Si-diodes in Second World War radar & Communication equipment

Smart-ass ping-pong plonked... no redeeming social value ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
         America: Land of the Free, Because of the Brave

You mean facts, as opposed to second-hand rumors you think you remember? Guilty as charged.

John

--
"and Maybe"

"Could have been"

"Or maybe"

are facts?

Maybe in your fantasy world, but to me it all sounds like conjecture.

The _fact_ is I originally commented to the OP that his quest might be
made easier by considering the 1N23 as a starting point for silicon
detectors, but then amended my original comment to reflect that the
first time I had contact with them was in the \'50\'s, when I thought it
was new.

I guess, according to you, I was right in the first place.

Oh, well... 

I guess your way is to clam up if you think you\'ve made a mistake
unless someone calls you on it.

That\'s not my way.  If I think I\'ve made a mistake I own up to it as
soon as it looks questionable to me without waiting for anyone to prod
me into forced admission.

Your mileage certainly seems to vary.
   

BTW, for some reason I\'m prohibited from accessing the Rad Lab series
via:

http://www.jlab.org/ir/MITSeries.html

so would you be so kind as to post the text referring explicitly to
the 1N23 and relevant deployment dating?

Thank you ever so much...

JF

Certainly not. Those were conjectures on why you might have thought that these parts were new in the 1950's. As if it matters.

Those words do usually imply conjecture, yes.

If it's so important to you to be right, you might consider checking facts before posting. But do whatever's fun.

Nope, too much work. The books are widely available.

John

--
Certainly seems to matter enough to you to keep posting about it.

Jlab (used to be CEBAF) has a 1/4 mile racetrack electron accelerator, pumped by klystrons driving cool shiny superconductive cavities with megavolt-per-meter fields and Q's like 1e8 or something. Their site has some interesting stuff. We did the electronics that measures all the liquid helium temperatures and levels, and the microsteppers that tume the cavities. That was probably the last major CAMAC installation anywhere.

John

--
Right. Thanks, :-)


JF

Did they even _have_ silicon diodes in WWII? I remember when they announced the first transistor, some time in the early 1950's.

Thanks, Rich

Very much so. Point contact types.

Graham

That's the whole point. The poster hasn't gone to the sources, and decades later the details are garbled.

I pulled out two ARRL books from the early seventies, and they had contradictory information on at least the 1N21 and I think the

1N23, a data table in the Handbook listing it/them as germanium, the VHF manual saying it/them as silicon.

The earliest magazines I have are some QSTs from 1947, and I can't find them at the moment. I know I saw some early ads for semiconductor diodes in QST, and if it wasn't 1947 then the next eariest issues I have are 1953. The fact that the ad I remember seeing just a few years ago did have the 1N34 sort of blows out the comment in this thread that the "1N" numbering didn't came till the late fifties. Indeed, one reason we know of early semiconductor diodes is that there was surplus after the war and we saw them, or at least read about them. I always thought the APX-6 was WWII surplus, and it used diodes in the "1N" series; go back and find the conversion articles and that might give insight (though of course, if the information was garbled at the beginning, then such later sources won't be reliable).

But that sort of source is going to be far more definitive than our memories, of the time or of what we read years ago.

Even that book about radar development in WWII, which did seem fairly "sane" technically, it's not a technical book so even if I could remember what it said about diode development, it might not necessarily be accurate.

Michael

Yup. Most of the WWII radar diodes were silicon point-contact types, Schottky diodes actually. The best 1943-vintage mixer parts were about as good as any packaged schottky you can buy today... 0.2 Vf, 0.2 pF, decent noise figures to 30 GHz.

The point-contact transistor was invented at Bell Labs in 1947. Most of the relevant semiconductor theory - bandgaps, hole/electron conduction, doping - was well understood by about 1940. The RadLab guys didn't develop a PN-junction diode or the transistor because their mandate was to develop radar to win the war.

John

I'd question that "was well understood" part. The description in the Buderi book makes it pretty clear that before 1940, people working with semiconductors (key being at Bell Labs) didn't have a very deep understanding of what was going on. It was only in late '39 and 40 that they got serious ideas that they could actually control what was an essentially empirically-understood phenomenon by changing the amount and type of impurity. The description of things going on then as "increasingly curious properties" of silicon doesn't seem to fit very well with "well understood." But maybe Buderi didn't do a very good job documenting that particular work, and missed the depth to which the phenomena were understood.

Cheers, Tom

Artech has never been cheap. :-) At this price, I think they figure they're appealing mainly to University technical libraries and perhaps some larger companies' internal corporate libraries... or perhaps they're thinking the market for such historical documents is quite limited, thus "necessitating" the higher price to cover their costs?

Just checking the footnotes in the radlab book, it looks like most of the serious theorizing (ie, stuff that worked) was published between

1939 and 1942, "about 1940" by my standards. Potential barrier diagrams and Fermi levels and such were in books published in 1940. Mott and Schottky seem to have published the first non-silly diode theory stuff (non-backwards!) in 1939 and 1940. This got a lot more serious between 1940 and 1943 as MIT poured in money and talent.

John

--
John, I must say you\'re simply amazing!

Being able to postdict the butterfly effect is a gift few of us have.

JF

--
Talent, maybe, but where do you think the money was coming from?

Hardly MIT, if their mandate was, as you state:

..."to develop radar to win the war." 

JF

V(f) @ 1 mA. Result < 180 mV. Thus Ge, not Si.

Certainly. 1N23s that i had were Ge also. Lost them on some move.

Gee, John. Where do you get schottky diodes with V(f) below 0.2 V at I(f) of 1 mA? All the ones i could find were over 0.33 V and mostly

0.4 to 0.5 V.