Strange no start problem

[C33]

Folks, maybe you can help me on this one. My '88 325 has been stood pretty much since last Easter in my parents farm yard and I'm try to recommission it.

With a freshly charged battery it will fire, fast idle, then settle down to a good steady idle. Switch off with the key and it restarts without a problem. If I happen to stall it and try to restart straight away it I have all sorts of trouble trying to get it to fire again, leave it half an hour come back to it and it usually goes straight away.
I've tried:
*Swapping batteries, made sure they're recording good voltages before attempting to start (that said I've not tested voltage drop during cranking but the engine spins healthily enough)
*Checked AFM is ok
*Checked TPS is ok (manual gearbox so the switch)
*Not tested blue sensor although I've tried with the sensor disconnected (open circuit) and with the connector shorted with a length of wire to give the two extremes of resistance and no change was seen in the characteristics.
*I get a healthy spark (in fresh air)
*If I whip out a plug after a no-start its wet so the injectors are firing and fuel pump running - I should add it's got an LPG conversion and when it doesn't want to start it won't fire on gas "emergency start"either but obviously doesn't give damp plugs.
*The ICV purrs with the ignition on but I haven't checked resistance readings

Is there any thing that happens when stalling that would differ to switching the engine off on the key? Does the car just need a fresh set plugs and a good heat cycle or two thru it as its only moved around the yard for a while and I doubt it's got out of its warm up phase in a while or is it more sinister? When the engine runs its fine, it's just getting the bloody thing started!

As an aside the engine bay has the loom and relay fitted for the lambda sensor. If I was to weld a boss into the exhaust downpipe and fit a lambda sensor does a UK spec ECU (173) have the facility to trim fuelling in response to lambda? It's something I'm going to do in order to go to a closed loop LPG system but it would be nice to know if it would help running on petrol too.

Regards
Iain

[Grrrmachine]

Based on this, I'd still be looking at the blue temp sensor and the wiring related to it.

To start with, if the battery has been dead for a long time the ECU will have lost its settings; you need to fire it up, let it idle for ten minutes, then turn it off. After that, take it out for an aggressive drive.

If it stalls, hiccups, kangaroos or backfires during that time, then there could be another fault, at which point you need to test the blue temp sensor FROM THE ECU!

[Brianmoooore]

Have you tried disabling the petrol system (pulling out the fuel pump fuse, #11) and starting it on LPG only?
It will help if you 'prime' the LPG system first by applying 12 volts to one of the LPG solenoids (usually a blue wire) for several seconds before you start it.
This will eliminate problems with blue temp sensors, leaking petrol injectors, and all the other common maladies associated with the inferior fuel that you normally start on.
After removing fuse #11, there will still be residual pressure in the petrol system, so try the gas start more than once.
Your ECU may have lost its idle settings if the car has been left without a battery, so it may not idle properly, or at all, without some throttle input, until it has been driven a bit and re learnt them.
The age of the petrol in your tank could also possibly be an issue. I wasted an hour or so at the end of last year, trying to get a petrol engined 'strimmer' to rev properly - an issue that was eventually cured totally by replacing the fuel in its tank.

[C33]

Hi Brian, yes, I've tried pulling the fuel pump fuse it try to eliminate over fuelling from a leaking injector (common trick to clear flooding on Rover V8's as the hotwire ECU has no flood clear on the cranking/throttle settings). It's had a gallon of fresh petrol in it so it's not all old fuel.

Idle doesn't seem a problem once it's running, I tried disconnecting the TPS at one stage and the idle revs rose slightly - almost like it initiates a fast idle mode, don't know if that signifies anything.

Anyway, I'll get new plugs on order (are NGK BPR6ES still thought the best even with LPG?), I'll check the blue temp sensor, throw in some more fresh fuel and see where I get to.
Regards
Iain

[C33]

Evening folks, still having trouble with the car. Took it for its MoT on Saturday, failed on hand brake adjustment (15% efficient when it needed 18% apparently). Tried to take it for it's retest this morning and the bloody thing wouldn't fire.

During the MoT it passed emissions without a problem, but the tester did comment it smelt a bit rich and wasn't idling very steadily. On the way back the car felt a little hesitant on petrol, flicked it to gas for a short while and it was fine after that. I've tested the temperature sensor this evening at the ECU plug and that is bang on what I'd expect in this weather, about 5000ohms. In free air I can get about a half inch long spark before the electrickery decides its easier to flow up my arm. I've put the battery from our Mondeo (current daily driver) in the car to eliminate a duff battery with poor cranking characteristics. I've also tried swapping in a "mild" chip into the ECU, but don't have a spare ECU to swap that in incase there's a dry joint or something that is causing the ignition timing to stray.

Is there any way I can test the timing, as far as I remember there are no timing marks on the crank pully to fire a strobe gun onto.

Any ideas anyone? It's rather frustrating that it seems to work well enough to pass it's emissions but totally fails to go at other times.
Regards
Iain

BTW has anyone else noticed NGK have stopped putting removable ferrules on the end of their BPR6ES plugs??? While on the subject of plugs I contacted NGK regarding LPG and spark plugs and got this rather non committal response
"The later plug ZGR5A is catalogued for your vehicle age which is slightly more projected than the plug type used for the earlier vehicles (BPR6ES).

We do have a range of specialist LPG plugs but not unfortunately for the later extra projected style plug. You may be able to use the earlier standard projected style plug in your vehicle without any issues. "

[C33]

Folks, this is driving me to distraction. I've gone thru and replaced all the gaskets on the inlet side of the engine, checked the hoses for splits and leaks. Put everything back together fired up on petrol at the first time of asking. Result. Take it for a run to the end of the road and it dies on me, try cranking and nothing, run home put in a spare battery and it fires on gas.

I've gone thru the following with the back off the ECU plug and to my mind it all checks out.

Pins 2, 14, 19, 24 are continuous to earth.
Pin 3; fuel pump runs when pin 3 is grounded
Pin 4; 20ohms between 4 and 37
Pin 7; with engine running, multimeter to ground gets the following voltages
1.4v at idle, 2v at 2k, 2.2v at 2.5k
Pin 12; 5v ref (about 4.9v with multimeter to ground)
Pin 16 and 17; earthing these pin gets clicking injectors
Pin 18, permanent 12v
Pin 22, 21ohms between 2 and 37
Pin 26, continuity checked ok between AFM plug and ECU plug
Pin 27, 12v on "run" and "start"
Pin 36 switches main relay when earthed
Pin 37 12v when pin36 is earthed
Pins 40 and 41 soldered to the wire to pin 37 (now disconnected)
Pin 44 3.2kohms (air temp sensor in AFM)
Pin 45 350ohms (coolant temp sensor with engine warm)
Pin 52 Earthed at closed throttle
Pin 53 Earthed at towards WOT

I think the only thing not checked now is fuel pressure. Is there any easy way to check this???
Any ideas why the car wont run now?

Regards
Iain

[Grrrmachine]

Is this an intermittent problem, or a hot-start problem?

The trouble with it being intermittent is that there's no point doing tests unless it's at the time the car won't start. Still, since you're a willing victim, take a look at this:

http://www.e30zone.net/e30zonewiki/inde ... oubleshoot
It sounds like a loose connection or a failing rather than failed component, so finding the source of the problem will be tough.

[C33]

That flow chart leads me to check for air leaks.
Its a virtually constant no start on petrol from cold. If its cold and I fire it up on gas immediately switch off it will fire on petrol. Again if I let it warm up it will generally fire on petrol, however the car never seems happy driving on petrol but seems ok on gas, I've only run 100yds up and down our cul-de-sac as its not got tax or test on it until I can be confident it will run to the MoT station.

Is there a minimum voltage the injectors need to see in order to open fully?

Regards
Iain

[C33]

Last time I checked they were wet, however today I've just switched it to start on gas and it's grudgingly started after a couple of revolutions.
Regards
Iain

[C33]

Hi,
C191 is good, I know thats a favourite failure point so that's been separated, checked, cleaned and had Deoxit liberally applied, infact yesterday when I did the gaskets every electrical connection got checked, cleaned and Deoxit'd (my brother uses it all the time on his kit out in the North Sea so just happens to acquire spare cans)

I'll check the coil / coil pin 15 next time it decides not to start.

What should the resistances across the coil be? I've a spare so I can check they're both equally good or bad.

Got a fuel pressure guage connected this morning, reading bob on 3bar, dropping to about 2.5bar when I suck for all I'm worth on the vacuum pipe. Took about a minute for pressure to drop to about a bar after I stopped the fuel pump. Also tried crimping the return pipe and running the fuel pump, it held pressure (circa 8bar IIRC) when I stopped the fuel pump which I think discounts a leaky injector?

Again what resistance should the leads give? I've several sets spare so again I can check they're all as good/bad as eachother.

The other things I've done is I've rechecked the plug gaps this morning and it seems to be behaving a bit better, firing on petrol gone for a run up the road and back. This was using the 164 ECU, it seemed happier on that than the 173. Also I've done a compression test, all around 8.5/9bar apart from #5 that historically has always been a little lower than the rest.

Regards
Iain

Regards
Iain

[Grrrmachine]

All the primary engine components are in the Wiki, so that you can check their data for testing:

http://www.e30zone.net/e30zonewiki/inde ... ition#Coil

(if something's missing, let me know).

[C33]

Thanks, primary coil circuit 1ohm, secondary coil circuit 7400ohms
King lead 1980ohms
#1 6000ohms
#2 5830ohms
#3 5930ohms
#4 5880ohms
#5 5970ohms
#6 5860ohms

Coil not to spec according to the wiki, how much tolerance is there on the wiki values?
Regards
Iain

[Grrrmachine]

I doubt there would be 50% tolerance; if the secondary coil is supposed to be 5k and you're getting 7.5k, I'd think about fitting that second coil you have. All the other numbers check out.

[C33]

Evening folks, I'm bordering the point of insanity now.

I've swapped out my modified injector loom (that had a feed off to a relay to cut the injectors on LPG) and swapped in a standard loom, LPG is now disabled apart for Heath Robinson jury rigging.

I've put in (although it doesn't properly fit the area) a known good battery out of my Jeep Cherokee to give loads of cranking current, battery voltage is now about 11.5v during cranking

I've swapped coils but also installed one of those inline spark testers between #1 HT Lead and plug, and get definitely get a spark.

After cranking I've whipped a plug out and its wet.

The only thing I haven't done is try the -164 ECU in place of the -173

Any other ideas before I start rocking back and forth in a dark corner???

Regards
Iain

[C33]

Folks, still no joy, it started fine when I got home after work last night, did the weekly food shop, tried again and nothing, it cranks, it struggles its thinking about going but like me with a hangover never quite gets going. Was thinking late on last night, the car has the 13button OBC fitted, could the relay module be in the process of dying and intermittently not giving a signal to pin 27? What effect does the ECU not getting 12v on pin 27? Does it not know to increase the injection pluse width (ie put the car on choke in old school terms) hence its wanting to start but the fueling is too lean for the engine to fire???

Are there any external obvious signs of a failing OBC relay module? Buzzing? Clicks when there should be clicks in certain conditions or vice versa?

Clutching at straws I know but...


Regards
Iain

[Brianmoooore]

This is a simple engine, and the fault will be a simple one! You're looking for something far too complicated, IMO.
I see you've had a fuel pressure gauge connected, but have you monitored the fuel pressure while you're trying to start it? It's not unusual for a failing pump to run happily on a full 12 volts, but to stop when running off the 8 or 9 volts it might see while cranking.
The result of this is that the fuel pressure drops right off just as the engine begins to catch, giving the symptom of almost starting and leaving wet plugs.

[C33]

Thanks Brian, I'll double check and report back!
Regards
Iain

[dialup_2001]

Does this car employ the ballast resistor on the coil? 12v when cranking and 9 when running? Try feeding the coil directly from the battery, before you start eating the foam from within the ripped drivers seat, splash out on a full 6 inline spark indicators if I were you just so you know your getting a uniform spark, maybe a air/fuel ratio meter on the lambda sensor to check mixture is correct. I know little about engines compared to many but looks like you need any input. I know the feeling.

[BenHar]

Does this car employ the ballast resistor on the coil?

No.

Ben

[C33]

Haven't touched the car since Monday evening, just gone out tonight after work, fires up repeatedly at the first time of asking with no issues at all. Took it for a quick run up and down the street, apart from slightly bogging down and needing a couple of revs when moving away it drove fine.

Regards
Iain

[Brianmoooore]

Haven't touched the car since Monday evening, just gone out tonight after work, fires up repeatedly at the first time of asking with no issues at all. Took it for a quick run up and down the street, apart from slightly bogging down and needing a couple of revs when moving away it drove fine.

Regards
Iain

Fits perfectly with my fuel pump theory.

[C33]

Smart arse!

I know different models have different part numbers on RealOEM but are all in tank e30 fuel pumps the same for all practical means? I've a spare VDO pump and sender assembly but I can't see any part numbers. It's either a spare I had for my old 318i 2dr or from a 325 tourer I broke 18months ago...can't remember which though.

Regards
Iain

[Brianmoooore]

The actual in tank pumps are interchangeable. You can use a complete two pipe assembly to replace a single pipe one, but not the other way around.

[C33]

Thanks Brian. The spare unit is a single pipe affair, but it's no real hassle swapping the pump into the existing unit, I think getting it out of the tank will be more difficult than anything else.

I'll book it back in for it's ticket and see if I make it this time!

Regards
Iain

[C33]

Afternoon folks, I've been running the car as my daily driver now since early April and put over 4k miles on it so I've got a bit more experience of this no start issue.

Starting on LPG is fine, on petrol it starts fine if stone cold, eg left over night, it also starts fine on petrol when hot, eg stopping after an hours drive for fuel, however when lukewarm for example the temperature gauge is in the blue section it is nigh on impossible to start on petrol.

Next thing I'm going to try is testing the blue sensor when lukewarm to see if I've a weird failure mode, but any other suggestions would be appreciated!

Regards
Iain

[Grrrmachine]

That sounds exactly like a Blue Temp sensor failure, or more specifically the wire feeding the signal to the ECU.