30-line Televisor

Home-made Televisior
Home-made Televisor

This was my second Televisor that I built soon after (possibly even during) the construction of my first Televisor. This one can display a 30-line Baird signal (with included sync pulses). It is also unique that it uses neon indicator lamps in place of LEDs.

Model "A" Televisor
Model "A" Televisor

My version was modeled after the Baird Model "A" portable Televisor, shown here from the front, and the rear with the back cover removed. The single knob in the front adjusted the motor speed, the framing was set using the knob in the back just above the motor. The picture appeared in the small rectangular opening. 

Wireless demonstration of television

The Model "A" in action.

The Neon Lamp

From the outset of this project I wanted to use an actual neon glow lamp instead of LEDs to keep with the time period, but immediately ran into design problems. Neon glow lamps with an electrode surface area large enough to be viewed directly behind a Nipkow disc only existed because they were made for television, and then only in limited quantities. Any originals that survive are more museum pieces than experimental items. Therefore I had to either find a suitable substitute or use an array of tiny bulbs. "Aerolux" style lamps were the first thing to come to mind. These were simply glow lamps with electrodes shaped into objects or designs that were popular novelties in years past. eBay was filled with them, but none had the large, smooth electrodes required for my purpose. Many of them were also housed in larger "globe" type glass envelopes, meaning that the electrodes would be located some distance from the disc, making it appear smaller. The second option was to create a light box, like the kind used for LEDs in modern Televisors, but instead using small neon indicator bulbs. These are the same indicator bulbs still being made for use in electric stoves, toaster ovens and irons, although they are gradually being replaced by LEDs. However, I found out that not all indicator bulbs are the same! Most bulk bulbs with actual specifications attached to them were quoted with minuscule currents, some rated for only 0.3 milliamps at 120v. 

During my search I also came across some new old stock INS-1 bulbs, made in the Soviet Union. These were used as dot indicators for Nixie displays, and were attractive to me for their built-in light focusing lens. The electrodes consist of a small cylindrical cathode and a larger sleeve-shaped anode. I bought some and quickly threw together a simple 4 x 8 matrix out of cardboard. Unfortunately, although they looked nice, they just weren't bright enough to use on their own when placed behind a suitable diffuser. Soon after that I came across some more conventional NE-2-style high-brightness bulbs online.

Daven Television Lamp

Daven Television Lamp at the ETF.

Neon Television Lamp

Similar television lamp energized.

INS-1 neon bulb

INS-1 bulb.

An NE-2 lamp operated by DC and AC current. By Ceinturion 17:47, 19 August 2007 (UTC) (Own work) [GFDL (http://www.gnu.org/copyleft/fdl.html) or CC BY-SA 3.0 (http://creativecommons.org/licenses/by-sa/3.0)], via Wikimedia Commons

These standard-style bulbs were actually preferred for my application for an unusual reason. Typically, when one of these bulbs are lit up, both poles appear to glow. That's because the lamp is running on AC current. In reality, only one electrode (the cathode) glows at any given time, but since the polarity of the current is switching 50 or 60 times per second, both poles appear to glow simultaneously. When run on a DC power source, only one pole will glow. Neon lamps will run fine on DC, but after a while a strange phenomenon called "cathode poisoning" takes place. This is related to metal sputtering and causes the lamp's striking voltage to rise when that particular polarity is used. If allowed to continue, parts of the cathode will stop glowing and eventually the lamp will fail to start. Most of these NE-2 style lamps have striking voltages of about 90 VDC. As I live in an area with 110 V mains, that doesn't leave much head room for an optically shielded bulb (more on that below) running on higher-than-normal current to strike reliably. Fortunately, this effect can be reversed to a certain degree by swapping the polarity manually via a DPDT (dual-pole, dual-throw) switch on the front panel. However, reversing polarities also reverses which electrode produces the glow, so, in the case of INS-1 bulbs and many DC powered lamps, the ring-shaped electrode (normally the anode) would be producing light rather than the disc-shaped electrode, reducing efficiency. NE-2 bulbs have a symmetrical electrode design, so theoretically, when placed in a light box behind a diffuser, switching polarities should not appreciably effect the light output.

home made neon lamp

When combined with the INS-1s, the box generated an adequate amount of light, although I had to overdrive them to get there. This extra current made both the bulbs and the resistors hot, so I stuck an old computer fan on the top for safety (the entire housing is made of cardboard). Even with the fan, the whole thing is dangerous and will have to be rebuilt at some point.

There were still other factors to consider, one of which is the "dark effect". Lamps started in darkness require a higher starting voltage than they would otherwise need in the light. The reason for this is because light (particularly in the blue to UV region) helps jump start the ionization process, called photo-ionization. In fact, in some lamps, a minuscule amount of radioactive material is added to combat this effect. Another approach would be to include a constantly energized blue or UV LED in the light box with the bulbs. An orange filter would be needed to remove the color cast from the LED.

Neon bulb array with diffuser and top cover removed.

With all that said, the rest of the machine is similar to my previous 32-line Televisor and is rather unremarkable. This disc was hand made using a process similar to that described in How to Make a Nipkow Disc. This version does not use graduated scanning, so the picture is slightly wider than it should be. The small DC motor is a bit unstable (I think the brushes are wearing out), so the picture bounces slightly, but this is a problem shared by many original Televisors and only adds to the authenticity.

30-line image

A digitally-generated portrait of John Logie Baird viewed in subdued daylight with original as viewed on a computer.

30-line image

NBTV "explode" pattern with original file for comparison.