On Wed, Jan 23, 2002 at 03:57:55PM -0600, Steve Budde wrote:
>
> OK Folks, last time I was looking for a logic check. Now I'm baffled,
> stumped, befuddled and stymied. My truck still won't start, I have
> installed a new starter, battery cables, and battery. Yes, I know the
> battery is good as it starts my Impala with the upgraded engine. All
> contacts are clean and lubed with dialectic grease. The last time it was
> running, it was running just fine so I'm not suspecting any mechanical
> difficulties. Where do I start??
>
As Robert Pirsig says in "Zen and the Art of Motorcycle Maintenance"
(paraphrased): "now it's time to get serious and use the scientific
method".
In case you have forgotten, this means that youy come up with hypotheses
and then tests to test those hypotheses. And write them down, both
the hypothesees and the tests and their results.
Also, I would recommend a technique that I developed working on bikes and
honed while debugging software: devise tests that split up the problem
into smaller and smaller pieces until you have discovered the part
which is broken.
An example in your case would have been to swap the known good starter from
the Impala to the truck. If the truck starts with it, then you know the problem
is in the truck starter. If it doesn't, then you know that the starter
isn't the problem, something else is.
Then you try the solenoid, etc.
When I do this I try to come up with tests that will split the
problem space in half, more or less. Then I know which half of the system
has the problem, and I can split that in half again. Once you have done
that a couple times, you have a much smaller problem space and can start
testing indiviual components, if you haven't already figured it out.
The examples above don't work that way because I can't think of a way to split
this particular problem in half... doesn't mean that you can't.
But the key to this technique is NOT to say "maybe its the ignition switch"
and test or replace that... it's to think of tests that will help narrow
down where the problem is. Once you have done that, then you worry
about what the problem might be. Making random guesses can often
get lucky (or be based on prior experience), but when that fails, it's time
to get serious about finding out where the problem is instead of
making stabs at it.
Eric
/// unsubscribe/change address requests to majordomo@autox.team.net or try
/// http://www.team.net/mailman/listinfo
/// Archives at http://www.team.net/archive/shop-talk
|