Distributed Amplifier

Search on the 'travelling wave tube" - TWT

I see it as a particular kind of distributed amplifier - as invented by Bill Percival at EMI Central Research in 1936 - but I'm biased having met (and liked) Bill Percival when I was working at EMI Central Research much later (1976-79).

Someone once patented a distributed-amp tube, about like you'd expect. And I vaguely recall a secondary-emission version of same.

Lots of people now make wideband monolithic GaAs distributed amps. Some go LF to 40 GHz or so. They are commonly used to drive lithium niobate electro-optical modulators, swings like 5 volts p-p or so.

We've use the Hittite 20 GHz parts to drive EOMs; worked fine.

The old Tek 545 oscilloscope used a tube distributed amp in the vertical section. It was outrageous.

I have a couple of them around here still, as souvenirs.

Weird, the output line drops impedance as it goes. That'll keep a stiff load for the transistors, but each transition wastes a little bit reflecting it back the other way; the input lines don't do that. Hmm, cascode too I guess, but what's the significance of the ~1/4 wave TL between transistor pairs? Peaking? Otherwise, doesn't seem so smart... Also appears to be thick film, given the odd colors.

Is there any real power in a hunk of LiNbO3 or is it just a capacitor?

Oh, they used them constantly. The epitome of tube tech was their 80MHz masterpiece (I forget how hybrid it was, or what number 5xx exactly).

Tim

Were you drunk when you wrote this?

amp.jpg

Funny. The OP reminded me for some reason of an old analog storage scope tube. In fact i have a Tek 466 sitting around somewhere awaiting repair. I just love the look of those old things. I foolishly got impatient when powering it up having bought it second hand from some charity shop. It probably hadn't been used in over 20 years so I supplied it via a variac and wound it up slowly - but not slowly enough. Got the old magic smoke emission at 2/3 full voltage and suspect an EHT paper capacitor to be the culprit. I'll tear it down for a look see next year. Does anyone remember the tek 466 and if so, what did they think of it, I wonder?

Oh, so you have actually done *some* work at least in your life then, Bill? ;->

I have one too, which worked fine when I got it but failed the instant I tried turning on the analogue storage feature.

I used a (then reasonably new) 468 back in the early 80s. It worked very well except that when you had the persistence cranked up too high the trace and the background gradually turned to uniform gray-green.

Cheers

Phil Hobbs

They usually look like 50 ohms. I think they are actually built as transmission lines, input on one end and a 50 ohm terminator on the other.

The 580/585. The CRT had distributed deflection plates, too, similar to this one:

The old secondary-emission storage things were awful, and aged badly. The HP mesh storage was really good, but lower writing rate. HP and Tek had cross-licencing deals, and I think Tek may have done a mesh storage scope, too.

Analog scopes were cool, but I wouldn't want to use one these days. I've got used to color, deep storage, signal averaging, all that neat digital stuff. Plus being able to lift a scope with one hand.

Yup. The travelling wave design is why you can get pi radians of retardation with only a few volts of signal. Normal Pockels cells take about 3 kV.

Cheers

Phil Hobbs

He was the janitor! You get to meet and learn how to recite things that sound intelligent, even if he has absolutely no idea what they are talking about.

Jamie

by

rch

Two of my patents date from my time at EMI. Their legal department was rema rkably receptive to "patent queries" and one of our less discriminating eng ineers held the lab record for the most patent queries submitted in one yea r, but the janitors didn't figure largely.

Here's one of my patents - for which I had submitted my only patent query, after fielding all sorts responses to what I had thought had been a stateme nt of the obvious, which clearly proved that idea wasn't obvious to those s killed in the art.

U.K. patent 2028503 "Improvements in or relating to ultrasonic apparatus"; assigned to EMI Ltd in 1978.

The other patent arose because my name had been tacked onto one of my coll eagues patents because the patent happened to cover a theoretical digital s olution the problem of interest (that I'd worked out and simulated) while t he bulk of the patent covered the practical analog solution that my colleag ue had not only invented, but also reduced to very successful practice.

Quite a lot. There are even three patents with my name on. My problem is th at I can't get work now - I'm 72, which doesn't stop me from applying for t he occasional job that asks for analog, digital and mixed-signal experience , but does make it unlikely that I'll be preferred over a less experienced but younger candidate.

Am 01.12.2014 um 09:50 schrieb josephkk:

A friend of mine works in a TWT factory here in southern Germany and so it happened that I got a guided tour. Precision mechanics everywhere, eroding machines, waste bins for metals you only know from the periodic system, melting glass, vacuum until you think all that had nothing to do with electronics.

Then you come to some rooms with arrays of these absolutely high end vector network analyzers, the shabbiest ones with

regards, Gerhard

get

(well,

(roughly)

so

Mark ----------------

the

sucks

The last paragraph sounds a lot like a Traveling Wave Tube. Still made and used. But nowhere near like they used to be.

?-)

Heh heh. :) Most of those elements should be just molybdenum and tungsten though, no? And nickel, but that's really pretty common stuff.

They're still flying TWTs on satellites, aren't they? I don't think I ever heard what they've got, say, orbiting Mars, if they've gone solid state at last or what.

Tim