Radar Tubes

For my first entry I'll put in my latest CRT project. I have a 7APB7A tube which I pulled from a Furuno radar head when I was about 12 years old. I've kept this tube around for all that time not really thinking of what I'd like to do with it.

This tube uses magnetic deflection and so is harder to use for experimentation since I didn't even have a yoke that fits it. My interest was piqued when a friend gave me a tube assembly from a B&K TV Analyst. These B&K units use a special ultraviolet tube and a magnetic deflection yoke. The yoke is of the correct diameter to fit my radar tube so for now I've borrowed it from the flying spot tube.

The P7 phosphor used in this tube is a long persistence type; this means after a trace is drawn on the screen there will be a slowly-fading image left behind. The choice of this phosphor for radar is due to the fact that radar scans a line-of-sight path which is rotating on a plane. At any given time there is a single trace being drawn from the center of the tube to its edge which has modulations of beam intensity to display landforms and objects in the path of the radar antenna's beam. This line is swept around the screen as the radar antenna rotates. The slow phosphor retains the image formed by the sweeping line long enough for the user to view the complete picture before the next trace is made. If a fast phosphor like that in a TV were used, the radar antenna would have to rotate at extreme speed in order to update the picture quickly enough for continuous viewing.

I have no idea ultimately what I'd like to display on this tube but I figure I'd start by building the circuitry around it to create a vector display. Such a display can accept any sort of voltage inputs to move the beam around on the tube. These inputs are known as X and Y. A third input, Z, is used to modulate the beam intensity so that it can be made brighter or dimmer (or even completely blank).

The first hurdle in getting any picture tube working is to set up a power supply for it. Bigger magnetic deflection tubes like this radar tube require a rather high voltage supply to provide sufficient acceleration of the electron beam in the tube to make a bright sharp spot on the face (about 10,000V in my case). This type of setup is known as post-deflection acceleration as it accelerates the beam after it has been deflected. The beauty of post-deflection acceleration is that it allows good deflection sensitivity while still giving excellent brightness and sharpness. If one were to accelerate the beam so greatly before trying to bend it, one would find it harder to bend as it spends less time in the deflection area and thus is less sensitive to its effects. Deflection sensitivity is very important in CRTs because making an amplifier fast enough to drive the deflection magnets or plates is hard enough on its own let alone trying to make it output more and more voltage.

I found in my junk box (more like junk wing-on-the-house) a nice pre-packaged high voltage supply which came from a 21" monochrome computer monitor. This supply probably develops about 15kV or so and can be used with my tube. Conveniently it also has a 400V output for supplying screen grids, focus anodes, grid voltage-dividers, etc with their own power supply. Looking up the specifications on the 7ABP7A revealed that for a typical 10kV post-accelerator voltage, the screen grid should be 300V more positive than the cathode while the control grid be about 20 volts below. The focus anode in this tube is unusual in that it is designed to operate at or near cathode voltage in order to effect reasonably automatic focus without the requirement for user-adjustment. I still added a control to allow very fine adjustment because I found that connecting the focus anode to the cathode didn't give optimal sharpness for all brightness settings.

The next portion of the power supply required for any CRT is the heater supply. This works just about like any vacuum tube: a heater must heat the cathode before it can boil off enough electrons to allow sufficient current for useful operation. Most vintage tubes use either 6.3V at 600mA or 2.5V at 1.5A. Since I have lots of tube-related parts around I just used a Hammond 6.3V heater transformer.

With the above in place I now have a working cathode ray tube. I can make it display a spot in the middle of the screen but I cannot move the spot about the screen. This is where I slip the yoke from my flying spot tube onto the radar tube. To drive the yoke linearly we need a very high impedance signal source; something that outputs a controlled amount of current in proportion to the amount of deflection desired. This is because magnetic field strength is directly proportional to the product of current and the number of turns in the electromagnet. If one were to use a voltage-control output instead, one would find that for any frequency above DC (0Hz) the current and thus the amount of deflection would be reduced as frequency increased. This reduction in current is due to the fact that the magnetic deflection coil is also a very good inductor and it is known that inductive reactance increases with frequency. For any constant voltage, an increase of reactance has the same effect as increased resistance in terms of total current in the circuit. This is shown easily in ohm's law: I = V/R where I is current, V is voltage, and R is resistance; try putting reactance in place of resistance and the same law applies for current if you don't care about phase (phase is beyond the scope of this discussion).

To effect a current-regulated driver for the deflection yoke, a device known as a transconductance amplifier is used. This amplifier uses a voltage input to control its output current. Most amplifiers people are familiar with simply develop an output voltage but it is also possible to make current the controlled factor instead. I will be adding a section on my transconductance amplifier designs shortly since I also use this for driving speakers in various special ways. Let's just be happy for now with the fact that we have a black box which drives the deflection yoke very nicely based on whatever input signal we give it.

Now we have a working tube with deflection capability. Time to give it some signals to work with. One of the most fundamental things to try is to make Lissajous figures. A Lissajous figure is generated when two separate sine waves are applied to the X and Y inputs respectively. As seen in the picture below a 5:4 Lissajous curve is displayed. This means that the ratio of frequencies applied to the X and Y axes is 5:4; for every four oscillations on the Y axis there are five on the X axis.

Below is an example of using my tube as an oscilloscope. I've got a sawtooth wave driving the X axis so that it scans smoothly from left to right and then quickly jumps back to the beginning for another scan. A proper scope blanks the trace on the way back (known as retrace blanking) but I just wanted to get an example going quickly.

Next is a great example of what a horrible mess I make when I'm testing out new ideas on my bench. If you think this looks disorganized and messy then you haven't seen too many engineer's electronics benches!

I will follow this project up once I've built an enclosure for the tube and the associated electronics to make it into a nicely packaged X-Y display. Information on the transconductance amplifiers will appear eventually as well!

Previous page: CRT Fun
Next page: Flying Volt Meter