Plasma Speaker

Another old project that I'm just getting around to writing about now. This is a simple plasma speaker, built around a flyback transformer with the core shims removed driven by an arrangement of a TL5001 PWM chip feeding a FAN73832 high and low side N-type MOSFET driver.

It plays pretty well, though the bass tends to distort. A low pass filter on the audio input would not be amiss. The FETs run warm, but not too hot to touch.

Sprucing up an old radio..

I picked up a trashed Crosley 1930's radio at the yearly local antique radio association's swap meet/flea market at the request of a then-girlfriend who wanted to make a gift of it for a close relative. It was in truly terrible shape, so I didn't feel too bad about the possibility of making things worse by being almost completely inexperienced with wood restoration. The speaker was shredded along with the grill cloth, all the electronics were covered in a thick layer of grease and dust and rust, and it had been painted. Repeatedly.

I used a "green" paint stripper with a plastic scraper, and discovered three layers of paint on top of the original finish: glossy black, glossy red, and flat black. I had to remove the original finish as well, which was the most resilient of the coatings.

After many sessions of scraping and scrubbing, the radio was down to bare wood again. After carefully re-gluing some areas of the veneer that had delaminated, I used a number of stains and varnish to make it all pretty again. In the end I did not end up restoring the radio to functionality, as that was less important for my friend. I did still replace the speaker cone and clean up the electronics as much as I could without complete disassembly. 

Everyone was pleased with the result in the end, though I know I could have done better. Perhaps there will be a next time..

Power Supplies for Plate and Filament of Homemade Triode

After revisiting the ongoing project of a crude home made thermionic valve or triode vacuum tube, I decided to dispense with the old Variac-controlled adjustable 60Hz power supplies. I made a new high voltage supply using a simple flyback converter topology, followed by a high voltage linear voltage regulator made from discrete components.

The low voltage supply is still 60Hz, but it is also followed with an adjustable regulator, in this case a LM350, a 3 amp device.

The high voltage is run through a Bogen T725, with a rotary switch selecting an output tap for audio output. This can be seen in the upper left corner. The low and high voltage supplies each have their own electrical cord so I don't have to use both at the same time. I suppose switches would have worked, too.

The power supplies got their first workout in an experiment to see if I could use a silicone rubber gasket in place of the usual RTV for sealing the glass envelope of the triode. It did not work well, but that is in part due to the irregular shape of the sealing surfaces. The filament's life was short and smoky, and I was unable to receive any signals. I completely rebuilt the regenerative radio set for these experiments, so I do hope I didn't miss-wire something. It could also be the fact that I'm attempting to use grid-leak biasing this time around, whereas in the old setup I used a separate variable bias supply for the grid (+/- 17Vdc IIRC).

Hysteresis/Induction Motor

This is something I built a while ago, but I wanted to describe it in some more detail. It is a simple alternating current motor that spins a thin disk of metal balanced on a sharpened point of graphite.

The circuit applied roughly equal current to each of the coils, phase shifted by 90 degrees. The large AC capacitor provides both phase shift and current limiting to one coil while the other is simply fed through a high wattage resistor. I strung together 12 smaller resistors in series/parallel and even attached heat sinks with thermal grease to get a suitable dissipation power and reasonable resistance of 17.5 ohms.

The two coils act together to produce a rotating magnetic field, as one always precedes the other by 90 degrees. Four coils would produce a much more symmetrical field, but two will do the job. As the steel mason jar lid is magnetized by the increasing field from L1, that part of the steel becomes attracted to the simultaneously falling field from L2, and gets pulled around toward L2. Much the same way, when a rising field near L2 magnetizes the steel that same part of the steel is repelled by the falling field from L1, and pushes the disc from L1 toward L2. This might not be all that is going on, but it suffices for me as an explanation of why this simple induction motor functions at all.