Talk:Transformer

proper theorectical treatment

We should have a proper theoretical treatment of how a transformer works, including treatment of:

• impedance transformation
• current ratio and voltage ratio
• non-linear effects and losses

Perhaps an elementary introduction using induced e.m.f. balance and conservation of energy is the right way to get into it? -- The Anome 08:10 12 Jul 2003 (UTC)

conversion to motion=loss?

I'm a little uncomfortable with

The energy converted to motion is one form of core loss.)

since some of that motion is not lost immediately, but stored as elastic compression. I expect resonating at the mains frequency might ameliorate the loss (or might destroy the unit!) and an inspection by a power-systems EE might be valuable here. --Jerzy(t) 01:19, 2004 Mar 5 (UTC)

There is a whole lot about energy storage that could be included. (I remember Professor Jerrold Zacharias pointing out that the trick in a necessarily simplified introduction was not teaching anything that would have to be unlearned later.) There is not only mechanical energy storage, but magnetic and even electrostatic energy storage to consider in a real transformer. You are right, some of the stored energy is recovered, some is dissipated. I am uncomfortable with leaving out any link to ideas related to power factor, the issue that the current and voltage waveforms may not be in phase. Maybe there is an article on reactive power to link. Also of practical importance is that transformers sometimes have non-sinusoidal waveforms applied. Harmonic distortion of nominally sinusoidal power waveforms can be a big problem. The imperfect relation between B and H is critical in practical transformers. A power transformer can overheat and fail if there is a significant DC component to the current, for instance. And you can not arbitrarily change the voltages applied - sucessful use depends on more than the turns ratio.

It is possible to have significant unintentional transformers. Some examples:

• AC wiring practice always uses two wires, and they are always run together.

If you run the two wires in two separate metalic conduits, for instance, you have created a simple transformer with a single turn secondary - the conduits, (and a single turn primary, the two wires) - the conduits can get quite hot.

• Twisted pair is often used in balanced circuits to minimize inductive coupling from ambient fields. However, if you have two circuits and you cross wire the twisted pairs, the circuit will look fine at DC, but transient signals in one circuit will transformer couple into the other circuit in a mysterious manner. The twisted pairs have become a transformer! The first time I encountered this wiring mistake it took a while to figure out.

This is all good information, which I shall add to the article if I get time. Unless you do it first, of course. ;-) -- Heron 22:14, 5 Mar 2004 (UTC)

Pacinotti

I removed this line:

"Antonio Pacinotti (1860)"

from the list of inventors, because I can find no evidence that Pacinotti had anything to do with transformers. He did, however, invent a dynamo in 1860. Perhaps there was a mistranslation somewhere. -- Heron 20:06, 24 Jun 2004 (UTC)

Power transformers

I've added some content pertaining to power transformers, hysteresis, and fire-resistant cooling fluids. I'm not sure D'Arsonoval belongs here . I removed the pejorative description of PCBs as "toxic waste" since the people who invented PCB and thought it was perfect for the job didn't think of it as such. --Wtshymanski 17:21, 11 Dec 2004 (UTC)

Autotransformers?

Ignition coils and flyback transformers aren't autotransformers. An ignition coil only has three terminals, but there are two discrete windings that happen to share a ground connection. An ignition has far too high a ratio to benefit from an extra 12 volts on the primary. Similary with a flyback transformer.

I don't think a flyback transformer is "resonant" - as I understand them the flyback in a TV set captures the energy that's been stored in the deflection yoke, but is not a resonant circuit itself. I suppose an ignition coil is sort-of "resonant" since it does rely on energy storage as part of the spark generation process. I'd like to add some stuff about considerations in design of power transformers, in terms of trading off no-load and load losses...if this article hits 32K it will be time for a separate article on power transformers. --Wtshymanski 22:33, 16 Dec 2004 (UTC)

I can't say whether your average flyback is an autotransformer, but most flyback transformers used in ordinary television sets are definitely resonant. A common test for a suspect flyback used to be to apply a pulse and see if the transfomer "rang" (using an oscilloscope to monitor the voltage waveform of the ringing). If it did ring, the flyback was good, but defects like opens or shorted turns would kill the resonance. IIRC, they usually resonate at about 7 times the horizontal scan rate; this produces the best "flyback" of the trace.

I don't know how this works in multisync computer monitors; perhaps someone else can explain how they can be used at a wide range of sweep frequencies (or even a continuous range of sweep frequencies. At one point, I think multisync monitors used switched capacitors; perhaps nowadays high-voltage generation is separated from horizontal deflection and the horizontal deflection is just "brute-forced"?

Atlant 22:30, 28 Jan 2005 (UTC)

Years ago, several green screen monitors were discarded at the University. I pulled out the flybacks for use in high voltage bias supplies on our beam transport experiment. I conducted some simple tests to determine resonant frequency of a bare flyback and step up ratio. The secondary coil resonance was about 65 kHz, if I remember correctly. The step up was quite high, can't remember what it was though. The primary resonance was expectedly much higher. The sweep circuits were probably lifted from a standard B/W TV, so it's likely the sweep was standard [NTSC] of 15.75 kHz, so it wasn't exactly resonant. Like most solid state flyback circuits, it had a typical diode/cap tripler. My guess is the anode voltage was around 15 kV, so the flyback itself peaked at 5 kV. I think I was able to get as much as 30 kV out of it before the secondary broke down and arced out. Madhu 19:40, 21 August 2005 (UTC)

Step-across transformers

"step across" transformer appears to be a Wikipedia artifact only and is not actually used in the industry. I've been unable to locate a reference to that term either in my local library or on the Web, barring a bunch of quotes from Wikipedia itself. I've deleted this Wikifiction. --Wtshymanski 20:51, 28 Jan 2005 (UTC)

capacitive transformers

i heard there is a dual to the transformer, the capacitive transformer or electrostatic transformer, which I believe is just two coupling caps carrying AC to and fro so the circuits are floating relative to each other. does this deserve an article? are they used anywhere? - Omegatron 01:23, Mar 19, 2005 (UTC)

Capacitive voltage transformers exist and are used by utilities for high-voltage (greater than 66 kV) metering. They have a capacitive voltage divider but also have a dual-winding transformer to couple the divided voltage to the metering circuit. They tend to have lower allowable burdens than a wound transformer but can be made economically at higher voltage ratings. Another difference is that even though they decrease voltage, they do not increase current as found in wound electromagnetic transformers - an ampere drawn by the load is an ampere drawn from the primary circuit. And of course they can only reduce voltage, not increase. The article is talking about magnetically-coupled transformers and these "capacitive transformers" aren't really "transformers" in that sense. One could also mention optical "instrument transformers" but these don't use windings coupled by magnetic fields, either. --Wtshymanski 04:35, 19 Mar 2005 (UTC)

Hmmm... What I heard of was a purely capacitive device that operated in a way analogous or complementary to a transformer. This could be as simple as a coupling cap for all I know. Or it could not exist. It is the dual of an inductor transformer, as a capacitor is the dual of an inductor. By "optical instrument transformers" do you mean optocouplers? - Omegatron 00:33, Mar 24, 2005 (UTC)

I see what you mean, but I've never heard of a "dual" to a "magnetic" transformer. What the utilities call a "capacitive voltage tansformer" is really only a capacitive voltage divider. In a "magnetic" transformer you can have multiple turns around the same magnetic flux; in a "capacitive" device I don't see any way to do anything analogous. The lines of the electric field stop at the plates, so each plate can only "link" electric flux once. So you can't get the very useful property of changing voltages and currents so long as VI is roughly constant.

There used to be capacitive multplexers used to isolate analog signals from input to output - (may still exist, outside of my field) - these used two sets of relays, one to momentarily connect a capacitor to the input signal, and a second to connect the capacitor to the measuring circuit. This provided as much isolation as you could get in a pair of relay contacts, but wasn't a "transformer" as such.

No, an optical current transformer uses some hi-tech physics to turn a varying magnetic field into varying polarization of a beam of light - this is called the Faraday effect. For example see www.nxtphase.com/sub-products-optical_ current-nxct_current-transformer.htm , or of course ABB has them also at www.abb.com/global/seitp/seitp332.nsf/ 0/177a4854b28ce5d0c1256eaf002cc96f?OpenDocument - (what would we do without Google?). These devices are only called "transformers" because that's what utilities like to call them; it would be more accurate to call them "voltage/current transducers". An optocoupler usually is just an on/off signal, and doesn't provide an analog value. A quote I read says "Optocouplers are just signal fires writ small."--Wtshymanski 16:17, 24 Mar 2005 (UTC)

Huh. I guess I heard wrong. Don't forget about piezoelectric transformers. - Omegatron 02:13, May 28, 2005 (UTC)

a static electrical device

why "static"? - Omegatron 01:29, Mar 19, 2005 (UTC)

As opposed to devices with moving parts - a transformer changes voltage with (in principle) no moving parts, unlike, for example, motor-generator sets. --Wtshymanski 04:35, 19 Mar 2005 (UTC)

The current edit now says "no moving parts" - this isn't as accurate as "static". Prefixing "magneto-" or "electro-" shows that there is a root meaning, which meaning is the one I had intended. Large power tranformers, (at any rate), have numerous parts that have to keep moving for the transformer to keep operating.( When I was a kid I would have thought "transformer oil" was a joke, like "propwash" in pilot school - the oil moves!) And of course there are a couple of varieties of transformer where the major elements do move with respect to each other, but that motion isn't the basis for energy transfer between primary and secondary. --Wtshymanski 16:27, 24 Mar 2005 (UTC)

So "no moving parts" is not perfect, but I don't like "static", either. It doesn't convey the right idea with just that word; plus newcomers will associate with this kind of static. It should just have a sentence describing what you meant by static. - Omegatron 02:13, May 28, 2005 (UTC)

Strange edits

Is this work in progress or should we revert to an older edition, that's less scrambled up? At this point the article has duplicated sections and some text has been cut. What's up ? --Wtshymanski 20:16, 21 Mar 2005 (UTC)

i'll fix it; sorry. work in progress. - Omegatron 20:30, Mar 21, 2005 (UTC)

Sorry to "experiment" on a live article. In the future this will be less ugly. see Wikipedia_talk:Extended_image_syntax#Multiple_images_in_one_frame for progress. these three articles have such tables: Transistor#Types, transformer, BJT. - Omegatron 20:35, Mar 21, 2005 (UTC)

Poly Phase Transformer

Does any one know how these work or something like that? Is this just three seperate transformers? Thanks

No, true polyphase transformers have three sets of windings that share a single set of core laminations. The usual arrangement is to have a core that is shaped like the letter "E" but with the open end closed. One set of coils is placed around each of the "legs" of the E.

I think this works out nicely because the phase relationship means that some of the magnetism from one set of coils is opposed by magnetism from other sets of coils, allowing a proportionally-smaller core assembly than would be required by three completely-separate transformers.

Here are some images from out on the web click..., click..., or click. The second and third clearly show the wiring of the HT side of the transformer.

Atlant 14:53, 15 Apr 2005 (UTC)

Name change

The name "transformer" is ambiguous, so I propose to change the name of this article to "electrical transformer" or "electric transformer". Which name is the more correct one? --surueña 16:24, 2005 May 24 (UTC)

Then again, the electrical use of the term may be the principal use, in which case things are just fine as they are. Certainly, when I look at the two dab alternatives we have today, I think they pale in significance to this usage.

Atlant 16:34, 24 May 2005 (UTC)

may i ask what other meanings you consider significan't enough to warrant such a move rather than just a link to a disambiguation page link at the top of the article? Plugwash 16:35, 24 May 2005 (UTC)

I'm not proposing the change because the term has other common meanings, but only because it is ambiguous and this can be easily corrected moving the article (and redirections are cheap, so it is not a performance penalty). In other articles it may be not obvious what are the destination of the link, so it is better to use the "whole" name. Another option is to have one redirection (or two) from "electrical transformer" or "electric transformer" to this article. I'm not a native English speaker so, are both names correct? Thanks --surueña 18:28, 2005 May 24 (UTC)

I am a native English speaker and practicing electrical engineer, so I can assure you with considerable confidence that the devices are never called "electric" transformers or "electrical" transformers. Transformer is a perfectly un-ambiguous word and has no other common meaning aside from the toy line. In my opinion any re-naming of the article would be a serious error and would be misleading to Wikipedia users. Articles about the toys should spell out what they mean if they have "overloaded" the meaning of transformer. Thanks again for the effort to build consensus. --Wtshymanski 21:34, 24 May 2005 (UTC)

OK, last retry, I promise :-) What about a page named "Transformer (electrical)" or similar that redirects to this one? Thanks a lot --surueña 23:23, 2005 May 24 (UTC)

I think it's unecessary, and pointless, but considerably less of a problem than renaming this article. Will anyone actually type "Transformer {electrical)" when looking something up? Unlikely. Regards,--Wtshymanski 05:07, 25 May 2005 (UTC)

This article should stay under "transformer" and everything else should get parentheses. And I grew up playing with transformers (toy).  :-) - Omegatron 23:28, May 24, 2005 (UTC)

You're normal, I grew up playing with transformers (electrical) ;-) Madhu 18:57, 2 September 2005 (UTC)

Object. The largest ambiguity is due to those toys, which are disproportionately well-represented here on Wikipedia-at least the toy/movie articles are titled Transformers (toy line), etc. Neither of the proposed alternative names is more correct. I've spent a few hours looking up articles with "Transformers" and changing the links to the toys, movies, etc. where the electrical apparatus is not meant; anyone else who can spend 20 minutes on this would be a help. Thanks for asking first, though. --Wtshymanski 18:21, 24 May 2005 (UTC)

I've found a lot of Transformer (album) too, maybe there are still some of them. --surueña 23:23, 2005 May 24 (UTC)

I've got to admit, if I ever see a transformers (toy) that transforms into a transformer...I'm going to buy it. What would it's name be? "Pauwels" maybe, or "ABB"...--Wtshymanski 14:55, 11 August 2005 (UTC)

---

So I always used to think that wall warts drew power while there was nothing connected to their outputs, since they are still a connection between the two wall terminals. Of course I forgot that it is an inductor which looks like a large resistance to the AC line. Mention this? - Omegatron 02:14, May 28, 2005 (UTC)

transformator shortcircuited even at AC?

The article says: A transformer winding should never be energised from a constant DC voltage source, as this would cause a large direct current to flow That is totally clear to me.

When thinking about AC on a transformator, i still think of a shortcircuit - how can it be, that there are still 98 % of the energy transported on the "other side" rather than just flowing through the first coil and therefore shortcircuiting a current source or voltage source.

Still, when to do say a current being DC, and when AC? Is it still OK to use a AC of 0.00001 Hertz (that is around one revolution per day) on a transformator? --Abdull 13:00, 24 July 2005 (UTC)

That's a good question. To answer it, let's go back to first principles. Take a transformer, and remove the secondary winding (or, equivalently, just leave it open-circuited). You now have a single winding, the primary, wrapped round an iron core - in other words, an inductor. Now connect that inductor to a DC voltage source. As with any inductor, the current will start at zero and rise linearly with time. Most of the energy goes into building a magnetic field in the core, and just a little goes into heating the wire through resistive heating. After a while, the core will become saturated - that is, it can't contain any more magnetic energy. At this point, the device stops being an inductor, and all you have left is the resistance of the wire. You now have a low-resistance piece of wire connected across your DC voltage source, so a large current flows and the wire gets hot. This is a bad thing to do to a transformer, and will probably destroy it.

For our next trick, let's repeat the experiment (assuming the transformer is still in one piece), but using an AC voltage source instead of DC. This time, the magnetic field builds up for one half-cycle of the AC waveform, but then decays to zero and starts to build up in the opposite direction for the second half-cycle. Because the AC voltage keeps reversing, the magnetic field repeatedly builds to a finite value and then decays. If the frequency is high enough, the magnetic field will never reach its saturation value and the core will always behave linearly. If the frequency is too low, then the core will saturate before the end of the first half-cycle, and the wire will get hot. The limiting frequency depends on the construction of the transformer and the size of the AC voltage. Roughly speaking, if you halve the frequency, then you must also halve the voltage rating of the transformer primary. --Heron 15:01, 24 July 2005 (UTC)

So just to drive that point home, if the transformer has enough inductance, then it will work at that 8.64 cycles/day frequency that was being discussed. But it would require an awfully large amount of inductance (so many, many turns around an enormous core). But if it could be manufactured, it would work.

Atlant 21:20, 24 July 2005 (UTC)

explosions

I don't want to be sensationalist, transformer explosions are rare, but maybe their should be something about them in this article. There was recently a large explosion in SF. [1]

I think this would be a worthwhile addition (and I assume it mostly happens when water gets into the coolant and breaks down its insulating capability). We are starting to get to the point, though, where we might want to factor "utility transformers" (or something) out of this article into their own article.

Atlant 00:26, 22 August 2005 (UTC)

• • Looks for Answers in San Francisco Blast

• • Pacific Gas and Electric Company Will Expand Inspections of All Underground Facilities in San Francisco

Scotty