Posted by Johny B Good on November 15, 2011, 2:18 pm
So... that confirms the generator in question as being a dynamo.
A modern alternator removes the need for laminated magnetic core
materials from the bit that rotates, allowing for solid steel poles and a
co-axially wound field coil. A design change that permits significantly
higher rotational speeds compared to the armature of a dynamo.
In which case (leaving aside the issue of friction), don't forget the
almost fixed 2 volts drop per brush (4 volts total) in the commutator
circuit of the classic dynamo.
Which losses also apply to the dynamo.
However, a further improvement in efficiency can be obtained by using a
permanent magnet based rotor and using a switching regulator to
efficiently convert the varying (with speed) output voltage to the desired
DC output voltage.
This principle is used in the "inverter" based generator sets such as the
Honda EU3000i with its "eco throttle" feature where the DC output is used
to power the 50/60Hz 230/120 volt inverter module.
Whilst the inverter based genset designs might seem to have introduced
additional losses (most notably that of the inverter module), the
efficiency difference is insignificant compared to the two benefits
conferred which are:
1) Decoupling of generator speed from the AC output frequency which allows
engine speed to be used as the main form of output power regulation when
the eco throttle feature is enabled.
2) Decoupling of reactive loads (in particular, leading currents due to
excess capacitance) which can adversely effect voltage regulation in the
classic (cheaper) AC generator based genset designs. In fact it this
'feature' that makes them so unsuitable for powering Personal Computers
(especially when protected by an in line UPS).
The output voltage regulation of a switching sinewave inverter remains
totally unaffected (within defined limits) by such reactive loads. For
example, a cheap 2.8KVA genset I once owned, would, on its 230v regulated
output setting, overvolt to 280v when connected to a 4.7 microFarad PF
correction capacitor borrowed from a 20W fluorescent light fitting. An
inverter based genset would suffer no such deficiency.
Regards JB Good
Posted by Daniel who wants to know on November 16, 2011, 6:06 am
I think even just adding a rotor position sensor (simple hall effect reading
the rotor poles just like in a PC brushless muffin fan) and replacing the
regular diodes with MOSFETs would enable synchronous rectification and get
rid of the diode forward drop. The fun part here is that the alternator
would also try to run as a BLDC motor when the engine speed dropped too low
so it would need a circuit to prevent it.
BTW I did not know how much was lost in the carbon brushes of the dynamo, I
just know I have a "3 bobbin" (cut-in/out, current, voltage) 12v mechanical
voltage regulator in front of me that I plan to use on an externally
regulated Chrysler/Bosch alternator to make a rotary 2 stage battery
charger, not efficient, but an interesting project nonetheless.
I still like the Toyota hybrid method, no alternator, just a PM rotor
motor/generator (MG1) and a regulated DC-DC converter for 12v battery
charging. It is kind of an inverter generator integrated into a car.
Posted by vaughn on November 16, 2011, 1:32 pm
This is the wave of the future. Another advantage is that they are compact and
can be designed for a wide range of rotor speed, so they can be driven right off
the crankshaft or possibly combined with some other engine accessory. These
things are gaining traction in homebuilt aircraft for just that reason.
Posted by Johny B Good on November 16, 2011, 6:00 pm
The figure I quoted was off the top of my head. I've since visited the
where a figure of 0.8 to 1.0 volt per carbon brush contact is quoted (a
total of 1.6 to 2 volts).
Those regulators use a 'switchmode' control of the field current, relying
on the inductance to average out the varying duty cycle of pulses
generated by the vibrating regulator relay. I'm not sure how well it would
work with an alternator's field winding (solid steel core). In theory,
there's no reason why it can't be made to work.
Switching converters, with optimum component choice, can exceed 95%
efficiency, making them a better choice over the traditional field
regulated generator designs.
Incidentally, the commutator brush gear on automobile dynamos typically
limited output to around the 20 to 30 amp mark. Alternators were able to
significantly raise this limit by use of high current silicon diode
Regards JB Good
Posted by m II on November 16, 2011, 11:32 pm
-----BEGIN PGP SIGNED MESSAGE-----
Johny B Good wrote:
I beg to disagree.
It could be argued that laminated armatures are far stronger than one
piece, cast versions. Each lamination has a hardened 'skin' on it that
forms during the manufacturing process.
They are certainly more vibration resistant than cast pieces. The reason
they are cast is one of economics. It's cheaper than assembling laminations.
As for the co-axially wound field, it would make no difference. The
reason alternators can spin so high is the fact that it's the *field*
spinning instead of the main windings. The rotor wire is a lot smaller
than that of generator, resulting in a great reduction of centrifugal
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
-----END PGP SIGNATURE-----