Subject: RE: Who done it? Philbrick Amplifiers.
Date: Thu, 3 Mar 2005 10:18:08 -0500
Thread-Topic: Who done it? Philbrick Amplifiers.
From: "Sheingold, Dan" <Dan.Sheingold@analog.com>
To: "Robert A. Pease", Joe Sousa, Dan Sheingold
Wow! I'd forgotten half of these! Trying to peer through the clouds, I'd guess you were at least 90% correct on most of the stuff I remember.
I'll just insert my (few) comments in square brackets.
Dan
-----Original Message-----
From: Robert A. Pease
Sent: Sunday, February 27, 2005 6:14 AM
To: Joe Sousa, Dan Sheingold, Robert A. Pease
Subject: Who done it? Philbrick Amplifiers.
Hello, Joe Sousa, I will try to define which guys designed WHAT old Philbrick products. As far as I can.
But before you post this, you must get approval from Dan Sheingold. If Dan and I agree, you can post it on your web pages.
If Dan is not sure, or if I am not sure, we might ask Bruce Seddon. But Bruce is very busy this month, moving to Vermont. If there is an item and we are not sure, you could post it with a big question mark. Does that seem fair, Dan?
PRELIMINARY list : First Draft.
---- PHILBRICK's VACUUM TUBE STUFF:
(Power Supplies)
SR-400 - Roger ("Tim") Noble.
R-100A - Tim Noble?
R-100B,
R-300,
R-500,
R-600 - all 4 engineered by Bruce Seddon. Many are working reliably in 2004.
K2-W - designed by George Philbrick AND FRIENDS. ( Per text by DHS)
K2-X - ? George Philbrick and/or Roger "Tim" Noble. ? [Used 6AN8. The plates glowed a pretty red under load]
K2-XA - redesigned by Tim Noble using 6BR8A and bootstrap bias for the Neons, about 1961 K2-XA - redesigned by Tim Noble using an NPN follower, to hold the Pentode's Cathode at -120 volts.
About 1962. This version is ~ 2 x slower than the earlier
K2-XA. * See RAP's column.
K2-P - Chopper-stabilized] [ing] amplifier - by GAP and friends?
K2-PA - a version of the K2-P with slightly revised grounds. [Probably Bruce Seddon]
K2-B- Booster - designed by - ? [George? Tim?]
*** CAUTION, the filaments should be biassed at - 150 VDC, or the tubes may become unhappy and set fire to the whole assemblage.
K2-cases that are black or yellow-brown are NOT fire-retardant. [Also not safe against shorts to ground]
* K2-WJ
* K2-PJ
* K2-YJ
* K2-BJ - all engineered by Bruce Seddon. Military versions, built on terminals on PC boards (rather than the K2-W's rats-nest.) These used mil-type tubes, and used no Neon lamps (which were unacceptable due to radiation prohibitions in some military projects such as submarines) but instead used zeners. [The neon lamps needed ionizing radiation to ensure reliable starting, originally provided by radium paint, later by lamp manufacturer (GE) incorporating traces of radon gas. NE2 became NE56 (58?)]
(((" K2-J" - never sold by Philbrick. This was a version of Loebe Julie's 1942 amplifier, with positive feedback added to bring up the gain from 2000 to 30,000. One was built by RAP in 2003, using 6SL7GT's. Just a half-baked idea of RAP. )))
USA-4 - Bruce Seddon. [There was also a military version, USA-4JX, see p. 10 of the Applications Manual]
USA-3 - George Philbrick AND FRIENDS? Did Bob Malter help? It looks like some of his ideas.
[The first version of the "Universal Stabilized Amplifier" I ever saw, on a narrower board, was George's USA-2 design. Bruce and Bob redesigned it to be manufacturable and work properly, and--because it needed a larger form factor--called it USA-3.]
SK2-V (+/- 100 volt output using 12AX7, 12AU7, and a matched pair of 6CW4 Nuvistors, in Walter Kern's metal package) SK2-P - Chopper-stabilizing amplifier (in Walter Kern's metal package) SK2-B - Booster using 7119 triodes. (in Walter Kern's metal package)
*** The basic design of these 3 amplifiers was by GAP (and friends) and optimized and put into production by R.A. Pease
-------------------------------------------------------------------------------
SOLID-STATE AMPLIFIERS.
P7 - researched by George Philbrick, but he never got it into running order. **
P2 - designed by Bob Malter, and put into production ~ 1961. Very valuable and very profitable. **
** Refer to the Chapter in Jim Williams' book on this subject. I am sure we can get permission to put this on the web-pages.
P2A - when the production of 2N384-type germanium transistors was stopping, about 1968, George Philbrick designed the basic P2A, using silicon transistors.
A very finely optimized design. I helped on some of the frequency compensation, and production engineering.
SP2A - a plug-in version of the P2A, in Walter Kern's SP package. RAP's layout.
SP2B - a 20-mA version of the SP2-A. The 20-mA booster went on a mini-board.
SP102 - a +/- 100 volt redesign of the SP2A. All the RF amplifiers ran on the same 1mA, stacked up in series. Slightly ingenious.
--------------------------------------------------------------------------------------
*** The above amplifiers were RF-coupled (5 Mcps). Below are the ordinary DC-coupled amplifiers.
P65 (also known as 6141) - Alan R. Pearlman - ARP. 6 transistors.
PP65 - George Philbrick's layout, a potted module, after ARP and Tim left to form NEXUS - George designed the package, in co-ordination with Walter Kern.
P65A and PP65A - a souped up version of the P65, to 2 mA output.
Probably a collaboration of RAP and George Philbrick. It was feasible because the 2N1132's beta was improved from 7 min....
The 6151- or - P75 - ARP. (Darlington Version of P65)
**** (I'll be happy to let ARP explain why HE and Sidney Darlington invented the Darlington at the ~ same time - but Darlington published first.....) (If you have any \ trouble getting that story out of ARP, I'm willing to help.) It's great history.
P25A and PP25A - RAP's work; a FET version of the P85. Using Amelco's FETs (NOT matched by Amelco, but matched by GAP/R) that had 1500 Umhos at 300 UA. Good Ig, better than 150 pa. One of the first good FET-input amplifiers. The PP25A was potted in yellow epoxy.
P85A and PP85A - RAP's work - high CMRR redesign of P65, using 8 transistors. See also P25A.
RAP did the PC board layout on both P85 and PP85. P25A and PP25A used the same PC boards as the '85's. The PP85A was potted in red epoxy.
P35A and PP35A - John I. Smith's design, a complete redesign of the
P75. It used FETs in the second stage, to give better drift and stability. A nice improvement.
Grey epoxy?
P55A, PP55A - a downgraded version of P65, with lower beta
input-transistors. Looser IB specs.
P15, P95 - no such animal! Never existed, never built. I never heard of any such thing proposed.
(But think of the fun we could have!)
PP25AH - a discrete-component version of the Q25AH.
PP85AH - a discrete-component version of the Q85AH.
Q25AH - a small hybrid op-amp with about 12 transistors, assembled to Philbrick's specifications by Amelco Semiconductor (Mt. View CA) The inputs were well-matched FETs driving a diff-amp. In a 12-pin TO-8 package - about 0.6" diameter. RAP's design.
Q85AH - same as Q25AH, but with NPN input transistors, instead of
FETs. "Microhybrid" .
T52 - same as Amelco 809BE. RAP's design, Philbrick's first monolithic operational amplifier.
T82AH - Same as Amelco 806BE; Not RAP's design.
P45 and PP45 - RAP's design, to do 100 MHz Gain-bandwidth product. +/- 20 mA output.
RAP's layout. It used an NSC "134P2" PNP, Similar to 2N2604. PP45 in dark gren epoxy.
P65AH and PP65AH - high-speed version of P65, using 2N2907, rather than 2N1132.
It had 27 MHz gain-bandwidth product; but it was not purely symmetrical. (Neither was the P65 or P75). Details available - later. RAP's design.
P65Q and PP65Q - a micropower version of P65, drawing less than 0.9mA Iq. Output drive, +2, -0.5 mA. RAP's design, using high-beta 2N2907. I talked George into it.
LPP65A - low profile potted module, cut from 0.625" max to 0.5" max height. TO-5's were thrown out. The 2N2907 (TO-18) was much smaller.
LLPP65AH - a very low-profile potted-module version of PP65AH, about 0.22" high. Built with "Moly-Tab" transistors. I think I built only ONE. (But it was
cute!)
"U" such as - "P65AU, PP85AU, P2AU" - MANY of the products listed above could be made with plastic or epoxy-packaged transistors. Not all were, but
many were, and if a customer had enough VOLUME, he could get almost anything made with epoxy transistors....
SP656 - a chopper-stabilized amplifier using PP65 and a transistor preamp, mechanical chopper, and a 20-mA output stage.
SP456 - same as above, but using a PP45 100 MHz main amplifier.
SP656-P- version of SP656 using a photo-chopper.
Solid-State Power Supplies:
PR-150 - Tim Noble. It had mediocre start-up, and poor short-circuit protection, due to the use of mostly germanium transistors, and an imperfect design.
PR-300 - to replace the PR-150. Designed by John I. Smith (with comments from Bob Malter) All-silicon. Solid as a rock.
PR-30 - a little joint project with - Bob Malter, Dan McKenna, RAP?
Reasonably reliable (Unless short-circuited for a long time.)
PRH-120 - RAP's design, to put out +/- 60 mA at +/- 120 volts, for various hi-voltage boosters. Built into the PR-300 package.
THIS is not any complete list of Philbrick Amplifier Products, and it
is NOT a perfect list of WHO designed WHAT. But it is a good start.
--------------------------------------------------------------------------
Who designed all the Nexus Parts? DA-1, SA-21? We should ask ARP. Many of these were sold by Philbrick/Nexus Research Labs..... I am sure many were
designed by the team of ARP and Tim Noble, with other collaborators.
Who designed the 1009, 1011, 1026, ETC ETC ETC? Let's ask Arnie Liberman. He designed a lot of these, and/or he would know who did.
K4-FF function ["fitter"] generator. GAP plus friends? [FF had two rows of pots, one set to adjust 8 breakpoints, the other to adjust 10 incremental slopes. Of course the trouble with it as a function fitter was its discontinuities. George fixed this by adding adjustable high-frequency triangular dither at each break point, which produced parabolic rounding tangent to both adjacent slopes, giving a smooth curve. The points of tangency (and hence amount of rounding) were adjustable with dither amplitude in the FFR, a successor to the FF, with a third row of 9 pots.]
SK5-F - - GAP plus P.D.Hansen with production engineering by Mayer E.Wantman.
F2-V - 3D function generator - - GAP plus friends.
Computing modules:
K3-A,-B,-C,-D,-E,-H,-J,-K,-L,-S,-T,-sigma,-tau,-V,-Z]; K5-U, SK5-U - -
GAP plus friends...
[There were two early series of shelf-mounted modules, the 1/4-rack-width K3s, and the rack-width K4s--including the K4-MU (dual) multiplier, the K4-FF], the K4-CO (P-I-D) controller, and the K4-DY (which incorporated a second-order differential equation)]
Still not complete. But it's a good start.
------------------------------------------------------------------------------
RAP's OTHER products, that he designed for Philbrick Researches:
DACs? I designed the 4001 and 4002, Philbrick's first (16- and 14-bit DACs) . I may have helped on other DACs, but not much. About 1970?
V-to-F converters: 4701, 4703, 4705, 4707, 4709, 4711, 4715, 4721 - all designed by RAP
F-to-V converters: 4702, 4704, 4706, 4708, 4710 , (4722?) - all designed by RAP. About 1969-74? P and PP85C; P and PP65C; P and PP25C - "compensated inputs"; details later.
P45AL and PP45AL - a 1-MHz version of the faster P45.
Other circuits designed by RAP: PT&H, PPT&H. 6628.1. and - probably several other bootleg products. ....
John I Smith's other Circuits: SPT&H. SP-FF Function fitter? (NOT the SP-ARF)
All for now - this has been a BIG project, just writing down all this history. !! / rap