Simple... but of genius...

A long-tailed pair as an input stage... with a single-ended output and, regretfully, without a dynamic load in the cathodes... a resisitve "shifting down" coupling to the second stage (the left section of the second tube)...

The two neon bulbs in series "shift down" the anode voltage variations... then a cathode follower boosts the output... These 680 k and 4.7 M resistors in series to the neon bulbs are a bit confusing... but obviously, the grid current is negligible... like today FETs... The 7.5 pF bypass capacitor shunts (for AC) the high ohmic "moving" network.

But what is the role of the 221 kom resistor connecting the cathodes of the second tube? It "smacks" of a positive feedback increasing the cathode resistor of 9.1 k - another famous idea (cathode degeneration). The 500 pF bypass capacitor shunts (for AC) the relatively high ohmic cathode resistor. Or maybe, the two sections of the second tube form another long-tailed pair? But it seems unlikely...

Another great idea is the ±300 V split supply...

I think there is no more ideas for unraveling... Of course, I could browse through Google to find some explanations... but it is so exciting to find yourself the new idea... especially when you do not know well tubes... and you have to convey basic ideas from transistors and semiconductor op-amps...

In this connection, maybe it would be interesting to compare it with the 741 op-amp internal structure? Oh my God ! Only 20 transistors versus 2 tubes! What a perfect simplicity!

By my opinion this resistor defines the positive feedback (DC) - 500 pF interrupts the feedback for higher frequencies. This solution gives us a high value of DC gain.

It seems it "interrupts" both the negative feedback introduced by the 9.1 k resistor (the cathode degeneration) and the positive feedback introduced by the 221 k resistor?

I estimate that the "higher frequency" will be hundreds of kHz, so the influence of positive feedback is dominant.

Sorry if killing the party:

http://www.philbrickarchive.org/

and expecially 

http://www.philbrickarchive.org/operational_amplifier.htm

I agree with the words "simple" and "genial".

.

All:

' 

In its time, the GAP-K2W was a widely used op-amp.

I have dozens of these (most are luxurriously equipped

with brand-new tubes) in my museum, And yes, the 221k

resistor provides a modicum of positive feedback which

raises the DC gain from pretty miserable to about 1,000.

Apart from the low gain, there was another penalty that

had to be lived with, and that was the power dissipation

of about 10W per amplifier, most of which (about 7,4W)

is simply wasted in heating the cathodes.

.

By modern standards, the K2W was a terrible op-amp.

Yet its ubiquity at the time provided the motivation for

Ray Stata and Matt Lorber to invest their own money

in a start-up, which they soon called Analog Devices.

That was over fifty years ago.

.

The Lone Arranger

Barrie, Indeed , Marvelous to read  this discussion about the K2-W.

For a moment, I thought I was 21 again.

Cyril and Josef are so very technically adept in touching the K2-W. 

I still think that "a long-tailed pair" is a fanciful term ! 

I Had to take a side trip down memory lane

to check the military ARC-5 that was my first Radio Transmitter, 1958.    

Barrie, thank you for the very nice "picture" - and comments.

Josef

Glen, I can not believe you're worked with the K2-W :-))

Josef

Josef,

Accepted !     :-))

Well, I have no such great claim for my experience, as you  may have,

only a familiarity that comes from interest and curiosity. .   

My minor C.V. is published  on my website.   

It Is not nearly so  good as those engineers in today's discussion. 

As a kid, 14, I obtained a full U.S. Amateur Radio License, K4KKQ, 

built up a Ham Radio Station, CW morse code only, 

used a military ARC5 for my first transmitter, 60 Ft tall Vertical antenna, 

and a 1940 Vibroplex Bug. ( Still use the  Bug every day, on the air. ) 

...

Back then, I read about where radio came from,

and then read more intensely in the late 60's on as serious schooling

enabled me to 'spell' the 'words' of electronic ideas I had dreamed of ...  

and   I learned to let  my dreams lead my thinking. 

That first Dance was an  inspiring joy 

...  my last Dances are about the people I associate with

and the way I can share my exeriences of my wonderful "Time" .

the apprentice, Glen  

Dear Glen, we all are apprentices still.

At least I think so (humbly) :-)

Josef

This is one of those circuts that I have been meaning to try out "for real" for quite a while.

I have all the parts, including the valves/tubes, in stock for my home lab. So it's "just" the small matter of finding time to get round to it :)

Once revealed the brilliant ideas behind the legendary K2W tube op-amp, can we do the same with the famous 741 op-amp circuit solution?

I have asked another question as a continuation of the present Josef's question about K2W:

https://www.researchgate.net/post/Can_we_reveal_the_brilliant_ideas_behind_the_741_op-amp_circuit_solution_of_genius

.

NOTE: This thread got redirected at some point,

.

Susan:

.

If I can help in any way in your quest for hollow-

state perfection, please let  me know. I think the

lessons learned in designing CMOS amplifiers

may be of some value; but the big problem here

is that there is no opposite-polarity: type of tube!

.

Nonetheless, it is an interesting question as to

whether one can improve (in ANY measure)

on the K-2W performance, allowing, say, up to

a doubling of power consumption. The heaters

are the worst offenders, but there's a family of

low-power tubes designed for battery radios

which might be useful here, And,  of course

there is no solid justification for using a total

of 600V for the signal path; though one may

be obliged, in the spirit of the venture, to stay

with a +/-100V output.

.

This looks like fun!

.

Barrie