LoL- my cheap function generator makes both waves with a positive bias just fine..  No problem with diodes.

Any recommendations on 2 channel amplification?  I think regular lower frequency is what we need here.  Figuera ran his off a motor with 16 contacts which make 1 sinewave per rotation.  So he certainly wasn't running 100's of hertz..

Hi Floodrod,

Good to see the Sinewave waveforms can be generated (too easy - right!). Note that your generator can probably drive
a short circuit for about 60 seconds (check specs) so it does have some decent drive capability.

Citfa's Class A Amplifier is a good way to drive the coils and likely the simplest to implement.

Another option is to use a Voltage Follower Op Amp (a single supply rail or, if you have it, dual rail). Attached a few pdf
files that show the idea, but there's probably a lot better information available as well. More complex (?) but more options
and controls.

SL

Citfa's Class A Amplifier is a good way to drive the coils and likely the simplest to implement.

Another option is to use a Voltage Follower Op Amp (a single supply rail or, if you have it, dual rail).

Thanks guys..  I ordered 2 Class A 1969 amplifier boards coming from China.  So it will be a few weeks.  I may order some amazon parts and tinker to make one while I wait.  But circuitry isn't my strong suit.

I think a regular more efficient amplifier may work..  Those Class A's are hot inefficient beasts.. 

Signal generator set to Triangle- then run the Stereo amplifier output through a bridge rectifier.  It flips the negative side to positive and evens out nicely.

Or if you want to mimic figuera's steps, you can use Post or Neg Ladder through a rectifier.  Again it flips the negative side to positive in an equal manner. Now it look like the resistor rig.

Then just use 2 channels and set the phase to 180.  <--  WRONG

EDIT-  With FWBR's, you need to use 90 degrees phase shift to get 180 degrees output..

(using regular sine is close, but it V's at the bottom a little too much, unless your signal generator can make a half-circle sinewave. )  But that triangle looks pretty dead-on to me..

Now I don't have to wait weeks..

I think a regular more efficient amplifier may work..  Those Class A's are hot inefficient beasts.. 

Signal generator set to Triangle- then run the Stereo amplifier output through a bridge rectifier.  It flips the negative side to positive and evens out nicely.

Or if you want to mimic figuera's steps, you can use Post or Neg Ladder through a rectifier.  Again it flips the negative side to positive in an equal manner. Now it look like the resistor rig.

Then just use 2 channels and set the phase to 180. 

(using regular sine is close, but it V's at the bottom a little too much, unless your signal generator can make a half-circle sinewave. )  But that triangle looks pretty dead-on to me..

Now I don't have to wait weeks..

FWIW - Your FG has AWG (store arbitrary signals) so you should be able to store an "Up Ladder" + "Down Ladder" sequence
(and vice-versa for your other input) and replay them in a loop.    I haven't tried it so just guessing? But might be worth a look.

However, you might have to "build the waveform" using the software supplied with the generator [another option].

SL

FWIW - Your FG has AWG (store arbitrary signals) so you should be able to store an "Up Ladder" + "Down Ladder" sequence
(and vice-versa for your other input) and replay them in a loop.    I haven't tried it so just guessing? But might be worth a look.

However, you might have to "build the waveform" using the software supplied with the generator [another option].

SL

The image I posted is using a regular stereo amplifier.  I got the wave full positive bias after the amplifier...

Scope is hooked up to the DC side of bridge rectifiers on the output of the stereo.

The image I posted is using a regular stereo amplifier.  I got the wave full positive bias after the amplifier...

Scope is hooked up to the DC side of bridge rectifiers on the output of the stereo.

Floodrod - Quite inventive I must say - looks good!
{{ Maybe I should remove my last few posts - good info but they clutter the thread? (unless ur a mod and can do it) }}

Gyuasun - Thanks - I have the latest (but looks like the last update we'll see; unfortunately).

Great progress, you could also try out a rheostat, the swiping arm you slide back and forth manually hereby emulating movement of the G component, but current flows continually and isn't interrupted which may or maymnot be desirable. 
imho the fields don't oppose, the purpose is to make one N pole grow and the other S shrink, inducing variation and a flux flow WHILE net change of flux remains zero. Or creating a magnetic current, the analog is displacement current which is an electric field that varies, magnetic current is considered hypothetical, a mathematical construct, but the same is said about displacement current. 
Purely hypothetical: Since the time rate of change of each coil is equal and opposite, the mutual induction cancels the self-induced emf, resulting in a non-inductive coil (V1=L1.di1/dt + M.di2/dt, say for a single period during rise of 1 sec, i1(t) = 5t and i2(t)=-5t+5, then di1/dt=5 and di2/dt=-5, and V=0 if L=M).

Figuera: 
PRINCIPLE OF THE INVENTION - Watching closely what happens in a Dynamo in motion, is that the turns of the induced circuit approaches and moves away from the magnetic centers of the inductor magnet or electromagnets, and those turns, while spinning, go through sections of the magnetic field of different power, because, while this has its maximum attraction in the center of the core of each electromagnet, this action will weaken as the induced is separated from the center of the electromagnet, to increase again, when the induced is approaching the center of another electromagnet with opposite sign to the first one.
Because we all know that the effects that are manifested when a closed circuit approaches and moves away from a magnetic center are the same as when, this circuit being still and motionless, the magnetic field is increased and reduced in intensity; since any variation , occurring in the flow traversing a circuit is producing electrical induced current .It was considered the possibility of building a machine that would work, not in the principle of movement, as do the current dynamos, but using the principle of increase and decrease, this is the variation of the power of the magnetic field, or the electrical current which produces it.
The voltage from the total current of the current dynamos is the sum of partial induced currents born in each one of the turns of the induced. Therefore it matters little to these induced currents if they were obtained by the turning of the induced, or by the variation of the magnetic flux that runs through them; but in the first case, a greater source of mechanical work than obtained electricity is required, and in the second case, the force necessary to achieve the variation of flux is so insignificant that it can be derived without any inconvenience, from the one supplied by the machine. Until the present no machine based on this principle has been applied yet to the production of large electrical currents, and which among other advantages, has suppressed any necessity for motion and therefore the force needed to produce it.

"when the induced is approaching the center of another electromagnet with opposite sign to the first one."