[snip]
Elaborating: The above is fine for two fixed references. For a fixed hysteresis about whatever reference the user wants, the current diversion trick still works, but you play sizing games with the two input diff pairs to create the hysteresis... likely what Microchip does. ...Jim Thompson
What's funny is, a particularly small SuperJunction FET (might not've been available at that moment?) would probably get there. BUT, only above 20V. You'd have to arrange some sort of over-the-top Baker clamp to prevent it from getting suck in the mud.
Tim
(a play on the resistors-approaching-an-IC picture you may've seen before)
Tim
My copy of RC-20 is dated Dec 1960, but there certainly could have been earlier printings with the same cover.
Some of the most interesting -- functionally speaking -- included CRTs with funny stuff on the face end. Cutouts for test patterns. Binary (or gray code) patterns (ADCs up to 12 bits, in as many microseconds as you can sweep the beam!). Scan converters -- between video standards, and for stranger purposes like vector to raster conversion.
Even high speed waveform capture: Tektronix had an oddball 7000 series mainframe where the MCP (microchannel plate) face experienced an inverse storage effect. The high speed "write" beam would zip through in a few nanoseconds, blasting electrons out of the (poorly conductive but highly secondary-emissive) MCP layer, which would momentarily deactivate the written area. The write itself would produce a visible trace, but not a lasting image. But because the MCP gets deactivated in that area, the plate current multiplication can be measured, and therefore the image reconstructed. So, a raster or feedback tracking scan, immediately following the triggered event, would recover the waveform, which can be stored digitally using very modest ADCs. I believe the claim was the first >= 1GHz BW, single-shot capable, DSO.
Tektronix made all their own glass (and a vast amount of ceramic, especially for the most precision CRT parts). They were *damned* good at what they did, at least at their peak of knowledge somewhere around then (i.e., early to mid 80s I think?).
To this day, it's still slightly cheaper, or at least can be more junkbox-friendly, to slap together some old-stock tubes for a SW linear than to go out and buy an RF part. Especially in the UHF+ range where all the LDMOS still remains obscenely expensive. I had bought some of the cheaper ones, PD57006-E, which seem to be little more than a 2N7000 with heatsink tab (presumably, a little better optimized die as well, and slightly larger for actually equipping that heatsink tab).
That such a comparison can still be made, in this 21st of centuries, seems strange. Surely if I can purchase 10 gigaflops of computing power for the same price, there is something missing.
Flipside of that being, I recall a few years ago, some articles were going around, bragging about how we'll be doing away with magnetrons! It was laughable back then, too... :)
Not through any fault of their own, of course.
I choose to view only excellent material on mine*; which is peak-of-its-day Trinitron technology. One of these days I need to look for a 2048x1536 model. Or two. And an extra air conditioning unit...
*Which begs the question, what is S.E.D. doing here. :^)Tim
Early digital telephone systems used a single-tube companding ADC. An electron sheet was deflected by the audio signal and hit a grey-coded anode pattern. Pretty neat.
We have a 7104 (on a cart!) and occasionally still use it.
Tek sold a French-built scan converter scope (double-ended CRT) with 7 GHz single-shot bandwidth. You can still buy it!
(Nice logo, huh?)
The short ceramic mesh CRTs just weren't as pretty as the old, long glass tubes.
That is fun.
Of course men were men back in those days. When I was a kid, I recall seeing an article in (iirc) Electronics Illustrated about using a huge synchronized colour filter wheel (all controlled by tubes) to colourize a B&W TV. I can't imagine the spousal explosion that would have resulted from some guy trying to put that in the family room. A Fred Flintstone type would never have gotten away with it.
That sounds like a Lithicon. When I was in grad school, we had a Lithicon scan converter, which you could scribble on in XY mode any way you liked, and it read it out in RS-170 video.
That was replaced by a giant, rack-mounted, wire-wrapped Grinnell image processor system. (It was one of the ugliest things I ever saw--the rack was this absolutely bilious yellow colour.)
Cheers
Phil Hobbs
On a sunny day (Sat, 18 Apr 2015 15:58:47 -0500) it happened "Tim Williams" wrote in :
I used to play with this sort of stuff:
We had plumbicon cameras, I designed and build my own vidicon camera, and in one studio U worked there still were image orthicons in use:
Never used the iconoscope AFAIK.
Happened to spot the RCA Receiving Tube Manual in the book pile...
...Jim Thompson
-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: skypeanalog | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at
Yessir. That's the copy I was working from, before switching to trannies in 1963-4. What a grand time!
But I struggled to understand how to effectively design with transistors (looking at you, h-parameters), until a 30-minute session with Carver Mead (Cal Tech), who demonstrated the beauty and grand world of Ebers-Moll.
No tubes since 1963. Except for 20kV, 15A beauties, seriously discussed in WW-2 submarine patrol reports. E.g., 6C21, 6D21 and 715B. Three in parallel for 1MW. Newer tubes like the 4PR60B pushed pulse power to 1MW. But, "six 527 triodes as a switch... output of 10MW, 25kV at 400A, requiring a grid-driving power of 700kW." Whoa, trannies need not apply! But I did manage to reach the 100kW level with stock Fairchild MOSFETs.
One of the unforgettable moments in my life (we're excluding girls here) was when I made a curve tracer out of my Knight Kit oscilloscope and connected it to a transistor. Revelations.
The RadLab books are cool, especially vols 5 and 6, "Pulse Generators" and "Microwave Magnetrons." They were getting megawatts of RF at 3,
10, and even 30 GHz by the end of the war.
The RCA book appears to be 1960.
The other:
These actually *sold*!
Maybe this is speculative, a "tube bubble."
Somebody at Svetlana should start making copies. They'd have to be careful not to make them too much better than the originals, or even phools would become suspicious. ;)
Cheers
Phil Hobbs
Have something to add? Share your thoughts — no account required.
Ask the community — no account required