25 Dec
2025
25 Dec
'25
9:58 a.m.
On Dec 23, 2025, at 10:09 PM, Brian L. Stuart
<blstuart(a)bellsouth.net> wrote:
On Mon, Dec 22, 2025 at 01:42:54PM -0500, Paul Koning wrote:
Sure, that makes sense. And there's a fair amount of power involved to produce fast
pulses of significant amplitude.
It reminds me of the CDC 6000 series mainframe console display. That uses what amounts to
oversized oscilloscope tubes -- electrostatic deflection tubes, but 14 inches diameter vs.
the usual 5 inches or so you would find in a typical oscilloscope. We sometimes referred
to them as "radar tubes" but classic radar displays use electromagnetic
deflection (with a rotating yoke).
Since the display rate is pretty fast -- a vector-drawn character takes about 2.5
microseconds, with 100 ns per stroke step -- and the deflection waveforms are at least
several hundred volts, the deflection final stage is a push-pull pair of 3CX100A microwave
transmitter tubes, rated at 100 watts at several hundred MHz. In the DD60 display they
are DC amplifiers, with 2 kV anode voltage. Not the usual environment for a computer
technician to work in.
paul
I glanced at a few of them. Was surprised to see
an inverter using a 6L6. Then looked a little further and came across an element that
uses an 807.
The 6L6 and even more the 807 are rather substantial power tubes. 807s are commonly seen
in amateur radio transmitters in the 1940s and 1950s, good for 50 watts or so power
output. I have a couple, they are very nice tubes. I don't remember the 6L6 ratings;
20 watts, perhaps? Both are also rather large, the 807 especially, when compared to
typical "receiver" tubes.
Good catch. The majority of the 6L6s are used in parallel pairs in
the circuits identified as "transmitters." They are drivers for
transmission lines that can be up to about 80 feet long with taps
about every foot. The pulses they're transitting are about 2
micro-seconds long. So they really are like small RF transmitters
pushing long transmission lines pretty hard.
When I first saw those designs, I thought in terms of over-design
from the context of TTL signals on busses measured in inches. Then
again, we're talking about a 50 volt swing charging up a big long
coax capacitor. So the more I've dug into it, the more respect I
have for the design. In fact, there are also pulse amplifiers
that can be inserted to bump the signals on long lines.