Diagnosing "engine no start" problem
- Thread starter geogan
- Start date
seanie
Well-known member
If you have no spark it's unlikely you have problem with injectors also. When using spark tester I assume it's an extension lead that goes between plug and coil?
If so what type of coil on plug is it? Does the coil not need to be earthed to head to work? Some do some don't depends on type of coil.
If it does redo spark test with coil earthed to something
If so what type of coil on plug is it? Does the coil not need to be earthed to head to work? Some do some don't depends on type of coil.
If it does redo spark test with coil earthed to something
seanie
Well-known member
Coolant temp sensor fault generally would cause cold start problems due to incorrect fueling on start up. I have no experience with that issue on a subaru but can't imagine it would cause a no spark condition unless it shares its electrical circuit with some other sensor related to the ignition system. It's possible, I don't have wiring diagram to hand to check.
Also possible it is coolant sensor and not checking spark correctly.
If you scope cam and crank sensor together are the tdc pickups correlating?
The odds of an unrelated fuel injector fault and no spark condition happening together are highly unlikely unless you have a power supply/earth related problem or tdc signal problem or ecu
Also possible it is coolant sensor and not checking spark correctly.
If you scope cam and crank sensor together are the tdc pickups correlating?
The odds of an unrelated fuel injector fault and no spark condition happening together are highly unlikely unless you have a power supply/earth related problem or tdc signal problem or ecu
seanie
Well-known member
I have never had this issue on a subaru so anything I am spouting is generalised and not subaru specific.
I think you are looking too deep into this
There isn't a lot going on, verify
Timing
Check spark
Check fuel
Compression
You say 3 pin coil on plug, I'm assuming igniter is internal to COP, assuming this if you have 12v and ground and what I would imagine is 5v digital signal on 3rd wire, that 3v if read on multimeter is 60% duty cycle so coil has everything it needs to fire spark again assuming it doesn't need to be earthed by mounting bolt (check that). I can't see 4 coils failing together.
Quick way to check spark assuming timing and compression OK , unplug 4 injectors and give a shot of c?rb cleaner into intake and see if it kicks.
I think you are looking too deep into this
There isn't a lot going on, verify
Timing
Check spark
Check fuel
Compression
You say 3 pin coil on plug, I'm assuming igniter is internal to COP, assuming this if you have 12v and ground and what I would imagine is 5v digital signal on 3rd wire, that 3v if read on multimeter is 60% duty cycle so coil has everything it needs to fire spark again assuming it doesn't need to be earthed by mounting bolt (check that). I can't see 4 coils failing together.
Quick way to check spark assuming timing and compression OK , unplug 4 injectors and give a shot of c?rb cleaner into intake and see if it kicks.
drop it to someone who knows what there doing would work out cheaper than lashing parts at it,have you asked a mechanic to look at it ?,my uneducated guess is alarm/immobiliser
OP
OP
geogan
Petrolheads
Yes hopefully nothing to do with injectors.
The tester I used was a spark strength tester. It replaces the actual spark plug and plugs directly into coil-pack the same as the spark plug would. On other end of tester is an earth clamp which is just grounded on engine ground anywhere. It allows adjustment of gap to test strength and according to instructions, a coilpack should be able to spark the longest length from 2-3cm (compared to other weaker ignition systems).
But there was not a single spark when I originally tested, this is because the coils are not getting the signal from ECM at all. Should be seeing 12v pulses from ECM to trigger the coils. But apparently the ECM has decided to not send the signals - and this is the black box problem, because it won't tell me why not! If only the ECM would say "i'm not sending spark signal because I expect to see X,Y,Z and I'm not seeing it, so no spark signal!"
I can directly connect oscilloscope to signal pin on coil connector or to ignition control signal on ECM output and see same thing. I might re try these tests again today since I haven't done them for a while with all the other works/fixes/tests I had to do.
OP
OP
geogan
Petrolheads
If that's the water temp sensor on bottom of water pump behind main timing belt then that's going to be a pain to get to - I had the whole thing apart and already put it all back together again (belts/covers/crank pulley/radiator/pipes) so not really in mood to do that again. I presume no way to get to that without taking everything out again??
OP
OP
geogan
Petrolheads
Unfortunately I only have a cheap scope that does one signal only at a time. But the crank and cam sensor waveforms looked correct from what I can see. And as I said, the RPM was increasing in OBDII monitor, which comes from crank sensor output.
That's why I was trying to get RomRaider Logger to work as its logging is coming straight from what ECM sees. But it appears that "free" RomRaider is only free as long as you also buy the 200 Euro Tactrix cable to make it work!
Also that's why I took the engine apart to check the timings on the actual timing belt... and all marks appear to line up correctly on cams when crank sprocket mark is at top.
Last edited:
OP
OP
geogan
Petrolheads
Well considering how many hours I have spent on it, that would have been a massive charge in labour hours at any garage...
Now obviously it is easier for them if they have a few parts to swap out and test, but I don't know if they keep spare random sensors lying around for every model of car! Only main Subaru dealer might have those, and they charge through the nose for that service.
I have avoided the obvious and quick testing of "lashing out on parts" up to now... the only parts I got (apart from diagnostic tools and equipment which I don't mind buying as they are useful always) was service parts that needed changing anyway (spark plugs/fuel filter/air filter/brake pads). The starter was probably on last legs anyway... old one was not in great condition, been slow cranking from cold for a few months. The MAP sensor I bought because it is right next to IACV which was what i fixed last time this no crank problem happened and immediately fixed problem last time, but I got that as a future improvement/upgrade. The crank/cam sensors I got just now as I am sick of people saying "its the crank sensor - change crank sensor", even though diagnosis and logic says it IS working. The cam sensor was only ?20 to buy.
I thought of alarm/immobilizer yes, but it appears to not be that... I mean I know what it does when immobilized (no engine start or power at all anywhere)... when I enter code as normal, I hear relay clicking on and all systems get power, and car cranks as normal. Before code entry, there is no power anywhere. It's aftermarket expensive Clifford G5 system I have for years.
Also the car is not running at all so "dropping into garage" is not a cheap or easy option since it requires a tow truck to be hired.
So yes I WILL at last resort throw it into a garage if I exhaust all reasonable options to fix it myself... but not yet! ;-)
Unless there is any Subaru experts here that can come out and look at it in place? I would pay them if they fixed/diagnosed the problem quick!
Last edited:
seanie
Well-known member
You certain about that?
Not sure about subaru but any 3 pin coil packs I have worked on have 5v digital pulse signal.
You said you measured 3v on signal wire when cranking, if that was with multimeter it only shows average voltage for signal which 3v would be a legit signal if it is indeed a 5v supply.
You may be correct with 12v just wondering if you are assuming that or did you have access to the correct info
OP
OP
geogan
Petrolheads
Check the wiring on the injectors as I know it was a problem on a new age before and caused an intermittent Start problem
OK I will look into this if the crank/cam replacements don't fix problem
OP
OP
geogan
Petrolheads
Timing: Checked belt and marks lined up still
Check Spark: No spark. no signal from ECM. New plugs. Coils appear functional (resistance/voltages).
Check fuel: According to Tomei FPR guage in engine, there is fuel pressure as soon as key turns and Walbro pump primes. On every crank, can hear Walbro re-prime pressure. No idea about injectors or condition, but can't see four going together instantly.
Compression: didn't check this now, checked a few years ago and was normal, but again, all four at once makes no sense.
OP
OP
geogan
Petrolheads
No, I said there was a 3V peak-to-peak signal coming from CRANK sensor... which I think is normal (it is analog sine wave signal)...
The ignition pulse signal coming from ECM to coilpacks is actually supposed to be 13/14v which is full battery voltage + alternator I presume.
Yes I have factory service manual electrical details, which all match up to my car.
OP
OP
geogan
Petrolheads
" it is cam/crank sensor problem" - well I hope to hell that is what it is!
I got Clifford fitted myself a few years ago. As far as I know the original Subaru alarm system is replaced completely with Cliffords own immobilizer as part of alarm (it has Blackjax immobilizer).
OP
OP
geogan
Petrolheads
A lot of garages I went to in past seem to treat Subarus as some magical race car that they are afraid to touch - have got "we haven't got the right gear for that - you need to bring it to Subaru dealer" type thing...