Hybrid Car – More Fun with Less Gas

eeked out another 8 volts on the hydro - Page 10

register ::  Login Password  :: Lost Password?
Posted by Michael B on October 19, 2008, 11:25 pm

It was the photo of the setup that looked like a vehicle alternator
was being used.
Need strong magnets? Get the two from the magnetron portion of a
discarded microwave unit. You see them being thrown away regularly,
easy enough to scrounge the magnets.

Posted by z on October 19, 2008, 11:49 pm

It uses the same housing as a GM alternator but its PM from

I started with an old electric motor but it didn't cut the mustard so I
splashed out on one of those -- once I proved to myself I could make the
wheel spin :)

Posted by Michael B on October 20, 2008, 1:03 am

Gotcha. Damn, they are pricey.

Posted by z on October 21, 2008, 5:31 pm

yeah.. they seem well made though.  Great communication and the guys are
very helpful.    I didn't want to screw around with winding my own
alternator or making a PMA myself.  It was winter and the water was flowing
and I wanted that power NOW :)

Posted by Neon John on October 22, 2008, 2:28 am

Seems like a good time to create a new thread.

Here's how to achieve the same results and more on the cheap.

The only objection to a standard car alternator is that the field consumes
considerable power.  The slip rings impose practically no load and the brushes
last forever so they're not a concern.

OTOH, a PMA has a major problem, as Z has found out - a fixed constant between
RPM and voltage.  Not so terribly big a deal if you can gear the alternator to
the correct speed to match the prime mover but a major problem with direct
drive applications.  There are other secondary bad effects such as the torque
requirement if the magnet selected is too strong and the alternator thus must
be run at a significantly lower speed than design.  Losses accrue both in the
drivetrain and in the low frequency generated in a stator designed for higher

There is a solution that solves all these problems.  The hybrid field.  The
hybrid field contains permanent magnet(s) that supply all or nearly all the
field necessary for normal operating conditions.  the rotor ALSO contains a
small WOUND FIELD, connected to the slip rings, that is used to trim the field
to what is needed at any given moment.  The wound field can either add to or
oppose the PM field, depending on the polarity of the supplied current.

Physically, the rotor spider is pressed apart and the bobbin-wound field coil
removed.  A suitable donut magnet(s) is/are selected that fills most of the
space taken by the wound bobbin.  Then a smaller bobbin-wound coil is mounted
next to the permanent magnet, the leads attached to the slip rings and the
assembly pressed back together.

A significant bit of math can be involved or one can simply experiment to
develop the coil and magnet strengths required.  I built such a unit from a
Delco alternator of 80s vintage.  I wanted to minimize the power the custom
voltage regulator that I'd designed would have to supply.  I also wanted to
eliminate the large voltage spikes that occur when a conventional alternator
has its load suddenly reduced or disconnected.

If the alternator is nearly fully loaded and turning at idle or cruising speed
(not very fast) then the field is fully or nearly fully energized.  There is
considerable energy stored in the field.  If the load is suddenly reduced, the
regulator attempts to reduce the field current to bring the output voltage
back down but it can only do so at a rate at which the energy in the field can
be dissipated.  That can take a second or more, depending on the regulator
design.  Meanwhile the alternator can be producing >50 volts.

I've worked on cars where the helpful (NOT) next-door-neighbor shadetree
mechanic "tested" the alternator by pulling a battery lead.  The theory was
that the car would keep running if the alternator was good.  The test worked
in the old generator days.  Especially when alternators were fairly new (up
into the 80s), and the other electronics weren't designed to withstand such
load dumps, the resulting voltage spike would burn out every electronic device
in the car.

I once had a 76 Datsun Z come in my shop that was rendered inert from just
such a stunt.  The EFI computer.  The radio.  The tach.  The electronic
ignition box.  All smoked.

Anyway, I had a good deal of ham radio gear in my car, gear that required a
lot of power.  I was adding a large alternator to my car but I wanted to avoid
this kind of problem in case a battery wire came loose.  This idea occurred to
me, an idea that turns out to be quite old.

The way I developed mine is thus.  I assumed an engine speed of 1500 RPM,
normal engine speed at freeway cruising.  Figuring back from the pulley ratio,
I computed the alternator speed.  Seems like the step-up ratio was nearly 2:1

With the alternator spinning at that RPM and full amps being drawn, I measured
the field current and voltage.  Field current was the main parameter.

I took the alternator apart and suspended the rotor horizontally from string
slung over a joist.  I forged a piece of hot rolled steel to fit the curvature
of the rotor that covered maybe 45 deg.  The amount didn't matter.  I welded
an eye to the arc of steel.

I connected a power supply to the slip rings, set the amperage to the same as
that I measured while the alternator was alternating :-) and added weight to
the eye until the steel piece pulled away.  I did this several times and
computed statistics to make sure the test was repeatable.  It was.

I now knew how much "suck" the rotor needed to develop against that steel
piece to achieve the same output.  I pressed the rotor apart, removed the coil
and started substituting donut magnets.  Ceramic speaker magnets weren't
strong enough so I ended up ordering some Alnico magnets (rare earth magnets
didn't yet exist.  

Experimentally, I determined that I could generate 5-10 pounds of 'suck' with
just a few watts of power through the aux coil so I sized the Alnico magnets
to achieve about 10 lbs less 'suck' than with the original coil.  I didn't
have a bobbin the right size so I went to my friendly local electric motor
shop owned by a friend of mine, wound a circular coil of the correct
dimension, laced it securely and dipped and baked it.  No bobbin needed.

The process of multiple assemblies and disassembles pretty much ruined the
rotor's press fit so I got a new one, pressed it apart, put the magnets and
coil in it and pressed it back together.

The analog regulator I designed had a bi-polar (H-bridge) architecture that
could drive the trimmer coil to either polarity.  It had to add a little
magnetism at idle but subtracted some at high speed.

I ran it in the car long enough to know that it worked great.  The regulator
was tiny and usually had to supply far less than an amp to the field.  The
life happened, new job, lots of travel, move to another state and the project
got sidelined.  I think that I still have the alternator around here

For a micro-hydro or wind application, this would be a perfect solution.  Rare
earth magnets would make getting the right PM strength easy.  For a fairly
constant speed application like a micro-hydro, little to no electric field
trimming would be necessary.  If the water flow deviated in either direction
for whatever reason, a simple trim (presumably by an automatic regulator) of
the field would bring the voltage back up.

The trimmable field can also serve, in effect, as electrical gearing.  If the
prime mover is capable of high speed but low torque, a low field intensity
causes the alternator to make the desired voltage at high speed with little
torque demand.  OTOH, if the prime mover is a low speed high torque unit (say,
an old-fashioned wind mill), a strong field lets the alternator generate the
necessary voltage at low speed while requiring high torque.

John De Armond
See my website for my current email address
http://www.johndearmond.com  <-- best little blog on the net!
Tellico Plains, Occupied TN
If stupidity hurt then they'd be putting morphine in the water supply.

This Thread
Bookmark this thread:
  • Subject
  • Author
  • Date
please rate this thread