charge batteries via inductive coupling

[TonyMatthews]

Hello
I need help figuring out the coil structure for a toy I want to
build. I have a rubber ball with a battery inside. I want to recharge
the battery without a galvanic connection. I'm looking to get about
300 ma charge current. The ball will sit in a cup shaped recharger.
And the coil in the charger will be powered by a wall wart with a
nominal 12 vac output and around 500ma or less.
What I need is help in figuring out the shape of the coils and their
positions and the number of turns and so on. I am fairly good with
most electronics but never could grok this inductive stuff. The math
is above me. If I could just get a bit of advice on some basic aspects
like for instance:
1. Would the coil shape for the charger cup be better as a say 1.5"
donut shape mounted in the bottom of the cup just below the surface
with the other coil in the ball just above and stacked like two
donuts, be better than a 3" coil in the rim of the cup with the ball's
coil like a smaller donut in the hole of the larger charger coil's
donut?

2. How do I determine the number of turns I need to limit the current
drawn from the supply to it's rated power output. assuming 60hz at
around 12 vac?

3b. should I be looking at increasing the frequency driving the coil?

3. I would prefer an air core but it would be nice to know if that was
a mistake.

4. given the coil on the primary side is x turns what should the coil
on the secondary side be. I need 5 volts or more at 300 ma or so. To
run the ciruitry and have the over head to charge the battery.

5. How can I predict the effect of seperation on the power exchanged
by the two coils? I would like at least an inch between them is this
feasible? And if not what number should I use?

 

[Vladimir Vassilevsky]

Hello
I need help figuring out the coil structure for a toy I want to
build. I have a rubber ball with a battery inside. I want to recharge
the battery without a galvanic connection.

The coupling between the coils is proportional to the dipole moments of
the coils and inverse proportional to the 3rd power of the distance
between the coils. The dipole moment of a coil is roughly proportional
to the volume of the coil x number of turns x core permeability.
Ideally, the coupling is independent of the frequency; however it is
easier to get better matching and lower losses at the reasonably high
frequency (~tens of KHz).


Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com

 

[ChairmanOfTheBored]

Hello
I need help figuring out the coil structure for a toy I want to
build. I have a rubber ball with a battery inside. I want to recharge
the battery without a galvanic connection. I'm looking to get about
300 ma charge current. The ball will sit in a cup shaped recharger.
And the coil in the charger will be powered by a wall wart with a
nominal 12 vac output and around 500ma or less.

Just buy an electric toothbrush from OralB, take it apart and use the
elements it has. They charge in their stand without contacts by way of
inductive coupling. There is no way that they operate at 300mA charge
current though. I think you need to key that figure down a bit
realistically speaking. Or maybe not... :-]

 

[ChairmanOfTheBored]

3. I would prefer an air core but it would be nice to know if that was
a mistake.

Transformer that use a core needs to have the windings around said core
in order to pass the energy from one winding to another.

An air core configuration is best because it creates the core coupling
through the air, regardless of the distance between the windings (up to a
point, of course). Winding orientation is also an important factor. The
turns have to be parallel to each other to maximize coupling efficiency,
and minimize any losses.

You'll have to mark the ball so that you know which side goes into the
cup receptacle of the charging stand (or put some Braille on it for sight
impaired applications).

 

[MooseFET]

A bit of extending on Vladimir's comments is worthwhile:

The coupling between the coils is proportional to the dipole moments of
the coils and inverse proportional to the 3rd power of the distance
between the coils.

It only gets to 3rd power when the distance gets big. Look up the
equation for the field from a coil vs distance and use this to figure
how much field is cutting the receiving coil. Ideally, the ball will
fit into the charger such that its coil is concentric with the driving
coil.

The dipole moment of a coil is roughly proportional
to the volume of the coil x number of turns x core permeability

x the current.

This last is important because as you make the coil bigger to get more
moment, you raise the inductance thus making the drive to get a given
current harder to do.

The efficiency depends largely on the number of pounds of copper you
are willing to use.

..

Ideally, the coupling is independent of the frequency; however it is
easier to get better matching and lower losses at the reasonably high
frequency (~tens of KHz).

Use Schottky recifiers. Since you are charging a battery, you know
the load side voltage.

If the coupling is loose: Ideally, you want the load side resonant and
to have its impedance matched by the charging circuit. Making the
primary side resonant can make it much easier to drive.

 

[MooseFET]

On Dec 9, 11:43 am, TonyMatthews wrote:

Read the Vladimir reply and the extending of it. After I pressed
"send" I though of another comment I should have made.

[....]

3. I would prefer an air core but it would be nice to know if that was
a mistake.

It doesn't rise to the level of mistake but, adding a core will reduce
the amount of copper you need to use. These days iron is much cheaper
than copper so it would be worth adding some.

Remember that the iron must not go between the two coils. If the ball
fits down inside a cup that makes the windings concentric, you want
the core to be inside the ball. This would be troublesome from a
structural point of view. If you have the cup coil's below the ball's
coil, adding a core below the cups coil could improve the coupling.

ASCII ART:


OOO OOO <--- Ball coil
OOO OOO

OOO OOO <- Power coil
IIIIIIIIIIIIII
IIIIIIIIIIIIII
IIIIIIIIIIIIII <- Iron core
IIIIIIIIIIIIII

The core causes the lines of force to take the longer path around the
Power coil. This makes the field at the Ball coil larger.

 

[gearhead]

Hello
I need help figuring out the coil structure for a toy I want to
build. I have a rubber ball with a battery inside. I want to recharge
the battery without a galvanic connection. I'm looking to get about
300 ma charge current. The ball will sit in a cup shaped recharger.
And the coil in the charger will be powered by a wall wart with a
nominal 12 vac output and around 500ma or less.
What I need is help in figuring out the shape of the coils and their
positions and the number of turns and so on. I am fairly good with
most electronics but never could grok this inductive stuff. The math
is above me. If I could just get a bit of advice on some basic aspects
like for instance:
1. Would the coil shape for the charger cup be better as a say 1.5"
donut shape mounted in the bottom of the cup just below the surface
with the other coil in the ball just above and stacked like two
donuts, be better than a 3" coil in the rim of the cup with the ball's
coil like a smaller donut in the hole of the larger charger coil's
donut?

2. How do I determine the number of turns I need to limit the current
drawn from the supply to it's rated power output. assuming 60hz at
around 12 vac?

3b. should I be looking at increasing the frequency driving the coil?

3. I would prefer an air core but it would be nice to know if that was
a mistake.

4. given the coil on the primary side is x turns what should the coil
on the secondary side be. I need 5 volts or more at 300 ma or so. To
run the ciruitry and have the over head to charge the battery.

5. How can I predict the effect of seperation on the power exchanged
by the two coils? I would like at least an inch between them is this
feasible? And if not what number should I use?

How about charging the ball _inside_ a solenoid, where
the magnetic field is very strong.
Put a marking on the ball to enable the user to orient it correctly,
so that the coil inside the ball aligns coaxially with the outer
solenoid.
Don't use a toroid.

 

[Paul Hovnanian P.E.]

I see a problem with the geometry here. For maximum coupling, the base
coil and the coil in the portable unit should align along a common axis.
With a spherical device, how can you ensure that it will be placed in
the charging base with its coil aligned properly?

 

[John Tserkezis]

I see a problem with the geometry here. For maximum coupling, the base
coil and the coil in the portable unit should align along a common axis.
With a spherical device, how can you ensure that it will be placed in
the charging base with its coil aligned properly?

Position the coil around the circumference, and along with the iron core,
weight the ball so it falls or rolls when dropped into the cradle.
Have the primary coil in the cradle positioned alongside where the ball
secondary coil would be.
This would enable a completely spherical ball, but I foresee positioning
problems depending on the type of ball (fluffy??).


An alternative would be to make it a "girl" ball and create a cavity in
which the charging coil could be inserted. Something along the lines of the
Oral-B electric toothbrush and cradle (and other brands).
300mA might be a bit much if you want to keep the charging receptacle small
though. You would be forced to use a larger size ball if you need a large
receptacle.

 

[Jim Thompson]

But, being a ball, a random orientation is likely - it IS a toy and the
owner/operator may well not follow any written instructions.

It seems, since the OP says there's a battery "inside", you just adapt
the air-filler port to accommodate a power jack.

...Jim Thompson

 

[MooseFET]

But, being a ball, a random orientation is likely - it IS a toy and the
owner/operator may well not follow any written instructions.

Some adults may prefer a toy that quits working a few days after
Xmas. If not, three coils in the ball solves that problem.

 

[MooseFET]

It seems, since the OP says there's a battery "inside", you just adapt
the air-filler port to accommodate a power jack.

That assumes that there is an air filler port. I have a ball that
makes all manner of noises when it is bounced and flashes LEDs. There
is a port for the sound but if that was deleted, the ball would be
near enough solid.

 

[MooseFET]

I see a problem with the geometry here. For maximum coupling, the base
coil and the coil in the portable unit should align along a common axis.
With a spherical device, how can you ensure that it will be placed in
the charging base with its coil aligned properly?

Three coils on 3 axis solves this but not perfectly. A coil at a 45
degree angle has less coupling.

 

[rebel]

A bit of extending on Vladimir's comments is worthwhile:
(snip)

Ideally, the ball will fit into the charger such that its coil is
concentric with the driving coil.

But, being a ball, a random orientation is likely - it IS a toy and the
owner/operator may well not follow any written instructions.

 

[ChairmanOfTheBored]

I see a problem with the geometry here. For maximum coupling, the base
coil and the coil in the portable unit should align along a common axis.
With a spherical device, how can you ensure that it will be placed in
the charging base with its coil aligned properly?

If you had read the thread, you would have noted where other mentioned
such a need. A simple mark on the ball will suffice.

 

[ChairmanOfTheBored]

But, being a ball, a random orientation is likely - it IS a toy and the
owner/operator may well not follow any written instructions.

Instead of a cradle, make sure the place the ball sits has the
semisphere for the ball to sit in and a back, like a chair back, that
runs up the back of the ball. Place three coils in the charger. Two in
the base, and one up the chair back. Have the circuit detect which one
is driving the coupled charging circuit the best, and turn the other two
off. Or, if they are all in series, just leave them all on. One
orientation will charge perfectly. Most others will charge well, and one
or two orientations will charge miserably, but will still charge.

 

[JosephKK]

MooseFET [email protected] posted to sci.electronics.design:

A bit of extending on Vladimir's comments is worthwhile:



It only gets to 3rd power when the distance gets big. Look up the
equation for the field from a coil vs distance and use this to
figure
how much field is cutting the receiving coil. Ideally, the ball
will fit into the charger such that its coil is concentric with the
driving coil.

Orienting the ball is part of the problem. Any workable solution
includes orientation.

 

[MooseFET]

MooseFET [email protected] posted to sci.electronics.design: [..Ball being charged..]
Orienting the ball is part of the problem. Any workable solution
includes orientation.

Yes include:

3 coils in the ball solves this problem.

The shapes of the coils will need to be distorted a bit a the point
where they intersect. I don't think this will reduce the coupling
area enough to matter.

 

[AndyS]

Yes include:

3 coils in the ball solves this problem.

The shapes of the coils will need to be distorted a bit a the point
where they intersect. I don't think this will reduce the coupling
area enough to matter.

Andy comments:
Or... you can include multiple coils in the charger and phase shift
them
so you have a rotating charging field.... much like the rotating field
in an
induction motor......
That way, the charger is slightly more complicated, and the ball
is simpler...

A 3 axis rotating M-field isn't hard...... just think about it for
a
couple minutes and you will understand....

Andy in Eureka, Texas

 

[Paul Hovnanian P.E.]

Position the coil around the circumference, and along with the iron core,
weight the ball so it falls or rolls when dropped into the cradle.
Have the primary coil in the cradle positioned alongside where the ball
secondary coil would be.
This would enable a completely spherical ball, but I foresee positioning
problems depending on the type of ball (fluffy??).

An alternative would be to make it a "girl" ball and create a cavity in
which the charging coil could be inserted. Something along the lines of the
Oral-B electric toothbrush and cradle (and other brands).
300mA might be a bit much if you want to keep the charging receptacle small
though. You would be forced to use a larger size ball if you need a large
receptacle.

All of this assumes that the intended use of the ball can tolerate the
asymmetry in balance or shape.

Here's another idea: Put a small motor in the charging base that will
rotate the ball through contact with a friction wheel. Mount the wheel a
bit off center so that the ball wobbles and eventually it will reach a
position where a suitable coupling threshold is reached, at which time
the motor shuts off.