I just found this group and I don't see any currently posted info on this problem.
My Subaru is a 1993 Legacy (wagon, 4wd, a/c) with 151,000 miles. When it is running it is a super little car. Great handling, decent mileage, very comfortable.
I bought it used about 4 years ago. From the first it always started instantly, hot or cold. One morning in Fall of 2004 it got cranky. When I cranked it, it would fire, but not stay running. Repeating this process from 6 to 18 times would get it going. [This was in a garage with a temperature never below 60F.) After the first start, it always started just fine the rest of the day.
I took it to local service station. I dropped it off at night so they could see the "first start" phenomenon first hand the next morning. They couldn't find anything wrong.
Our nearest Subaru dealer is not all that close, so dropping the car off the "night before" is not attractive.
We are a Subaru family; my wife has a '99 SUS. I'm retired, so I took my wife's car up to the dealer for it's annual service. While I was there I asked them about my problem. If they had suggested I bring the car in and leave it overnight, I would have done so. Instead he said to me, "Oh the older ones do that." That wasn't much of an answer, but that's all the answer I got.
The problem came on quite suddenly; one day it was fine, the next day it was not.
My guess is that there is a temperature sensor somewhere that is telling the central computer that the engine is hot when it is not. That would set the mixture too lean for a good start. My cranking and short run bursts may warm things upo enough that it will start.
Outside, with the temperature around 30F, requires more cranking than the battery can handle, but, with an outside power source, it will eventually start. I should point out that, at first, the cranking is quite vigorous: Both the starter and the battery are doing a great job.
One thing I noticed today is that the electric cooling fans come on when I turn the ignition to the "run" position, even though the engine is at ambient temperature, about 60F in the garage. If that is controlled by the same sensor as the mixture -- and why not? -- that would support the notion of a failed temperature sensor.
Have any of you seen this problem and have you any idea how to fix it?
Any help will be very much appreciated!!
Thanks
Ken