Aeroelectric's figure Z-7 is the first time I have seen an explanation of how
to handle over-voltage failure in the Rotax 912's electrical system and
thereby protect one's expensive radios. The fuselink idea is excellent also
in that it solves an awkward problem in an elegant way.
However, there are a number of items in diagram Z-7 which raised questions in
this neophyte Europa electrician:
1. How is an under-voltage alternator failure diagnosed? Are two alternator
fail lights needed now, one for over voltage and another for under voltage?
2. When the crowbar activates and trips the ALT breaker, wouldn't the SCR
turn off when the current falls, allowing a spike from the relay's collapsing
field to affect terminal "C" of the regulator? Is "C" internally protected by
a diode? Is there any info available on the circuitry inside the regulator so
one could reason through how to deal with it?
3. If the crowbar pops the breaker, terminal "C" is pulled down by the relay's
coil resistance. What does the output of the regulator do under this
condition? According to the Rotax manual, the output will fall if "C" is
floating, but no info there on what happens if "C" is pulled down. Could
holding "C" low while the output is disconnected cause a high output voltage
such that the regulator is damaged after a period of operation?
4. What is the current draw of the OV relay's coil? Could the circuit be re-
arranged to disconnect the alternator by activating the relay when the
voltage is too high, thus reducing current normal drain? Could the relay coil
then be powered from the (over voltage) regulator rather than the buss to
minimize battery drain in the failed condition? Must the drive to terminal
"C" also be removed (to reduce output voltage) by tripping the breaker? Is
any of this a good idea?
5. Why isn't the wire which activates the master relay's coil protected?
Seems like a shorted diode or master relay coil could cause a problem - low
probability, but possible.
6. Why is the wire to the other half of the master switch protected by both a
breaker and a fuselink?
7. Would a resistor/diode/LED across the ALT breaker act as an over-voltage
indicator? (Seems like the wiring would be easier since the breaker is in the
panel rather than on the far side of the firewall.) If one side of a
resistor/LED were tied to the buss, could diodes be used as an OR to drive it,
one diode from the other side of the breaker and another from "L", resulting
in one alternator fail indicator?
8. Is a sheet of aluminum needed under the regulator to increase the heat
sink area in fiberglass airplanes, thus improving reliability?
In my abysmal ignorance of the circuitry within the regulator, I assumed that
it contains a diode/SCR bridge plus regulator circuitry to adjust the SCR's
firing angle for voltage regulation. Attempting a FEMA of the failure modes
in this hypothetical circuit, and desiring a solution without an over-voltage
relay led to the following:
1. The regulator circuitry could fail such that the alternator output is low
or a diode/SCR could open resulting in low output. Diagnose via a lamp on "L",
or if the "L" drive should fail, a low voltage alarm from the voltmeter.
2. The regulator could fail such that the alternator voltage is high (high
conduction angle on the SCR's). Crowbar the ALT breaker connected to "C",
thus removing the power to the regulator circuitry - expect this to reduce the
output current to 0. The "C" terminal would float after the crowbar event or
be supplied with a few mils via the optional LED. Diagnose by an optional
resistor/diode/LED across the ALT breaker or if this isn't installed, via an
eventual low voltage alarm.
3. The diode/SCR bridge could fail in a shorted mode leading to a high AC
voltage on the output lead. This should open the fuselink in the wire
connecting the alternator to the system, eventually resulting in a low voltage
alarm from the voltmeter. What will "L" do here?
3.1 Should the fuselink to the system open due to a shorted diode/SCR, AC
could be applied to the 22000uf alternator output capacitor possibly causing
its catastrophic (smoke or a minor explosion) failure. Add a fusible link
between the capacitor and the alternator plus a reverse biased high current
diode across this capacitor -- the goal here is to prevent reverse voltage
across the capacitor and also to open the fusible link.
3.2 Should the capacitor/alternator fuselink open before the capacitor/system
fuselink then the capacitor remains safely connected to the system. Assume
that the resulting AC voltage does not harm the (failed) regulator?
4. Connect the alternator output to the system directly at the battery rather
than after the master relay. This allows use of a single pole master switch.
A shorted diode/SCR while the master is off would be handled by the fusible
links.
5. Optional crowbars could be inserted in the avionics circuit to protect the
expensive gadgets should all else fail.
The parts/weight tradeoff is minor: an extra fuselink and a high current diode
vs an over-voltage relay and a double pole master switch. Any insights on the
performance and safety tradeoffs between these approaches would be
appreciated, as would any information on the regulator's circuitry.
Regards, John A044
|