Hi Guys,
I've been thinking about this problem and my guess is that both your problems are down to compression or fuel atomisation issues. From reading the bosch common rail yellow back books, the engines only need the preheated fuel at minus temperatures and rely on compression and fuel atomisation to start the rest of the time.
Arkk, I see yours is an ambulance, high mileage? Have you done a compression test? If it is down you could get a head skim and a thinner gasket to bring it up to spec.
Hi Torq, No Ambo 578 has only done 175000km. Compressions ok. Ambient temp shouldn't be an issue, just come out of summer with Temps in the low 30's cel. Have just removed, cleaned,tightened every connector plug I could find in eng bay. Found plug on top of pump was not seated fully (3rd
piston deact ?), but made no difference. Used cold start Friday am to start it, had it running on & off for 90 min while testing various sensors. 5 hours later I started it numerous times and couldn't get it to fault. Saturday am it started first go, 3 hours later it struggled to start even using starting fluid, to the extent that I all but flattened a battery. Fuel pressure is definitely low when the fault occurs, but I can't determine whether fuel pressure is low due to cranking speed, or vice versa- (chicken or egg). Measured crank speed is ok, fuel pressure varies 60 -110 bar, leaps to 300 on start up. These systems have pilot injection, and I think the ECU has to detect adequate fuel pressure and crank speed before switching main injection on- perhaps someone out there could advise me on this.
To recap: mechanically everything tests ok; compressions fine, new HP pump & injectors, etc.
Electrically all injection related components have been replaced, with new or 2nd hand(off vehicles without faults- if that is possible for Fiats), some more than once, ck, cam, pressure sensors, press reg, etc. ECU has been substituted. On occasions faults log on Fiat Examiner scanner, but more often not. The ECU has to detect a fault for a certain no. of cycles before it logs a fault, so if a sensor was playing up it certainly should log a fault in the time it's taken us to get it started. The best theory anybody has given me yet came from a hillbilly looking dude from a country garage. I went thru its history when he was in my workshop , he tried to start it, stood back and said dryly "It's probably got a ghost in it, maybe someone died in there". Could explain a lot.
Thanks for replying Torq, any more ideas gladly received, even if we've tried them before, it could be some procedure we've applied wrong or overlooked. Cheers, Ray