The Prince of Darkness Strikes Again!

[JDNSW]

Well not quite, in fact this time I can't blame it on Joseph Lucas.

Yesterday it became apparent that the alternator was not charging. So I brought it into the shed, and started off by checking that voltages were present at the alternator. (I had an idea what was wrong - last time it was a broken brush spring, and the time before the bearings had gone).

However, I soon found the problem - the wire from the alternator to the starter relay had melted. This was one I had installed when converting to alternator about fifteen years ago, and obviously it had not been adequately protected (it seems to have chafed on the engine, but hard to say exactly as the destruction was almost complete). It also became apparent that the diodes in the alternator had failed (when tested using my lathe to turn it), either due to the current to melt the (fairly heavy) wire or the excessive voltage that was generated when the battery was no longer connected (external regulator alternator, so it was still being excited).

So I spent most of today installing the alternator I had on hand (out of a S3), which needed some modifications to the installation to fit. Worked when finally installed and cleaned up just before dark tonight.

John

[Aaron IIA]

It also became apparent that the diodes in the alternator had failed (when tested using my lathe to turn it), either due to the current to melt the (fairly heavy) wire or the excessive voltage that was generated when the battery was no longer connected (external regulator alternator, so it was still being excited).

John

I doubt the high current would have broken the diodes. The current output of an alternator is limited by the size of the biscuit. The diodes are then chosen to be capable of handling this current.

Older alternators are known to have diodes that blow when the voltage becomes a little high. Their peak inverse voltage is not high by today's standards. If the output was disconnected, the alternator would have been getting full excitation current by virtue of the external regulator. At speeds above idle, very high (one hundred or more volts, increasing with speed) voltage would have been present. This would have blown the diodes.

Aaron.

[JDNSW]

I doubt the high current would have broken the diodes. The current output of an alternator is limited by the size of the biscuit. The diodes are then chosen to be capable of handling this current.

Older alternators are known to have diodes that blow when the voltage becomes a little high. Their peak inverse voltage is not high by today's standards. If the output was disconnected, the alternator would have been getting full excitation current by virtue of the external regulator. At speeds above idle, very high (one hundred or more volts, increasing with speed) voltage would have been present. This would have blown the diodes.

Aaron.

I'll agree with your analysis, that failure is most likely to have been due to voltage rather than current. Certainly the alternator would be classed as 'older' - probably from the 1960s or 1970s. I'm not sure that the voltage would be a hundred just above idle, as the rectified output is only just above battery voltage at idle (remember a 2a idles at 500rpm), but certainly by the time it reached top engine speed before changing gears it would.

Just one question - what is a 'biscuit' in an alternator? I don't know the term. If this refers to the heat sink holding the diodes, this does not actually limit the current from an alternator - it specifies the current that can be delivered for an allowable temperature rise beyond which the diodes will fail, but does not actually limit the current.

John

[Aaron IIA]

The portion of the stator that is visible from the outside of the alternator is known as the "biscuit". This is the piece of iron that the coils are wound through, and the two aluminium ends clamp onto. The wider this is, the more magnetic field that it camptures, hence more current. Older units of about 30 amps are about 5mm thick, while newer units of 100 amps are about 25mm thick.

Aaron.