"...The K2-UX is probably the most priceless of all Philbrick artifacts... " from email sent to Dan Sheingold by Joe Sousa, dated 10-16-2011
Dan Sheingold replied on 10-16-2011:
You're right about the significance of the K2-UX. It was the very first free modular differential op amp to be produced for use outside of a conventional analog computer. The only production run was about 40 of the beasts.
They were made for a special algebraic simultaneous equation solver for Stone & Webster Engineering Company. I believe the computer solved the equations of vapor liquid equilibrium at various points in a distillation column. The computer was about the size of a small rolltop desk. Its main feature (other than the 36 or so brightly red-glowing K2-UX crowns) was a string of about 10 Helipots geared to be adjusted simultaneously.
The next step was the design of the K2-W , which used basically the same assembly principle but was much more compact. It was originally designed at a time when George was becoming disenchanted with neons for voltage dropping and was playing with Thyrite, which was of course much more compact than two strings of 3 neons apiece. (The NE-2s at that time--or because of the way they were used, in strings of 3--were noisy and not very reliable.)
The units were built up by point-to-point wiring (like some of the Pease constructs I've seen) on the octal base as a "monolithic" entity. The two halves of the case were dipped in MEK (methyl-ethyl ketone) and pressed together around the octal and noval sockets and the translucent red neon's crown, essentially as an exoskeleton. Joe, I believe Bob once sent you instructions for taking a K2-W apart. The same technique would probably work with the K2-UX.
Joe Sousa on 10-17-2011:
....I see resistor values of 2.2MegX3, 10k, 220k, 2.2k, 1Meg, 360k, plus that 500pF ceramic cap. These values don't line up perfectly with the K2-W, so the design may be slightly different. The 500pF cap is a perfect match....
Dan writes on 10-18-2100 in reference to the black disk adjacent to the two neon bulbs atop the module.
That's Thyrite--a silicon carbide disc, with attachment wires on both sides.
From: Terry Walker
To: Joe Sousa
Subject: K2 UX posting
Date: Thu, 20 Oct 2011 09:27:45 -0700
I viewed your recent posting on the K2-UX and found it fascinating. The form factor of the assembly certainly would have looked dramatic at the time, which probably accounts for its unconventional arrangement. However, I get the distict impression that the case, cute though it is, was designed before anyone tried to construct the circuitry that would go inside it. The internal construction looks like a nightmare for assembly, not to mention the question of heat dissipation of the various resistors which have significant power to dissipate. The long-term reliability and drift had to be terrible with all those hot molded resistors changing in that environment.
Are there any photos of the resulting computer still around, perhaps elsewhere on your web site?
I'll bet that after asembling a few of these, George decided to design the now iconic K2-W housing, which gives much better space internally for the components.
Date: Thu, 20 Oct 2011 23:57:31 -0400
To: Joe Sousa
From: Daniel Sheingold
Subject: Re: K2 UX posting
Cc: Terry Walker
If it's not in the Palimpsest (and it isn't), it probably doesn't exist. The project was proprietary to Stone & Webster, so we probably were unable to say anything publicly about it. There were photos of course, but they were lost in the mists of history. Of the 5 or 6 Philbrick employees of the time I'm the only survivor. But I was away in the Army for two years, and I never saw anything of the sort after I returned. The company that assembled it for us, Electronic Prototypes, might have had a photo, but they were traded to Jim Pastoriza, were integrated with his operation, and ultimately to Analog Devices many years later--and I'm pretty sure I've never seen it in ADI's possession.