Engine died during extended deceleration

[Eiron]

This happened twice on the same day, once in the early morning (08:00) & once in the early evening (17:00), under similar conditions & locations both times. 1990 Saab 900, 2.0L I4, fuel injection, non-turbo, 5 sp, 161k miles.

The first time was after driving for about two hours. There was an extended downhill section on the West (shaded) side of the mountain of about four to six miles. Temp was probably in the mid 20s. During this time I had the car in 4th gear & used engine braking to keep the speed around 50-55 mph. As we approached the bottom of the hill, I saw that the tach was showing 0 rpms. Trying to restart the car, I got only strong cranking. I pulled off to the side & checked the engine to make sure nothing had come off (all belts & hoses looked good). After about 3-4 min it started up & ran fine for the remaining 45 min of the trip.

The second time was on the way home, after driving for about 20 min. Now we were heading East (in the shade again) into the same valley as the first occurrence earlier in the day. Temp was probably again in the mid 20s. This was also an extended downhill, but the slope was more gradual on this side, so I had it in 5th gear to allow engine braking at a similar 50-60 mph. Again, near the bottom of the hill, the tach had dropped to 0 rpms. I pulled over & tried to start the car, but again got only strong cranking. We waited for about 3 min & the car started fine & ran w/o problems for the remaining 2+ hrs of the trip home. There were many more extended downhills on the way back, but no more problems.

The day before, I had pulled the plugs (NGK coppers) & regapped them all to the lower end of spec. I also checked all the ignition wires (1 coil & 4 plug wires) for proper resistance & they were all in spec & slightly lower than the new wires I recently purchased. I ran out of time & didn't get to check the rotor's resistance. There was no corrosion on the coil terminals. Since all of the old wires showed slightly lower resistance, I left them in the car instead of installing the new wires. I know these old wires have at least 4yrs/50k miles on them. (Can this be part of the problem?)

My initial thought was fuel pumps (it's got a main pump & a pre-pump), but I would've expected more problems on the extended uphill sections than on the extended downhill sections. So what else? Throttle body icing under perfect conditions? (Descending into the same valley under similar engine & climatic conditions? Air filter is a K&N drop-in.) Ignition problems of some kind? Automatic Idle Control problems? Throttle Body gunk? Other ... ?? And why did it start fine after a few minutes & run perfectly after both incidents?

Again, the car drove fine for 2 hrs (mostly uphill) before having a problem, & then again for 2 hrs (mostly downhill) after the second incident. There were no weird noises or actions coming from the car at any time before, during, or after the two problems.

Thanks for any suggestions.

[ February 19, 2003, 11:15 AM: Message edited by: Greg Netzner ]

 

[White 03]

Is the car a 5 speed stick or an automatic? On my truck if the engine turns over but the tach does not register it sometimes indicates a bad CPS (crankshaft position sensor).

If the hill is steep and the fuel tank is very low my Honda Civic will starve for fuel but then starts when it levels out. The fuel pick up must be in the back of the tank. I can hear the pump running when it has no fuel.

When the Honda is out of gas the engine continues to turn, it has a stick shift, so you hardly know it is not "running". I think the tach registers in this case though.

 

[TheLoneRanger]

Greg, I had the same thing happen to me with my carbed Bronco. After 5 miles of downhill driving in the mountains, the Bronco stalled and then re-started. At the time I suspected carb icing.

Was the tach showing 0 rpms with the engine still turning in gear? If so, it's an ignition problem. I would suspect the ignition coil or the ignition module. The problem will be hard to find, unless you can re-create the stalling and have a multimeter handy to check for voltage at the coil, module switching, etc.

If you were in neutral when the engine quit, then I would suspect the IAC (idle air control) or throttle body icing up. You could clean them, and maybe spray some WD-40 in the IAC.....

[ February 19, 2003, 09:27 PM: Message edited by: TheLoneRanger ]

 

[BOBISTHEOILGUY]

First thing that came to my mind was the icing up.

Reasons...

If wires/coils were breaking down, they certainly would do it more frequently and not under a no load condition such as down hill only. It would become more evident up hill as it would have to provide good ignition to burn the fuel for power while climbing almost seeming like a bad fuel pump unless backfiring became evident.

On each occassion you were traveling down a hill, on a cool area, both times using the engine to break, therefore allowing a lot of air flow into the intake and across the engine, and no gas thus the cyl's were not burning as hot as would be while climbing. Also while climbing, your not running the same speed up a hill normally as you would down hill, so taking more power to climb, it creates more heat in the engine compartment and not allowing as much air flow across the engine, allowing more heat off the exhaust to be fed back to the intake keeping it from icing. On the down hill side, you have more air flow across the engine and manifolds, less heat being produced because no throttle being used, so not enough heat being produced off the exhaust to be fed back to the intake allowing for the ice to form due to the ventury effect of the air system design.

One thing to keep in mind is during the rest of the driving, did you engine brake down the other hills like those times or do it as long as the two times you described earlier. I suspect not. That would account for the problem you have incurred. Also check to see if you have the air duct from your exhaust system attacted to the intake as it is what provides heat during this type of situation.

Thats where we get "carb heat" from on air planes to keep from icing up while flying.

 

[Eiron]

Bob,
Thanks for sharing the additional knowledge. This is the first time anything like this has happened to me, so I have no experience in how to deal with it.

My throttle body is warmed by coolant, not exhaust. Does this mean the warming temps are (potentially) much lower? Both times, you could watch the coolant temp gauge drop from normal (about the middle of the gauge) to cold (about 1/4 up the gauge) before the engine cut out.

The engine braking runs down the other passes were just as prolonged, but the fact that it happened both times in the same valley makes me wonder if the "micro-climate" in that area was simply better for causing problems?

Any other thoughts or observations would be greatly appreciated.

 

[White 03]

I think the tach going to zero is the key to finding out what is happening. I would bet it is connected to the ignition system like TheLoneRanger said.

[ February 20, 2003, 10:23 PM: Message edited by: White 03 ]

 

[TheLoneRanger]

Greg, my first thought, like Bob, was TB icing. But since the tach dies with the engine still turning, the problem is in the ignition system.

Bad sparkplugs or plug wires will not cause this. You're getting an open circuit in the ignition system, somewhere before the distributor. Since the tach takes it's signal off of either the coil or ignition module, it's one of those going bad, (or possiblly the ignition switch).

I can't correlate that with the long coasting, though. Let me think about this......

 

[TheLoneRanger]

By any chance, is that valley frequented by UFO's?

 

[Eiron]

Thanks guys, I appreciate the help.

White 03,
It's a manual. In this case, the car was running, in gear, at 50-60 mph when the tach dropped to 0. The tank was about 3/4 full when it happened the first time, & about 2/3 full when it happened the second time. In both cases, the grade was leveling out at the time of the occurrence. There were no obvious changes in sounds or noises at the time.

TheLoneRanger,
Would old ignition wires create a similar problem to what would be created by the coil or ignition module? Any idea what happens to the system to require a short wait before restarting? I know that the coil & wires are at least 4yrs/50k miles old, but I suspect they're actually all original (14yrs/161k miles). The resistance on all five wires checked out within spec (when cold), but that doesn't really tell me the condition of the insulation or what's going on during operation, does it?

 

[Eiron]

Good point! I'd forgotten that all your electronics die when a UFO scans your vehicle. I don't remember seeing any bright lights, but I probably wouldn't, would I? And I didn't even think to check my watch!! I was cutting across the valley, heading towards Vail. I've driven across the valley before, but never at this time of year. Maybe the other end of the valley (towards Breckenridge) has some "out of this world" ski package deals going on??

 

[TheLoneRanger]

"Out of this world", he he he. No "lost time"? Then we can discount the alien theory.

The only thing different about long coasting that I can think of is cylinder pressure. With the throttle closed, cylinder pressures would be low. The low pressures would allow for a lower resisted spark from the plugs. This would create a bigger spark and more current flow through the coil, thereby creating more coil heat. The heat could be causing either a momentary open, or shorting out the coil windings. After the coil cools a bit, then the car re-starts.

Sounds far-fetched? Maybe. If it happens again, try touching both the coil and the ignition module. Chances are the hottest part will be the faulty one.....

 

[Eiron]

Thanks, I'll concentrate on those two items. (I don't know where my ignition module is located, but I have a shop manual!) The coil is bolted to the cross brace above the radiator, which gives it a nice, short path to the distributor. I'll have to look, but the hood & insulation may be protecting it from cooling air circulation. I'll also look thru the shop manual to see if they list any trouble-shooting procedures for my coil & module.

And I'll keep an eye peeled for Mulder & Scully.