Page 86 - Analog Circuit Design Art, Science, and Personalities

Bob Pease

9. The Story of the P2-The First

Successful Solid-state Operational

Amplifier with Picoampere Input Currents

First. let us start with-

A Fable
Once upon a time there were two wizards who decided they wanted to play
golf. The first wizard stepped up to the tee, addressed the ball, and drove the ball
right down the middle of the fairway; the ball then bounced twice, and rolled, and
rolled, and rolled, and rolled right into the cup. The second wizard looked at the
first wim-d. Then he stepped up to the ball and drove a wicked screaming slice
off to the right, which hit a tree. bounced back toward the green, ricocheted off a
rock. and plopped into the cup. The first wizard turned to the second wizard and
said. “Okay, now let’s play golf.”

End of Fable

Once upon a tirne, back in the ancient days of the electronics art, about 19.58, there
were two wizards, George A. Philbrick and Robert H. Malter, and they enjoyed
designing operational amplifiers. in those days, that’s what they called them-oper-
ational amplifiers, not “op-amps.” George had the idea to use some of those new
“transistors” to amplify the error signal from a balanced bridge. up to a good level
where it could then be demodulated and amplified some more and used to form an
operational amplifier. Ah: but what kind of balanced bridge would this be? A ring
of conducting diodes? Heavens, no-George proposed a bridge made of 100-pF
varactor diodes, so that when the bridge was driven with perhaps 100 mV of RF
drive, the diodes would not really conduct very much and would still look like a
high impedance-perhaps 10,000 Mil. Then just a few millivolts of DC signal
could imbalance the bridgc and permit many microvolts of radio frequency signals
to be fed to the AC amplifier. Now, back in 19.58, just about the only available tran-
sistors at any reasonable price were leaky germaniums, and you certainly could not
build a decent operational amplifier out of those. But George got some of the new
2N344 “drift” transistors that still had some decent current gain at 5 Mcps. He ran
his oscillator at 5 Mcps, and after running his signal through the whole path of the
modulator and then four stages of 2N344 RF amplifier. and a demodulator, he fed it
into B DC amplifier stage with push-pull drive to a class AB output. And it was all
built as a quasi-cordwood assembly, with seven or eight little PC (Printed Circuit)
boards strung between two long PC boards. Since each little PC board had about six
wires connecting into the long PC boards, this was a kluge of a wry high order, and
not Tun to assemble or test, or to evaluate, or to experiment with, or to troubleshoot.
George called this amplifier the P7. Please refer to the schematic in Figure 9-1. I
know this is the right schematic because I still have a P7. Also, see the photograph
oi’a P7’s inner workings in Figure 9-2.

67