MembersPage/MarcellGal/PowerAndTraction/Alternator (2007-10-12 11:30:08)

Subpage of MembersPage/MarcellGal/PowerAndTraction

This engine has a 110A Bosch alternator, driven by the outer V-belt from the 3-belt crankshaft pulley.

Alternator connections

I guess the charger-control is integrated inside the alternator in these audis.

MembersPage/PaulF - Most smart alternators that I have dealt with fail safe to an output voltage of around 13.8V. Usually it is to allow the charging voltage to be lowered at idle to help pullaway. It is also used for feedback to the ECM as to the load on the alternator for torque adjustment.

There is some charging when engine running, 13.7V

It charges somewhat .. maybe suboptimal, but it works.

I think I found a wire with ring connector that could be alternator excitation. I heard that this is connected to VBATT+ through a small indicator lightbulb. I heard that this indicator (on the dash?)

MembersPage/ZoltanAlmasi - The ring connector wire come from the instrument cluster's charge indicator bulb. The bulb get 12V+ from the ignition key. This circuit needs for the alternator to proper inducing. Without bulb or when this circuit is open, the alternator don't induce up or just very slow.


Alternator belt

I cannot seem to get a proper drive to the alternator.

WTF is happening ?

So after I lost the alternator belt (snapped or jumped off ? dunno) TWICE: Than in a rush to find parking place downtown, I stuck a concrete and killed the bottom oil-cap and the right-ear of the charge-cooler broke.

It was a really bad combination:

Oil-change was about time anyway, and I have a spare bottom-oil-cap to install, and we ordered 2 gaskets already. But the broken-off charge-cooler ears are no fun, removing the cooler and welding Al has its costs.


Ghost current

When ignition is off, current draw from the alternator is 0.37A. This means I have to charge with a battery charger if parking more than 2..3 days so I can start the car after.

What draws the ghost current ?

I guess I have to pull fuses 1 by 1 to find out where the ghost current goes. Isn't there a hierarchy so I can test a larger number of subsystems with less steps ? If fuses are arranged in a tree topology, this would be nice. But I guess there is the main fuse, and 20+ "neighbor" fuses under it.