The Transistor Story
26th November 2009
Recorded by Ted Watt and by courtesy of the Amalgamated Wireless Australasia Veterans Association Incorporated.
My name is Ted watt. I was employed by AWA-AWV from December 1st 1941 to November 30th 1986. Initially I worked in the AWA toolroom while completing the mechanical engineering diploma at the STC (Sydney Technical College).
In March 1946 I moved to the AWV (the valve company) and over time had the good fortune to work on a range of mainly, newly introduced electronic components.
The germanium transistor was one such component. AWV had a licensing agreement with RCA in the US for vacuum tube manufacture. This was extended to transistor manufacture. On December 4th 1957 together with Henry Blanks, and engineer from AWA, we set off for the RCA plant in Somerville New Jersey. Henry was to cover engineering and test equipment, while I covered manufacturing equipment, processes and know-how.
Early in 1958 we did some preliminary work on the alloy transistor 2N109 from imported components such as the chip, indium emitter and collector dots, mounting tab, stem and enclosing shell. At the same time and area of about 6-7 thousand sq feet was being prepared together with services and controlled air conditioning. Some equipment was being bought from suppliers of RCA and some designed and built at AWV. By the end of 1958 we had two of the conveyor furnaces and production moved forward. Two more alloy furnaces were added latter.
When underway our starting point was from purchased single crystal ingots with a section like a loaf of bread about 3/4” square. Wafers were sliced with a diamond wheel, then lapped to thickness, diced into pellets with a diamond scriber, then chemically etched to thickness and then sorted using a roller micrometer.
Next came the alloy process, most critical in providing a consistent junction gap between the emitter and collector. Using graphite jigs the mounting tab (solder coated) pellet and collector dot were loaded and passed through the hydrogen fired temperature controlled oven. The process was repeated for the emitter dot.
This assembly was then spot welded to the stem. Fine nickel leads were added and connected to the dots with a hot hydrogen jet. Electrochemical etching then followed to clean up the dot periphery. Testing of the bare transistor followed using an oscilloscope. OK units were encapsulated in resin, then enclosed with the shell which had been filled either with silicone or an alundum slurry. This process was done in a cabinet with a nitrogen atmosphere.
Testing then followed and finally branding and packing.
The whole operation was labour intensive (all females). Items were small, large illuminated lenses and in some cases binocular microscopes were used.
Transistor Types we Made
2N109 family – made more of this one than any other. Considered a bread and butter line.
2N218-219 – Made a lot also. Bread and butter line.
2N270 – Physically larger, easier to make.
2N301 – Very large – easy to make – high yield.
2N247 – Difficult to make hence had a low yield.
1N…. – Point contact diode
The People
Henry Blanks left the company late in 1959. John Hooke* took over and one of his many contributions was the design and building of a semi-automatic final test unit for grading the 2N109 series. Previously some discretion by the test operator was necessary. John, of course, moved to York Street later. Then came David Saunderson. David had semiconductor experience elsewhere. At a later date he moved to Ashfield and then to North Ryde as Works Manager. Close by the manufacturing area was a lab controlled by the eminent physicist Dr Lou Davies. I don’t remember much of what went on there, suffice to say it must have been “clever” stuff.
There was a young engineer in the lab named Bill Crews. One day Bill decided it was not for him and he chose another path. Now, as the Rev Bill Crews he is a prominent person and well known and respected for the enormous amount of work for those people in need. Doug Sutherland, who was Works Manager at AWV had a physics – science – engineering background later controlled a microelectronics lab looking at the newer seminconductor silicon and integrated circuits.
I cannot add much here as I moved to other fields in 1964. Other people who worked on engineering , equipment and production were: Ken Smallwood, Bob Rose, Jack Warren, Brian Thorpe and Don Latta.
Problems:
Whether you are making transistors or 25” B&W picture tubes the problems are the same if you are involved competitively from infancy through to production. These are:
1 Your production volume is too low – must make more
2 Your yield is low – must improve
3 Your quality and shelf life – must improve
4 Your costs are too high – that’s the bottom line.
Believe me I have faced these problems many times.
Signed Ted Watt
*John Hooke was the son of Sir Lionel Hooke, Chairman of AWA and John succeeded him after his training through the company