|VK2DYM'S MILITARY RADIO AND RADAR INFORMATION SITE.|
THE G & R VALVE COMPANY.
In Australia the large wireless company Amalgamated Wireless Australasia, or AWA, made wireless valves from around 1920 and established the Amalgamated Wireless Valve Company (AWV) in 1932. Then in 1936 Philips set up a valve manufacturing plant and was followed by Standard Telephones & Cables (STC) in 1939. The STC valves were mainly for telephone repeater amplifiers but it produced transmitting valves during WW2.
However, not many people know that in 1923 there was another serious attempt to produce wireless valves for the Australian market.
During World War I Mr. Alfred John Garrod studied at Cambridge University, England, and worked in the Cavendish Laboratory with Professor J.J. Thompson, the famous electronics scientist. The Laboratory designed and made amplifier and detector valves, mostly for the Royal Navy, and Garrod learnt the techniques of valve design and construction.
After the war Garrod emigrated to Australia and shortly after met up with Mr. Sidney Radcliff who was an industrial chemist but was involved in the production of high vacuum pumps for evacuating X-ray tubes. At this time most valves had a residual gas content within the valve envelope which detracted from performance, and were called "soft" valves. Garrod realised the benefits of greater evacuation of the gases from the envelope and found in Radcliff an engineer with the capability to make a suitable pump. Garrod also had ideas on the design of valves, and felt he could design a better valve that did not conflict with existing patents which were controlled by the AWA consortium.
Garrod and Radcliff worked together for two years on the design and construction of their valve and on 24th January 1923 lodged a Provisional Patent Specification with the Patents Office. Their Complete Specification was lodged on 18th June and accepted on 2nd November and Australian Patent No. 10,811/23 was issued on 13th November 1923. In the patent pre-amble it mentions that Garrod lived at "Yundago", Hilda St., Wahroonga near Sydney.
Garrod was convinced that the elements of a valve must be precisely aligned for consistent performance and to this end his valve, a triode, was designed to allow final alignment of the three elements prior to fitting the envelope. To understand how this was accomplished refer to the sketch submitted with the patent (Figure 1). In Fig.1 of the patent sketch it can be seen that the lead-in wires (E,F,G,H) are brought through the valve squeeze (A), but are terminated in small diameter tubes (J,K,L,M). The filament wire (O) is supported by a rod (R) and held taught by a spring (P).
The plate construction is shown in Fig.2 and the grid in Fig.3 and it should be noted that each has a mounting stem protruding, (W) and (V) respectively. Now when the valve is assembled, the stem of the grid assembly, then that of the plate assembly are pushed into the tubes in the squeeze and can be moved vertically and rotated till they align precisely with the filament. The sketch of Fig.1 shows the plate and grid stems with a bend so perhaps some delicate twisting was needed for exact alignment. Once the elements were in correct location the tubes (J,K,L,M) were crimped mechanically to secure the parts in position. This is indicated by (S) although the crimping is not immediately obvious.
Apart from the claims for more precise alignment, this mechanical fixing did not require welding and consequent heat and oxidation damage, and there was no need to subsequently manually re-align and handle the elements, with the possibility of contamination.
Garrod and Radcliff registered a manufacturing company, the "G & R Valve Company" and set about obtaining equipment for manufacturing their design. Radcliff's expertise was instrumental in manufacturing mercury-vapour vacuum pumps to obtain a very high vacuum within the envelope. The factory was established "in Wahroonga, near Sydney" and I assume it was at the residence of Garrod. In 1923 Wahroonga was a distant bush suburb of Sydney, reached by dirt road, whereas today it is an exclusive residential area where a factory would be totally out of place.
By mid-1923 the "G & R 77" valve was in production, in small quantities, with lavish claims for its performance, see Photo 1. Their triode could also be used as a detector of course, but claims of "Perfection Detection" and "Ample Amplification" would make even a modern day advertising agent blush. New Systems Telephones, a manufacturer and importer of wireless components, was appointed the sole agent for G & R.
The valve went through a number of design changes during the period 1923 to 1925. Initially the brass base was from a standard bayonet lamp base of the 20's, minus the side locking pins but fitted with 4 base pins. This design is illustrated in the left example in Photo 2. Because this was non-standard for wireless valves, G & R supplied a special socket which could be mounted onto a sheet of bakelite or other material, but was also an adaptor to fit into the English sockets common at the time. Photo 3 shows a similar but different lamp base with the closely spaced base pins. Note too in this photo the crude sealing pip on the top of the envelope and internally the filament spring is at the bottom. Photo 1 and the right hand example in Photo 2 illustrate the later valve base used and the valve was available as either "English or American Base".
Notice the substantial differences in internal construction of the samples in each photo. When first released in 1923 the valve was about 25mm diameter and 90mm high. By 1925 the diameter had increased slightly, probably to match the standard valve base now used, but the internal construction and element location differs markedly between samples. I have no explanation for these variations. In John Stoke's book "70 Years of Radio Tubes and Valves", 1982, he describes the G & R 77 as follows: "A contemporary illustration of the valve shows it as having an axial filament surrounded by a coarse ladder grid followed by a widely spaced box-like anode. Judged by the standards of the day such a design was obviously inefficient and it is not surprising to find that the G & R valve was not a commercial success."
Even so, by mid-1925 the G & R 77 valve was being manufactured in the Wahroonga factory, with a production capacity of 500 valves per week and "adequate arrangements for almost unlimited expansion." In a contemporary report in 1925 the company was now said to be the "GEC Valve Works", with Radcliff as the technical adviser but working at Sydney University. GEC was associated with the agents, New Systems Telephones, but I have not read elsewhere of GEC being involved in valve manufacture in Australia. The report tells how some of the manufacturing steps were done by hand instead of machine, for better quality control. Each valve was given an aging test of 16 hours under operating conditions, with a 95% pass rate. It was claimed that over 10,000 pounds had been spent over the past 3 years to develop this "perfect" valve.
When operating as a detector, the G & R 77 required a plate voltage of 20 volts, whilst at 30 volts it was said to have peak amplification efficiency. The published specifications listed a filament voltage of 4-5 volts with a current of 0.26 amps, and a plate voltage range of 30 to 100 volts, with a plate current of 60 ma. at 60 volts. It was claimed that it would withstand 200 volts before showing any signs of gas ionisation.
When first brought onto the market the valve was priced at 23/6 which compared with around 30/- for imported De Forest and Radiotron valves. AWA had by this time increased production of a range of 4 valves and there were substantial duties on imported valves. However in 1925 there were moves to reduce the duty which led AWA to drop its prices firstly to 25/- then to around 17/6. G & R followed. During this period a popular weekly magazine, Wireless Weekly, championed the reduction of import duties and that led the magazine into conflict with the local manufacturers, AWA and G & R. There was no love lost between AWA and the rest of Australia, with AWA "the worst hated firm in Australia", but in a letter to Wireless Weekly, G & R made a plea for the retention of duties to secure it's survival.
In the end it was probably a combination of factors which led to the demise of G & R. If John Stokes summation of poor performance is correct, that information would have spread round the trade very quickly, whilst reductions in valve prices and increased imports would have made the market more intense, and undoubtedly AWA would have made life very difficult for any competition. There is no further news of G & R after late 1925 and it faded into obscurity. G & R is not even mentioned by Tyne in "Saga of the Vacuum Tube."
Australasian Wireless Review, June 1923
Wireless News, May 1925
Wireless Weekly, April 1925
Australian Patent 10,811/23
70 years of Radio Tubes and Valves, 1982
Fin Stewart, Australia's valve expert.
Please note, this article is copyrighted.