The Televisor Electronics
The replica DE Televisor, like the original, is built on a wooden base and connects to the "outside world" via two wire pairs. The first of these connects to the "neon" lamp and, in the original, would have been connected to the plate circuit of the final audio amplifier tube/valve. The second pair is for the motor drive and would normally have been connected to the mains supply. Motor/disc speed control was achieved by varying the voltage applied to the motor using a high-wattage wire-wound potentiometer (the large knob) and a series power resistor wound on a porcelain/ceramic coil form (the vertical white tube on the left). Both of these components are non-functional in the replica kit.
The kit comes complete with drive electronics that will process the video signal to drive the LED replica "neon" tube and a phase-locked motor drive to lock into the proper speed/sync and phasing to display the picture.
Video Driver
Back in 1934, the neon tube of the televisor would have been driven either from the plate circuit of the broadcast receiver or via a dedicated amplifier chain made up is resistively-coupled stages. While IC-based video drivers are far beyond the expectations of a 1934 enthusiast, I was not about to replicated a vacuum tube/valve amplifier. What comes close, in spirit, is a simple transistorized video driver designed by Klaas Robers:
A simple little circuit board was designed to accommodate a very slight variation of this circuit. First, the BC548 transistors were replaced by 2N4401s. Secondly, the 10K BLACK SET control was located on the front panel of Klaas' unit, where it probably should be if you are going to use a wide range of video sources. Since I intended to primarily use my NBTV Romscanner and the NBTV Association CDs, I opted for a PC-mounted pot on the circuit board.
Motor Speed Control
The sophisticated motor drive circuits are problematic from the perspective of authenticity of function. High-end televisors of the period used a variety of techniques to lock the motor speed with the incoming image, along with provisions to phase the picture. However, the DE Televisor had no such provisions and relied on manual control of the motor speed via a series rheostat. My intent is to use a pulse-width-modulated controller for manual adjustment of the motor/disc speed. This will make viewing and photography of the display very challenging, but the operation of the unit will be completely authentic and, all things considered, still easier then the challenges faced by a DE viewer in 1934!
I am using the DC power supply portion of the DE kit electronics, but control of the motor/disc speed uses a Carl's Electronics DC Electronic Speed Control circuit module:
Kits like this are universally available, typically using an NE556 dual timer to pulse-width modulate a power FET. The complete Instruction Manual for the kit, including a schematic, is available in PDF format by clicking the preceding link. In the case of the kit I received, the speed adjustment is achieved using the 10K potentiometer that you can see at the right end of the circuit board illustrated above. While this would probably be equivalent or somewhat better than a 1930 rheostat speed control, adjustment would be super-critical. To give myself a fighting chance to take pictures of the display, the 10K single-turn pot supplied in the kit was replaced by a 12-turn 10K PC-mounting pot with a 500 ohm 10-turn panel mounting pot in series. The strategy is to set the panel-mounting pot to mid-range and use the 10K pot to roughly set the disc speed. The speed can then be fine-tuned using the front-panel control. More on this on a later page.
Motor Speed Calibration
Speed control with the original Daily Express televisor consisted of adjusting the rheostat speed control until something recognizable was glimpsed on the display. This itself would have been a minor miracle given all the other unknowns faced by the stalwart television enthusiast. Of course, to display anything recognizable, the televisor has to be running VERY close to the nominal 12.5 revolutions/sec or 750 RPM (12.5 x 60). If you have no way of measuring disc speed, getting into the desired "zone" can be a long and tedious business. Using a strobe helps things a bit, but here in the 60 Hz colonies, it would take a crystal-controlled LED light source. That is easy enough to arrange, but doesn't provide a really big jump in functionality.
My solution is a bit "over-the-top", but should work really well. In the past, I have done a lot of work with an imbedded microprocessor module called the DOMINO. This little unit is programmed in a comprehensive version of BASIC, can do floating point math, and make very accurate frequency measurements. Simply put, the DOMINO will be programmed to make continuous frequency measurements from an optical sensor at the disc, calculate the conversion from frequency to the disc speed, and print out the instantaneous speed on a small LCD. If you have to manually adjust the disc speed, this is an ideal approach, since you always know if you are slow, fast, or right on!
Printed Circuit Board
The first version of the layout for a printed circuit board for the video driver and Domino circuits has been completed:
The Domino microprocessor module is located on the left end of the board. If a builder elects not to include the disc speed display, the board can be sheared at the dotted line to create a smaller board footprint. The first boards are now being fabricated (01 April 08). Once everything has been throughly tested, board layout files will be available on-line for those who would like to duplicate the project.
The following page links will be activated as the final electronic construction gets underway:
BUILDING THE VIDEO DRIVER/SPEED DISPLAY BOARD
TESTING
RESULTS
PROJECT MANUAL
UNDER CONSTRUCTION