So i just found my personal holy grail - a color 27 (incandescent fluorescent) in the recycle bin. Sadly, the cathodes are completely worn. When it finally lit after many start attempts, the current slowly dropped and after 30 seconds it was gone and started flashing again.

But since cathode failure is pretty common, i've been thinking of a way of powering these things without using the cathodes. A cold cathode discharge is not an option - they run way too hot. I'm a ham radio operator. I've played around with fluorescents before just for fun.
But what if i purposefully hook up a fluorescent tube to the transmitter?

I grabbed a completely fine CW 20w t12 tube (the dead color 27 one is at my new house, my ham radio stuff at my old)  and wrapped some aluminium foil around the ends, taking care that it doesn't contact the end caps or the pins.
I first hooked it up straight to my tuner - no luck, only very faint partial ionisation.
Then i grabbed a little coil i wound a long time ago for a crystal receiver. About 80 turns with a tap at 10. Hooked up the 10 turn tap to my tuner's unbalanced output, the 80 to one side of the tube and GND to the other. It's running on 3500khz.

Video: https://youtu.be/TvPJrV3VSUk

I think a purpose built power oscillator will be able to power the 65w lamp at near full brightness. I will have to figure out screening - with 65w of RF power (currently running 10w from my little yaesu ft-7) and E-fields of many hundreds of volts per meter, i am pretty sure that even with the short wires it will emit a few watts, which is enough to reach all of my country on 3,5MHz.

This may also be a way to dimly light your tubes without any risk on sputtering at all, i reckon.

Thing is, when most fluorescent lamps on electronic ballasts go EoL, they will generally get hot enough that the glass cracks and then you have vacuum loss and they will not light no matter what. 

Not all ballasts are electronic ones, so if you got a worn out one that still got its vacuum, this can be a way to get them to light.

Yes this works, but beware the glass under the foil gets rather hot and will crack too if you drive the tube hard enough. Ask me how I know )


Happy New Year to everyone here!

I think a purpose built power oscillator will be able to power the 65w lamp at near full brightness. I will have to figure out screening - with 65w of RF power (currently running 10w from my little yaesu ft-7) and E-fields of many hundreds of volts per meter, i am pretty sure that even with the short wires it will emit a few watts, which is enough to reach all of my country on 3,5MHz.

Fortunately or no, no way it gonna happen

Laws of electrodynamics tell for antenna to be efficient it shall be sized comparable to the wavelength, in practice at least lambda/2 dipole or lambda/4 whip across the ground plane. Anything smaller generally just pumps RF energy around the immediate vicinity with nothing radiated further. Your setup may radiate a little, but in the range of milliwatts.

Small antenna for lower SW bands is sure a long term wet dream of radio amateurs, but it is just physically impossible. Many times you can live with an inefficient antenna for reception assuming input stage has low noise and high sensitivity, however it just does not work for transmission when you have to fight with interferences using a real radiated power.

That's a lovely lamp, real shame it's damaged.

That's a lovely lamp, real shame it's damaged.

It is! But if i can find a way to get it here, i've been offered an identical lamp by another forums member
Probably going to be an interesting train ride, or a 'taxi' ride with another dutch lighting enthousiast.

If i can find a way to get this project to work well enough, i'll probably still keep this one running at a quarter power or something, just to look nice in the corner of a room.

Fortunately or no, no way it gonna happen

Laws of electrodynamics tell for antenna to be efficient it shall be sized comparable to the wavelength, in practice at least lambda/2 dipole or lambda/4 whip across the ground plane. Anything smaller generally just pumps RF energy around the immediate vicinity with nothing radiated further. Your setup may radiate a little, but in the range of milliwatts.

Small antenna for lower SW bands is sure a long term wet dream of radio amateurs, but it is just physically impossible. Many times you can live with an inefficient antenna for reception assuming input stage has low noise and high sensitivity, however it just does not work for transmission when you have to fight with interferences using a real radiated power.

I'll fire up the Yaesu ft101 and put an FT8 signal into it, we'll see how good of an antenna a 65w fluorescent lamp is 
I am aware of the limitations. However, i've also noticed that on short wave frequencies, the actual necessary radiated power to notice a signal across a couple hundred kilometers can be extremely small on good days. As in 'I cannot see my power meter deflect' which in my case means less than 100mW into a 1/8 or 1/4 wavelength vertical.

Although efficiency rapidly drops with smaller antennas, i've also heard my own CW signals across several countries with just a wire of 4m long on 80m, running 10w or less. I did use a somewhat proper ground of course.

Well, but originally we discussed a possibility of leakage of several *watts* RF out of a fluorescent tube driving circuit @3.5MHz. This just won't happen.

Running QRP some hundreds mW into a large tuned antenna has absolutely nothing to do with this. Also in fact transmitting distance (all other things equal - bandwidth, environmental noise level at receiver endpoint) is about proportional to transmitter field strength, that is antenna voltage, not antenna power. This makes a real difference of transmitting at 500mW and say, 10W much less dramatic. Though radio people love talking Watts )