Electronic Design

Bob's Mailbox

Dear Bob:
I really enjoyed your article on Loebe Julie (Electronic Design, May 3, p. 100). It made me think of the dc-coupled vacuum-tube circuits I designed at a major aerospace manufacturer between 1983 and 1987. Tube circuits in the '80s? Why not? I had to design several high-voltage dc power supplies using power tubes as the output device. Since the systems had one large tube, I figured why not build them with tubes from the output device on back to the input?

The first circuit was a 200- to 1000-V, 25-mA shunt regulator. It used an asymmetrical input stage using the triode and pentode halves of a 6AN8. This then drove the 6L6GC output tube. Level shifting between stages was accomplished with a zener diode. The circuit also used several TV-damper diode tubes for the pulse clipper.

Another tube circuit I designed was a 500- to 7000-V, 2-kW shunt regulator. I believe it used a 12AX7 diff amp driving a 12AT7 diff amp. This then drove two paralleled 4-1000A glass tubes. It was quite a sight to see the plates glowing orange during full-power operation. Several of these shunt-regulator chassis were electrically stacked with the top chassis at -21 kV.

The third circuit was one of my most fun projects. It was a 30-kV, 6-A, fast-respond series-pass regulator. I had to have it working in 10 days, as the customer for our product had been told that we had been testing it with a regulated source for the past year; and they were coming for a visit! The circuit had to respond to a load change from 0 to 6 A in 3 µs. This circuit used a 12AT7 input diff amp (or, as they used to call them, a cathode-coupled amplifier) driving a 12BH7 cathode follower that drove the grid of an Eimac 4CW25,000. The reference voltage was provided by VR tubes. It regulated during a 1-ms, 6-amp gridded-TWT pulse with only 50 V of droop and used only a 0.5-µF output capacitor. My rival's competing circuit used a 20-µF output capacitor for a droop of 300 V during the 1 ms pulse. Believe me, you DO NOT want to deal with 20 µF at 30 kV if you can help it.

This project was not without its disasters, however. Due to time constraints, I had to beg, borrow, and use parts already in stock. The 4CW25,000 tube and associated circuitry (filament supply, screen supply, enclosure) was originally a shunt regulator for a TWT collector. In collector service, the tube plate is at ground potential. So, standard industrial water can be used to cool it. But in my application, the plate potential was 5 kV dc with peaks to 35 kV during load arcs. Normally, a deionized-water heat exchanger would have been used in this case, but I didn't have an extra one. I did have access to an oil heat exchanger, though. This was pressed into service. But due to the high viscosity of oil and its low specific heat (as compared to water), it didn't do a good job of cooling the tube.

In my rush to get the circuit running, I glossed over this point and ran it anyway. After an hour of operation, with the system running full-bore, there was a tremendous explosion and smoke. The oil had gotten so hot that the water jacket on the 4CW25,000 had come unbrazed! Luckily, the big tube was in a metal enclosure or my very capable technician, Bill Fontana, would have been boiled in oil.

This wasn't the first explosion this project suffered. As I said before, due to time constraints I had to cut corners. I hastily threw together a prototype regulator that involved hanging 50 series-connected, 300-V, 5-W zener diodes, by string, from the ceiling. Our TWT consultant, Bernie Vancil (who badly needed this regulator), was present when this particular circuit was fired-up for the first and last time. The power supply (a cube 10 feet on a side and weighing five tons) was brought up and the circuit was regulating nicely.

I was standing about five feet from the string of zener diodes while Bernie stood on the other side. All of a sudden, the zeners vaporized in a terrific flash and boom. The look on Bernie's face changed instantly from one of utter confidence in me to one of zero confidence mixed with shock and horror. The blast was so unexpected and loud that I staggered away to sit down while people poured in from the far reaches of the plant to see what had happened. It was great!

Due to careful attention to detail (excepting the aforementioned lapses in engineering judgement), the three tube circuits lived long and useful lives without the need for servicing.
via e-mail

Dave, do you have any schematics I could see on these? I'm really impressed. I'm not saying that I couldn't design such circuits, but I'm not exactly well-polished in these kinds of designs. Nor do I have the confidence that I could get them to work by the first or second try. Maybe the third try?—RAP

Hi Bob:
Enjoyed your "Julie Stuff" article on the radar-directed anti-aircraft guns. I read The Invention that Changed the World after you mentioned it. I have long been a student of the technology advances of WW II, not to mention over the history of man. I don't think many people ever understood what a fantastic advantage 10-cm radar gave us (BUT that book by Buderi can change that! /rap), which in turn was made possible by the invention of the magnetron. The last day of the V1 attacks Hitler sent 104, and all but four were shot down.

Our skill in code breaking is another story. Of course, all this was top secret at the time. I suppose by the time the whole story was declassified, it was pretty old news.
via e-mail

I agree. Those were amazing stories! Best regards.—RAP

All for now. / Comments invited!
RAP / Robert A. Pease / Engineer
[email protected]—or:

Mail Stop D2597A
National Semiconductor
P.O. Box 58090
Santa Clara, CA 95052-8090

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.