Oscillator with two seperate coils.

[Groundloop]

Hi All,

Can anybody explain why and how this oscillator works?

I must stress that there is no inductive coupling between L1 and L2.
L1 and L2 is far from each other and mounted 90 degrees to each other.
R1 has a high Ohm value, typical 1M Ohm.

GL.

[Groundloop]

When I try to simulate this oscillator in http://www.falstad.com/circuit/
then it will not oscillate. Is this a bad simulator?

$ 1 5.0E-6 4.43302224444953 60 5.0 50
t 400 240 416 240 0 1 -8.394010612157514 0.6059893878354361 400.0
w 416 224 416 192 0
w 416 256 416 288 0
w 400 240 368 240 0
r 416 192 368 192 0 1000000.0
w 416 192 416 160 0
l 416 160 416 112 0 9.999999999999999E-5 -0.0033659982755748436
g 416 288 416 304 0
v 240 112 288 112 0 0 40.0 9.0 0.0 0.0 0.5
w 288 112 416 112 0
g 240 112 240 128 0
w 368 192 368 240 0
w 368 240 336 240 0
l 336 240 336 288 0 9.999999999999999E-5 -1.2933143714086734E-20
c 336 288 336 320 0 1.0E-10 0.6059893878354069
g 336 320 336 336 0
c 336 112 336 160 0 1.0E-5 9.0
g 336 160 336 176 0
w 336 112 416 112 0
o 6 64 0 35 0.625 0.00625 0 -1
o 13 64 0 35 0.00244140625 9.765625E-5 1 -1

GL.

[poynt99]

Are the inductance values close to actual?

How long are the power leads?

What type of transistor is it?

[the_big_m_in_ok]

Are the inductance values close to actual?

How long are the power leads?

What type of transistor is it?

Just a thought?
If the coils are separated by a lengthy distance, capacitance on the wires in question might affect the whole circuit, yes?   Older computers from DOS days had prototype wires as short as possible for that reason and also potential electronic signal timing delay effects by the long distances as well.

--Lee

[Groundloop]

.99,

I do not have the circuit anymore because I did build it some years ago.
But the values is close to what I did use at the time. The circuit was
soldered onto a small universal board. Transistor was BC546C.
The battery was a 9 volt battery and the wires was approx. 5cm long.

GL.

[poynt99]

Were the inductors wound on two small ferrite toroids?

And I gather it doesn't work if one or the other inductors is absent?

[Groundloop]

Were the inductors wound on two small ferrite toroids?

And I gather it doesn't work if one or the other inductors is absent?

The inductors was ready bought 100uH inductors that looks like resistors.
No, both inductors must be there.

GL.

[poynt99]

Do you think it would work with a BC546B? It is about half the gain of the BC546C, and the "B" is what I have in PSpice unfortunately.

[poynt99]

Do you remember what frequency it was oscillating at? Something around 1.5MHz?

[Groundloop]

.99,

The BC546B has half the Hfe as the BC546C.
I did not use the B so I do not know if it will oscillate or not.
Can you modify the Hfe in PSpice?
I do not remember the fundamental frequency now.

GL.

[poynt99]

Yes I can.

I doubled the gain in the model, and it still did not oscillate. 

I tried increasing the cap to 470p, nogo. I reduced R1 to 10k, still a nogo.

Maybe you should rebuild it? 

[Groundloop]

Yes I can.

I doubled the gain in the model, and it still did not oscillate. 

I tried increasing the cap to 470p, nogo. I reduced R1 to 10k, still a nogo.

Maybe you should rebuild it? 

If I have more of the transistors then I can solder it together again.
I will look for the parts and see what I can find.

GL.

[BEP]

Just a thought...

Was there some coupling action between the two inductors on the physical circuit?

I don't recall if Falstad has that function.

[Groundloop]

Just a thought...

Was there some coupling action between the two inductors on the physical circuit?

I don't recall if Falstad has that function.

The two inductors was 20mm apart and orientated 90  degrees
to each other. So I think the inductance coupling was not there.

Gl.

[poynt99]

According to GL's first post, he arranged them orthogonal to one another, so there should be minimal coupling.

However, I tried it with some coupling anyway, and guess what, it does oscillate now.

It needs only as little as 0.1 coupling factor. You could never cancel the coupling 100%, especially if they are built close to one another, so I think this is pretty good evidence that you still have some coupling there GL.

.99

[Groundloop]

According to GL's first post, he arranged them orthogonal to one another, so there should be minimal coupling.

However, I tried it with some coupling anyway, and guess what, it does oscillate now.

It needs only as little as 0.1 coupling factor. You could never cancel the coupling 100%, especially if they are built close to one another, so I think this is pretty good evidence that you still have some coupling there GL.

.99

.99,

You may be right. I'm soldering together a circuit now.

What I want to try is shielding the base inductor so that
no coupling can happen and see what goes.

Thanks for taking time to do the simulation.

GL.

[Groundloop]

The circuit I soldered today did not oscillate at all.
I will change the two coils to 100uH tomorrow.
(Did not have the coils today.)

GL.

[Groundloop]

When I added an antenna to the simulator then the oscillation starts.

$ 1 5.0E-6 4.43302224444953 60 5.0 50
t 400 240 416 240 0 1 -14.929013314318563 -9.506396041410826 400.0
w 416 224 416 192 0
w 416 256 416 288 0
w 400 240 368 240 0
r 416 192 368 192 0 1000000.0
w 416 192 416 160 0
l 416 160 416 112 0 9.999999999999999E-5 -15875.000553532427
g 416 288 416 304 0
v 240 112 288 112 0 0 40.0 9.0 0.0 0.0 0.5
w 288 112 416 112 0
g 240 112 240 128 0
w 368 192 368 240 0
w 368 240 336 240 0
l 336 240 336 288 0 9.999999999999999E-5 1.4929013364320398E-5
c 336 288 336 320 0 1.0E-10 -10.803797281957856
g 336 320 336 336 0
c 336 112 336 160 0 1.0E-5 9.0
g 336 160 336 176 0
w 336 112 416 112 0
A 416 192 544 192 0 0 40.0 5.0 0.0 0.0 0.5
o 6 64 0 35 40.0 26214.4 0 -1

GL.

[EMdevices]

This circuit is Dr Stifflers circuit minus the resonant coil connected to the collector.  To eliminate coupling you would have to go to extremes and orient the physical devices with high precision, but you don't want that.  However, it's not just magnetic coupling but capacitive coupling that played a part. We mentioned this to Stiffler in the beginning and now his circuits are on top of a conductive ground plane to enhance coupling.

EM

[giantkiller]

This is an interesting circuit.
I cannot get to Falstad's site at work to try this but I am thinking...

What if you make the antenna a large loop and parallel that with another loop that is hooked somewhere else in the circuit to transmit either in phase or out of phase to the antenna/collector loop? This makes it a [closed] feedback operation. I use [closed] loosely. We have all seen this type of unwanted operation before in places we did not want it. Well what if in this case and we do?

Oh wow! What if SM's collector is really a label to the antenna connected to the [collector of a transistor]?

[BEP]

We mentioned this to Stiffler in the beginning and now his circuits are on top of a conductive ground plane to enhance coupling.

I didn't think many caught that 

So, now he claims more free energy because he was provided correct information 

[ion]

Some time ago (maybe 2 years) I built a similar circuit using an FET, and could not explain the oscillation at that time.

It worked even with shielded inductors some 3 or 4 feet distance from each other.

Orientation of the coils did not affect the oscillation.

I can only guess Miller Effect has something to do with the oscillation, or possibly some other capacitive coupling mechanism.

[EMdevices]

I modeled the circuit and it easily oscillates, just need to play with the values.    I can get it to oscillate in bursts or constant, and here is a graph.    Based on the frequency, I can tell that the base curent is resonating through an even lower capacitance inside the transistor.

If we take in consideration the transistor model, we might find that this is nothing more then a Colpits oscillator!

EM

[Groundloop]

I modeled the circuit and it easily oscillates, just need to play with the values.    I can get it to oscillate in bursts or constant, and here is a graph.    Based on the frequency, I can tell that the base curent is resonating through an even lower capacitance inside the transistor.

If we take in consideration the transistor model, we might find that this is nothing more then a Colpits oscillator!

EM

EM,

I think this oscillator is more like a Clapp oscillator.
The Clapp oscillator is a Colpitts oscillator that has
an additional capacitor placed in series with the inductor.
The resonant frequency is set by the LC tank circuit on
the base of the transistor.

GL.

[Groundloop]

Today I checked to see if the coupling between the two coils matter.
I did use a larger coil on the base to get a low coupling between the
two coils, and the oscillation did start.

EDIT 23.04.2013 I did not get the component values correct the first time so here is the correct version.

GL.