Westinghouse Tube Development
Leading to the WD-11

Fig. 1

Fig. 2

Well before their affiliation with RCA, Westinghouse experimented with vacuum tube designs to produce a receiving tube to market. These efforts proceeded in various directions. They purchased a large number of Moorhead tubes that they used for testing and experimentation. They built a number of Western Electric style tubes with glass arbors to explore that method of construction. In the higher power area they made copies of the General Electric "U" tubes. The "U" was a WWI transmitting tube (CG-1144) that evolved into the UV-203 in 1921. Like others they also attempted to produce a tube which would not infringe the DeForest patents. Starting in early 1919 (this date from Howard Schrader's notes, see paragraph 5) they built a number of experimental types for this purpose, several of which are shown. The tube in Fig. 1 is basically just a diode and may have been used for filament tests. However, it was more likely tested as an external grid device by providing an external control element around the bulb. Fig 2 shows a similar tube with a flattened center section and an external winding as a control element. The external grid structure was well known by this time, but no one had been able to design such a tube that was as efficient in performance as the standard triode.

Fig. 3

Fig. 4

Another approach is shown in the tube of Fig. 3. In this tube a central filament has a plate on one side and a control element on the other side. This scheme, if it worked at all, would have been very inefficient. An attempt to improve this design is shown in Fig. 4. In this case the control element has been located closer to the filament than the plate element. All of these experiments were doomed to failure as the standard triode structure was inherently superior to any design which did not place the control element between the filament and plate.

Fig. 5

Fig. 6

Westinghouse gave up on a non-DeForest design and by 1921 had designed and marketed the WR-21, a receiving tube of standard design. Two versions were made, the WR-21(A), Style #319517 (Fig. 5), and the WR-21(D), Style #319513 (Fig. 6). The WR-21(A) was listed as an amplifier, while the WR-21(D) was listed as a detector or amplifier. The detector version had green lacquer on the tip. These tubes were both high vacuum (hard) tubes, and of identical physical structure. Tests on a number of these tubes (at 90 volts on the plate) revealed no significant differences in electrical constants, so the real distinction between the two is unclear. Perhaps the detector version was selected out of the overall production. These tubes used a base similar to the European type used on the TM and R types. No other US tubes are known to have used this base.

Fig. 7

Fig. 8

Fig. 9

Fig. 10

A mate to the WR-21 types was the WB-800 ballast tube, Style #319532. These tubes can be found in several types. Fig. 7 shows one type with a vertical element. Another version was made (Fig. 8) with a horizontal element. The same Style # was used for both constructions. The WB-800 was designed to be used with the WR-21 tubes, and one WB-800 was used with each WR-21. The only receiver to use this combination of tubes was the Westinghouse Aeriola Grand. However, most samples of the Westinghouse DR detector/amplifier unit used WR-21's without the WB-800's. The Westinghouse DR (Fig. 9) was similar in outward appearance to the well known DA, but was resistance coupled and used many of the same components as the Aeriola Grand. At least one DR has been seen which had UV shell sockets. The DR was not advertised and not placed into production, the few known survivors are believed to be engineering samples. After Westinghouse joined the RCA combine in 1921 they tried unsuccessfully to get RCA to accept a rebased WR-21 (Fig. 10) as a UV-201 equivalent. This tube was produced briefly in 1922. It is likely that the DR unit seen with shell sockets was intended to use this tube.

In a 1973 interview conducted by noted early collector Howard Schrader, Lee Sutherlin, long time Westinghouse engineer, provided some details of the WD-11 project. Sutherlin started with Westinghouse in October of 1920, and initially worked for H. W. Freeman in the Small High Vacuum Tube Group. It had already been decided to make a dry cell tube with low filament current and running from a single 1.5 volt dry cell. It had also been decided that the nominal voltage should be 1.1 volts to allow for battery run down. Based on earlier tests it was believed that the only practical filament for such a tube would be oxide coated. Westinghouse had little experience with oxide filaments, so Sutherlin was assigned the task of gathering information on the subject. He went to see the tube engineers at Western Electric who provided the necessary technical details to manufacture oxide filaments. The agreements at the time prohibited Western Electric from producing receiving tubes for direct public sale so there was no problem with sharing the information with Westinghouse. Sutherlin's boss, Freeman, left the group around this time and Sutherlin was put in charge.

An extensive series of tubes was designed and tested as part of this project, and a listing of the entire group was compiled by Westinghouse engineering. A copy of this list was given by Sutherlin to Howard Schrader. This list was dated July 26, 1921 and gave information as to the basic structure, filament current, measurements of u, Gm and Rp, and notes. The tubes were given T numbers running from T1 to T99. Not all the numbers from 1 to 99 were used, and for many of these types multiple samples were made. In addition each individual tube was given a tube number, the numbers ranging from 382 to 733. The list does not include all these numbers, in all 236 numbers are detailed. It is not known whether the list is incomplete or if some numbers were not assigned. It is assumed that tube numbers before 382 were used in earlier experiments for types like the WR-21.

Fig. 11

Almost all of the tubes in the series had cylindrical plates with the exceptions noted in the list. Tube numbers 382 through 588 used pure platinum for the filament base material, with platinum-iridium used on the later samples. The list does not show filament voltages, only filament currents. The series started with the T1 which apparently had no grid and was used only for filament tests. The T2 was also used mainly for filament tests but did have a grid as normal tube constants are given. A few of the early T numbers are shown as having flat or oval plates, but after the T24 all types were cylindrical. Several of the T14's were made using "Parts from W.E. Co. VT-3". A few of the earlier tubes have no T number assigned. One tube is shown as the T50/WD11, and a small number of others are listed as just WD-11. As might be expected a number of these experimental tubes had problems, with notations like "Very high resistance", "Blue glow", "Filament burned", "Unstable", and others. The tube shown in Fig. 11 was early in the series and could have been a T1-15, T1-16, T9, T10a, T10b, T15, T16, T18a, or T24. These types were all listed as having an oval plate. The earlier T types had filament currents ranging from .22 amps to .80 amps. After tube # 589 (platinum-iridium filament) the current was mostly in the .25 amp to .40 amp area.

Fig. 12

As the development progressed various things were tried, some of which show up in the notes. T69 was made with an "Electron trap on end", and T37 was made in both "Quartz spacers" and "Glass spacers" versions. T75 and T77 used a "2 mil round filament" and T71, T84 and T85 had versions with tungsten filaments. The tube shown in Fig. 12 used a short plate with an inverted V filament, suspended from the top by a mica support. The actual T number of this tube is not known. Although all the early tubes shown in this article have markings on the press, these markings do not seem to correlate with the designations in the list. The T numbers went as high as T99, but by mid 1921 the T88 had been selected as the best version. A total of 35 of the T88 tubes were made, many more than any other type.

After the decision to go with the T88, a demonstration was set up for Westinghouse VP Harry Davis. This demo used a detector and one step amplifier, where the amplifier tube could be switched in and out of the circuit to demonstrate the stage gain. The audio was a steady tone. Sutherlin describes how they enhanced the effect of the amplifier stage by adjusting the frequency of the tone to the natural resonance of the earphone used as a monitor. This, of course, caused the audio output to be louder. Davis was impressed by the demo and ordered that the tube go into production as the WD-11. The Westinghouse production people were shown the prototypes but said that the factory could not build such a tube. Davis ordered Sutherlin's development group to make about 500 tubes to prove that it could be made. About 300 were actually made in late 1921 before the production group relented and agreed to start production.

According to Howard Schrader's notes, the first of the WD-11's had a Shaw base, but this version has not been seen. The base of the production WD-11 was the same as the WR-21 except that the plate lead was made larger in diameter so that the WD-11 could not be plugged into a socket intended for a WR-21.

Fig. 13

An interesting early version is shown in Fig. 13. This sample uses a larger diameter plate than the usual WD-11, but smaller than that of the WR-21. The grid appears to be the same as the normal WD-11. Note the cork spacer in the stem, and the elaborate Z shaped upper filament support. The leads are brought through the base and wrapped around and soldered to the pins. This same method of lead connection can be seen in early production versions of the Westinghouse WB-800. The Westinghouse logo and standard license notice appear on the base, but the type number cannot be seen, perhaps worn off.

Fig. 14

Fig. 15

The tube in Fig. 14 is an early production WD-11. This tube is similar to that of Fig. 13 except for the plate size. Normal production markings are present on the base. Fig. 15 shows a later production version of the WD-11.

The original Westinghouse Aeriotron tubes had no getter, but by the time Westinghouse started making these tubes for RCA a lime getter had been added. Sutherlin was responsible for developing this lime getter. The Westinghouse WD-11 dry cell tube was rather fragile and subject to damage from rough handling. The GE dry cell tube, the UV-199, was much tougher and enjoyed far greater sales than the WD-11.