong with the air/vacuum lines/piping. Ever since I came to realize how important this is, it is now the very first thing that I check when I have ANY problem (or suspicion of a problem) with my VR2 or other guys here who come to my place. Easy 5 minute check and it saves so much time at problem-solving. Now don't get me wrong, I don't mean in this in a 'told-you-so' (cause I didn't suggest anything upfront :)) but it's just an observation and this is exactly why I started my 'grand-problem-solving-topic' with this air/boost/vacuum check :)

Again: good you found it, saving one more VR4 from being slaughtered!

[Nick Mann]

I'm not sure how I would have found it with a boost leak check. The reference/control line to the dump valve was fine, the dump valve was leaking internally. If a boost leak check had been done then the air leaking past the dump valve piston would have ended up back in the intake pipe work. (Or it may have been that it wouldn't have leaked at all under positive pressure, I have no idea.) Usually I locate boost leaks with soapy water or a good ear - the soapy water wouldn't have shown anything and I don't know if I would have heard air travelling through a recirc dump valve.

One thought is also that even if I had found it, the car wouldn't have started until I changed the plugs as well. It arrived to me as a car that would not start, so the coking up had already occurred.

The other thought is that if the car was running just a stock ECU and no MAP2, how much of a problem would this have been? The air escaping through the recirc dump valve would have still been in the MAF metered area......

Either way round, a boost leak test is never a bad idea! I don't take much convincing of that. ;)

[Davezj]

well done nick perseverance pays off.

i should have realised this as well that it could be an vac leak issue and not just an electrical one. It is a good reminder how important is it to monitor the vacuum status of the engine and not just the boost.
a good thread for anyone doing any diagnostic work on there car as a guide of stuff to check.

a boost/vac gauge on the dash connected to the plenum would have shown the issue.

[MarkSanne]

Yeah it seems like this one was a combination of things. And I didn't think of re-circ, as I'm so used to vent-to-air BOV's (most of the guys here have VTA). Good points Nick (as can be expected from one of the much, much experienced VR4 guys around)!

[Nick Mann]

Hindsight is a wonderful thing, isn't it?! :P I've got an alarm system to try and reinstall, a cam cover to put back on, etc etc. /pan

In this particular case, the dash gauge wouldn't have made the issue much quicker to find - the MAP2 ECU displays boost on the laptop, so it was only the length of time required to get the laptop that delayed the questions. During cranking I didn't see any issue with the vacuum. I'm not even sure I should have? Even when the MAP2 had displayed a low vacuum, I questioned the MAP2 rather than the vacuum. I've simply never seen a BOV leak like that before, so I couldn't imagine what the issue was when there are very few places for the plenum to leak. Potentially if I didn't have a second MAP2 I might have decided that the MAP2 was at fault and ripped that out too.

[hbkuk1]

Nice! Well done :)

[Trotty]

Glad u got it sorted. I was pretty sure with all the craking and wet fuel the plugs would have been pretty average. At least got it started to find the real problem.

[carfanatic]

Having witnessed a few non starters in my time, sometimes a tow works when nothing else does, assuming its a manual. CJ

[Nick Mann]

It's running now thanks, Colin. I have now removed the dump valve completely as it kept causing the car to "hiccup" whilst driving around. Now just the tiny misfire on WOT to investigate and the car may be saleable.