The Transformer (Part 4)

Elihu Thomson was a geeks geek. It has been said that as a boy, his first invention was an electric friction generator built from a wine bottle. He became a science professor at Philadelphia's central High school. This wasn't a public high school, this was a different type of institution they did, and still do, grant BAs in addition to high school diplomas. I explain that because he was a professor, not just a teacher. In 1880 with another teacher, Edwin Houston, he started the Thomson-Houston Company. He was 23 years old, and Houston was 33. They were focused on the new hot item: arc lamp systems. More here. This set a huge number of modern events in motion.

The Thomson-Houston Electric Company was founded in 1883 by a group of shoe manufacturers based in Lynn, Massachusetts. Those investors, led by Charles A. Coffin, bought out Elihu Thomson and Edwin Houston's American Electric Company from their New Britain, Connecticut investors. Like the Gaulard and Gibbs transformer, the Thomson-Houston had a number of patent conflicts. But unlike Gaulard and Gibbs, they had the backing to pull a Google and just buy up the damn patents. In 1889 they bought out Charles Brush and ended their biggest problem. More here.

Their 2nd  biggest problem was Westinghouse. In 1886 Westinghouse took it's improved Stanley transformer and installed it's first arc light system in Buffalo at an existing Brush Co. station. That same year Westinghouse won out on a patent for AC distribution over Thomson-Houston. To his credit, rather than fight Westinghouse endlessly patent-by -patent, Thomson convinced them to share. By 1887 Thomson-Houston was manufacturing Westinghouse arc systems, and Westinghouse had a license to sell Thomson-Houston arc lighting gear. The deal lasted 2 years, ending when the Westinghouse patent was declared invalid. It didn't matter, Thomson had been hard at work trying to circumvent the patent anyway. The MBAs call it coopitition now.

Thomson was on a 2-year tear, patenting  more transformer models; different shapes, different laminations, different cores and eventually a core that sat insulated in an oil bath. When the Westinghouse patent died, Thomson-Houston patents were unavoidable. Westinghouse will owned the Albert Schmid improvements to the Stanley design, but they were otherwise in a sea of Thomso's patents. In 1890, Fleming wrote that there were 100,000 lamps using the AC dynamo in the Thomson-Houston system. He went on to describe their transformer.

"The field magnets are of cast iron, and are made in two parts, the lower one bearing the pedestals for the armature bearings. Each casting carries projecting inward from it one-half the total number of radial pole pieces. The number of polar projections varies from eight in some of the smaller sizes to twenty-two in the largest. The armature is a laminated iron cylinder, and has applied to its surface a set of flat coils which form one layer of double-insulated copper wire of square cross section. This armature winding consists of a number of flat oval coils, equal in number to the number of polar projections, wound in a lathe on special formers, and curved so as to fit the cylindrical surface of the iron core. There is a space of three-quarters of an inch or more In the centre of each coil which, in some machines, is filled with an insulating slip or core, and in others with a second separate coil used for exciting the field of the dynamo. The armature coils are securely bound to the armature core by German silver wire, and are covered with insulating strips of mica. These coils are either connected all in series and in series with the external circuit..."

You can see from that description that this transformer also draws heavily on those that came before it.Nonetheless the Thomson-Houston Co. patents were legit because they were narrow and specific. In 1892, Thomson-Houston merged with the Edison General Electric Company and history rolled on. In 1909, Thomson became the first recipient of the Edison Medal, bestowed by the American Institute of Electrical Engineers AIEE, what we now call the IEEE. He was president of the organization from 1889-90, and acting president of MIT from 1920-1923. He died in 1937. Edwin J. Houston was president of the IEEE twice actually, he took a different path, forming a consulting firm in 1894 with Arthur Kennelly.


There is a lot more to be said about Thomson-Houston and GE. I recommend the book Innovation as a Social Process: Elihu Thomson And The Rise of General Electric by W. Bernard Carlson. There were other important transformers that followed but most of them were a part of the natural progression to make components smaller. Here's a short list of models I didn't get into for those of you who are obsessive enough to read more.

• Charles Brush Transformer - 1881
• Lowrie-Hall Transformer - 1886 
• Mikhail Dolivo-Dobrovolsky  Three-Phase transformer - 1888
• Swinburne-Hedgehog Transformer- 1890
• Oerlikon Transformer - 1891
• Siemens Cable Transformer - 1891
• Frankfort-Laffen Transformer - 1892
• Schuckert Transformer 1894
• Elkhart Transformer - 1894

The transformer (Part 3)

 To his credit Westinghouse licensed both the Ferranti transformer and the Gaulard and Gibbs transformer. That's pure capitalism. He committed to a system (AC) and didn't trouble himself with the particulars. (In some ways it reminds me of Google's handling of their patent troubles.) But Westinghouse did bet on his pony, William Stanley. He had the good fortune to be the man who designed and installed electrical lighting at the 5th avenue Westinghouse offices.  George's brother H.H. actually made the introductions. George Westinghouse liked his work, and hired Stanley as his chief engineer for his Pittsburgh factory in 1884.

In 1885, Westinghouse imported a number of Gaulard-Gibbs transformers and a Siemens AC generator for Stanley to play with. Using this as research materials (ok and a bit of patent violation) William Stanley, Jr. built his first transformer and funny thing...it was a lot like the Gaulard and Gibbs's model. For the record, he also had access to the patent papers on the ZBD transformer.  But Westinghouse was still flirting with DC power and that infuriated the moody inventor. They weren't ready for a divorce so they got a trial separation in December of 1885. Under the new deal Stanley worked out of an office in Massachusetts. By March he had lit up a small neighborhood with his transformer and an AC generator. It was the first such system in America. The book Networks Of Power by Thomas Parke Hughes described it in detail:

"The length of the transmission circuit from central station laboratory to village center was about 4,000 feet. Connected to it were thirteen stores, two doctors' offices, one barber shop, the telephone exchange, and the post office."

 His claim to the first electrification of a network is indisputable. But what parts were actually his own? The generator was a Seimens and the transformers derived from Gaulard-Gibbs.This is hard to say for certain, but this is the system that Westinghouse mass manufactured and installed for a decade. Maybe I should back up a bit. In 1872 Electrical world was already reviewing a Mordey Transformer. William Morris Mordey’s transformers were manufactured by the Brush Electrical Engineering Company. This type of transformer is called a "shell transformer" beacuse the coils are completely surrounded by iron (except the ends).His electrical work was so important that a magnetic field behavior in dynamos was named for him.

• Mordey Effect (noun) A phenomenon observed in dynamo armatures. At full loads the hysteresis decreases. The effect is thus expressed by S. P. Thompson. "When an armature core is rotated in a strong magnetic field, the magnetization of the iron is being continually carried through a cycle, but in a manner quite different from that in which it is carried when the magnetizing force is periodically reversed, as in the core of a transformer. Mordey has found the losses by hysteresis to be somewhat smaller in the former case than in the latter."

Both the Ferranti model and the Westinghouse model descended from Stanley's patents are also similar to the Mordey transformer. In all three models the circuit is made of thin rectangular iron plates, with coils wound through punched holes. So they're all manufactured and assembled about the same way.  The obvious difference among all the models I've mentioned is that the Gaulard and Gibbs model was an open circuit. All the others were closed. I think the incestuous nature of these developments is pretty clear now.  But lets get to the reason I need a part 4 for this series:

There were three large American companies making transformers by 1890: Edison, Westinghouse were fighting AC vs. DC. But there was a third; The Thomson -Houston Company. In 1892, financier J.P. Morgan engineered a merger between the Edison interests and Thomson-Houston. The resulting company was named General Electric. More on their role Monday.

The Transformer (Part 2)

Lucien Gaulard and John Dixon Gibbs first exhibited their transformer in London in 1881.  They later exhibited the invention in Turin in 1884. They had tried to patent it in 1882 but failed because of a few related devices but mostly because of their arch nemesis Sebastian Pietro de Ferranti. Most sources claim/intimate that Ferranti "invented" the transformer in the process. Gaular and Gibb thought otherwise, but they lost in court. (Nonetheless it was the basis for the ZBD transformer.) Eventually the patent struggle drove Gaulard  insane. He died in an institution (Sainte-Anne Hospital) in Paris.

In reading the history now,  Ferranti's claim seems exaggerated. At that time Edison and Tesla were fighting it out AC vs. DC. We all know that AC won that fight, but at the time it wasn't clear to everyone. Ferranti bet on AC, as did Westinghouse and everybody that made that bet won big. But it appears that Ferranti got the credit not for being first, but for having the prettiest patent filing.

Ferranti was born in 1864. So he was in his early 20s as all this was ramping up. In his early teens he was a bit of a gear head experimenting with arc lights, and dynamos. Mr. de Ferranti designed a dynamo with William Thomson (aka Lord Kelvin) in old engineering text books they call it the "Ferranti Dynamo." He was supposedly 16 at the time so that would have been around 1880. He started work for Siemens in 1881. Remember the ZBD transformer was developed by 1885 so Ferranti has a small time window to figure this all out and get to court in time to piss off Gaulard and Gibbs.

This Ferranti Dynamo (or it's components) is largely what most history books are claiming is his transformer. Obviously dynamos are not transformers, but they do have some things in common... like those coils.  This wasn't his only widget in that era. He was also making alternators and armatures. All of them were metal bodies and wound copper coils. He took out a patent on his modifications to existing transformers in December of 1885: patent #15,141 "Improvements in Electrical Converters."  His transformer was made of iron strips laminated with paper on one side, bundled and shellacked. A cast iron frame supported the bundles. Cleverly he also patented his distribution system to set up transformers in groups or parallel mains. Because he got these two ideas on paper, we generally consider it the birth of the modern transformer. John Fleming wrote a comprehensive definition in 1897. I'll paraphrase for brevity:

1. When the primary circuit has its terminals connected to primary mains it should exhibit always a constant difference of potential between its secondary terminals, independent of the secondary load

2. There should be no waste of energy other than that due to the resistance of the conductors or circuits.

3. There should be perfect and permanent insulation between the primary and secondary coils and between both coils and the iron core.

4. There should be no sensible electrostatic capacity between the primary and secondary coils.

5. When the transformer has its secondary circuit open, no sensible current should flow through the primary circuit from the primary mains.

6. The loss of energy in transformation should be as small as possible at all loads.

Fleming described the ZBD transformer as a major improvement on Gaulard and Gibbs. But by 1885 Ferranti was off and patenting specialized transformers for telephony with Addenbrooke. It's important to note that Ferranti had a good lawyer. His lawyer recommended he set up private manufacture of his alternators to keep it out of the hands of Siemens. He later married his lawyers daughter and gained a very useful father-in-law. In 1886 he was appointed chief engineer for Sir Coutts Lindsay & Co. Flush with cash and backers he challenged the Gaulard and Gibbs patents in 1886.

The problem was that Gaulard and Gibbs papers and hardware had already made the rounds of the technical magazines. Most patent improvements in that time window start with Gaulard and Gibbs; not Ferranti. Westinghouse had bought half a dozen and shipped them to Pittsburgh. And crucially, William Stanley was in Brooklyn, NY, corresponding with Gaulard and making his own improvements...

The Transformer (Part 1)

Simply put, a transformer is a device that transfers electrical energy from one circuit to another through inductively coupled conductors i.e. coils—that's plural.  Because of that magic word... "coils" this post relates to my series on induction here. I'll try not to repeat myself too much. Try to think of this as a sequel. We use transformers to step up or step down the voltage in a circuit, or to convert AC power to DC power. (like wall warts)  This is more important than you think. Everything in the world runs on AC power, and that requires DC to AC conversions.) In 1802, Humphry Davy invented the first light bulb and it began an international human obsession with lighting up everything.The transformer was destined for ubiquity.

Most sources agree that the first transformer was an induction coil, invented by Rev. Nicholas Callan in 1836. His was not even the first induction coil with two coils. But he was probably the first experimenter to figure out that the relationship between the primary and secondary winding and the number of turns of wire. (That's directly proportional by the way. The more turns the secondary winding has than the primary coil, the more EMF it produces. But Callan wasn't using his coil as a transformer. He was using it to increase the voltage output from a battery. It was a critical step, but not exactly a modern transformer. It might have been even more important in a way that Callan suggested in an 1857 that joining together the secondary circuits of a number of coils might be able to power an arc light. That got some peoples attention, possibly even Mr. Bláthy.

The first big name in transformers is Ottó Titusz Bláthy. The term "transformer" itself is generally credited to Ottó Bláthy.  He was an electrical engineer in Hungary who started his significant work around 1880. He was employed at the Ganz Works in Budapest. they built ships, submarines, cars, planes and even power plants so Bláthy  had access to a big shop. He worked with Miksa Déri and Károly Zipernowsky in developing the ZBD model alternating-current transformer in 1885. This was a toroidal-type transformer for use in their incandescent AC powered lighting.

They were overcoming the short comings of DC: heat loss, distance limitations, etc.They developed an array of improvements to make a primitive transformer work better.  Zipernowsky  had developed the shunts and  Déri largely performed the testing. But our man Otto here designed it's iron core. He was their youngest engineer, just 22 years old. The end result was a set of 75 transformers that stepped down 1.35 kV to power over a thousand light bulbs. They had started the project with a Gaulard-Gibbs transformer. More on that tomorrow...

Unun vs. Balun

I think you could make a palindrome starting with with "Un balun y un unun..."  I don't think I'm the right man for that task. But I can tell you what they are and what they are for. balun is a difficult word to research being both a surname, an engineering term, a city,  and a slang word in Spanish meaning  balloon. They are both types of transformers and sometime I'll have to write a post about those as well.

• Balun [bal-uhn] noun

A device that allows a symmetrical antenna to feed an asymmetrical line. It is balancing an unbalanced input. It's a portmanteau word combining balanced and unbalanced. In it's early use it was almost always called a "balun transformer." We have largely dropped "transformer" from it's paired use. Balun already intimates transformer.  I hae fourd it used in catalogs from 1950, and it appears to have come into use in the decade prior.  The earliest use I found was in a 1947 catalog and was describing a "Balun bazooka" then described it as "a device for transforming from BALanced Line to UNbalanced."  [Their own capitalization] the timing leads me to believe that the contraction is military in origin. Here are a couple confirmed early uses:

Very High-Frequency Techniques, Vol. 1 - Harvard U. - 1947
"For a fixed position of the balun short-circuiting bar, the induced voltage E tends to increase with the frequency"

Hewlett-Packard Journal , Volumes 1-13 - H.P. Company  - 1949
"When this condition exists, there is theoretically no phase error introduced by deviations in balun legnth."

Radio Engineering - Terman, Frederick- Mcgraw-Hill - 1947
"At the same time, the balun does not introduce an unbalanced load to the balanced system, since the impedance across 2-3 approaches infinity..."

• Unun [uhn-uhn] noun

A device that allows an unbalanced antenna to couple with an unbalanced feed line. It's also a portmanteau word combining Unbalanced with Unbalanced. It's a less common device and more than likely a pun based on balun. This term has a much more recent vintage. I can't find it in any texts before about 1995. The earliest reference I found was in Issue 72 of the ARRL handbook for the radio amateur.