An intermittent open circuit on the alternator output terminal can
kill the diodes in an alternator. You can play hope and poke for a
long time and some day you may replace the correct item. or you can
run a simple test.

This is easy to test. For a 12 volt system apply a 12 ohm resistor
between ground and the alternator output terminal to get about a 1 amp
load. Turn on the master and measure with a DVM the voltage drop
between the battery + and the alternator terminal. Shake, thump and
move wires, amp meter circuit breakers, master relay etc and watch
for any voltage change. If the voltage changes more that a millivolt
or so find out what is loose and fix it. This method allows you to
measure EACH crimp connection, each wire section and each item in the
path that the 1 amp current is flowing in. V=IR If i= 1 amp then
V=R One millivolt is one milliohm. The method is called a Kelvin 4
wire ohm meter measurement. The DVM must be able to read millivolts.
Remember if the alternator outputs 60 amps then the voltage drop is 60
times what you measure with the one amp load. The voltage drop should
be less than 1 volt at full output.

John

On Wed, 15 Dec 2004 01:28:50 GMT, Chuck wrote:

Hi everyone,

This group seems especially knowledgable, I hope you don't mind a
question about a friend's plane.

Its a 78 Arrow and is being used by a FBO for training. For the last
month or so, they can't seem to keep the battery charged. Started a
month ago and the alternator was found not putting out anything.

Swapped it for a rebuilt alternator and two weeks later -- dead
battery again. Same thing -- no output.

Took the alternator back to the seller who tested it and confirmed
that it wan't putting out. Bought a brand new one and installed it.
Seemed to be putting out, but pretty low amperage except at high RPMs
(Arrow has a load meter, not charge/discharge meter) .

Exactly two flight hours later -- dead battery. Apparently, the
alternator isn't putting out again. And the A&P is at a loss. I
suspected the voltage regulator, but he says he's had it tested.

Has anyone else run into a problem like this before? Or can anyone
offer a suggestion on where to look for the root of this problem? I'd
like to help him get this fixed -- we're both working on Instrument
time and we're splitting the time between his Arrow and my Cherokee.

Thanks guys...



Chuck