I tried to rebiuld a Don Smith Device

[Farmhand]

I made a typo in my last post, I meant to writ 2k - 10k resistor from gate to ground.

..

[Google]

See attachment for circuit to alternate two 12 volt globes. The inverter circuit I modified to get the drawing attached is the 8th schematic down using the CD4047 as the oscillator chip. One wire from each bulb is connected to the 12 volt rail and the other is switched alternately through the two mosfet switches, a CD4047 drives the mosfets just fine at lower frequencies. The 1K resistors to the mosfet gates is not exactly how I would do it, I would remove those and drive direct from the CD4047 outputs to the mosfet gates and place a 2k - 10 k resistor from the gate to the ground to ensure the mosfets turn right off.

8th circuit down on this page is the one I modified to get the circuit attached. Don't be concerned with using mosfets they are great for switches.
http://circuit-zone.com/?cat=AC_DC_Inverters

Cheers

Can it be done like as in the attached picture, without destroying the audio amplifier ? 12 v 22 w bulb resistance is roughly 6 ohms, which is more than the rated impedance of the audio amp output of minimum 4 ohms.

Best,

[dieter]

(referring to that antenna receiver patent posted by google earlier) It seems to be not that incredibly simple when you need a half mile antenna. But interesting data about the frequency. But one of the more complicated concepts to verify.


About your audio amp question, I guess that would work, but you need a stereo frequency generator for the signal input. eg. a small astabile multivibrator or however it's called again. You can do that with two little signal transistors.


I have made a program that allows to draw a waveform with the mouse and save it as WAV, then play it at desired sampling rates from 1 to 200'000hz, but unfort. the pc soundcard has a basic output mixing rate that may be a harmonic to 44100 and results in moiré patterns/distortion of nonharmonic rates. So an analogue frequency generator or flipflop seems to be superior.

[Farmhand]

If a incandescent globe with a coiled filament is switched it produces spikes the same as any coil does ( see video below), if it's switched at the right frequency the filament will take time to "glow down" heat wise but the energy stored in the magnetic field of the coil of the filament will be released, and can be utilized. Under the right conditions it might prove to be more efficient than when the bulb is supplied straight DC or AC even, maybe. The resonant frequency of the light bulb filament coil may be altered with a capacitor, just like we can with any coil. 

Light bulb filament coil discharge. Again the video is rough but I'm not expecting a nomination for a Logie award  (Logie is the Aussie version of a Grammy I think).
http://www.youtube.com/watch?v=l89YgQIIqUY

..

[Google]

(referring to that antenna receiver patent posted by google earlier) It seems to be not that incredibly simple when you need a half mile antenna. But interesting data about the frequency. But one of the more complicated concepts to verify.


About your audio amp question, I guess that would work, but you need a stereo frequency generator for the signal input. eg. a small astabile multivibrator or however it's called again. You can do that with two little signal transistors.


I have made a program that allows to draw a waveform with the mouse and save it as WAV, then play it at desired sampling rates from 1 to 200'000hz, but unfort. the pc soundcard has a basic output mixing rate that may be a harmonic to 44100 and results in moiré patterns/distortion of nonharmonic rates. So an analogue frequency generator or flipflop seems to be superior.

Hey, iphone or samsung smartphone output is stereo output. We just need an android application that can generate frequencies higher than 20,000 hz.

Best,