"This compact and complete sampling system consists of a type 5T1A Timing Plug-In Unit and either of three Vertical Plug-In Units used with the Type 661 Oscilloscope. The 2 mv/cm sensitivity of the vertical units, in conjunction with their dc offset capability, allows 1000-to-1 vertical resolution. 100X time expansions and wide-range time position allows 1000-to-1 time resolution. To help make these features applicable to your problem, Tektronix offers a wide variety of probes, test fixtures, and accessories." (Tektronix, Catalog, 1963/64)
BRIEF SPECS
| Frequency Response | dc to 3500 MCycles |
| Input Deflection Factor | 2 mv/cm to 200 mv/cm |
| Time Base | 1 nsec/cm to 100 usec/cm |
| Ventilation | forced filtered air |
| Power | 450 watts |
| Dimensions | 16-7/8" x 13-1/8" x 23-3/4" |
| Weight | 49-1/2 lbs (24 kg) |
| Technology | tube, transistor |
| Price (for 1963) | $ 1,150 w/o Plug-In $ 1,600 for 4S2 $ 750 for 5T1A |
After Hewlett-Packard launched their model 185A sampling scope in 1961 Tektronix answered with the "lousy" type N plug-in for the 530/540/550 series mainframes - plus a bunch of trigger oscillators and delay cable box. While better to have than nothing Tektronix was not able to compete with hp until in 1963 by unveiling the type 661 sampling system. Different plug-ins were available - here the 661 is equipped with the 4S2A sampling and the 5T3 time bas plug-ins. The engineer behind the N was Norm Winningstad. He worked at Hughes Aircraft Electronics Division and was offered a job at hp's oscilloscope division to join their Sampling Scope team. After getting informed about the work he went to Tektronix developing the N (according to: "A narrative history of Hewlett-Packard from 1939-1990" by John Minck)
I did not yet restore this scope because I am still looking for the manuals. Power-up was successful but till now I was not able to get any meaningful display. So the scope is just down and dirty. Ah - no - the scope did not grew behind the house!
Sampling scopes are prone to failures by supplying more than the allowable dc voltage to the input. This blows up the special matched sampling diodes which are difficult to replace today. So best way is to use a dc block (same for spectrum analyzers). The 661 still appears in some application notes of Linear Technology - there are some circuit guru's who tells us that they are working with it...
A detailed discussion of the development of sampling technology you can find in "50 Years of RF and Microwave Sampling", Mark Kahrs, June 2003
