14

What does a black bar across a transformer symbol designate?

enter image description here

This appears in a power electronics context, and in particular in this case with substituting a DC transformer as an equivalent circuit for the guts of a DC-DC converter.

Adam Haun
  • 21,331
  • 4
  • 50
  • 91
scanny
  • 7,062
  • 5
  • 47
  • 87
  • 2
    I may be too woozy or tired right now, but, the bar as it is drawn here, to me, looks just like a transformer core drawn in badly/sideways. – Asmyldof Feb 01 '16 at 16:19
  • 2
    Ain't no such animal as a DC transformer (other than a DC-DC converter of some type.) – Ecnerwal Feb 01 '16 at 16:24
  • 2
    @Ecnerwal There is such a thing and they're cool (literally). – Spehro Pefhany Feb 01 '16 at 16:35
  • 1
    @Ecnerwal In this particular case I think the DC transformer is a construct used to develop a small-signal model of a DC-DC converter. That may even be the reason for the bar, to show it's not a normal or physical transformer. – John D Feb 01 '16 at 16:37
  • 2
    Looks like they've invented a symbol for a DC transformer, to represent the DC-DC converter. It's kind of a stupid symbol because it looks like the coils are shorted together. – Spehro Pefhany Feb 01 '16 at 16:42
  • 3
    @JohnD (and Spehro) -- Ah, you're quite right. Your comment got me thinking and I checked out Erickson 2ed *Fundamentals of Power Electronics*; on p41 he states: *"The solid horizontal line indicates that the element is ideal and capable of passing dc voltages and currents."* – scanny Feb 01 '16 at 16:51
  • 8
    I think it's the symbol for two bearded engineers trying to eat the same bar of chocolate. You may have to squint to see it. – Transistor Feb 01 '16 at 17:02
  • Yes you do need to squint to see it but how did you know that beforehand - do you often squint at computer screens LOL – Andy aka Feb 01 '16 at 17:22
  • I'm not sure if this a serious answer... What does it mean to the person actually fabricating a board with such a schematic? – nanofarad Feb 01 '16 at 18:02
  • It's definitely not serious. I've been working on a poorly-expressed SE question during the day and when I saw this I took the opportunity for some light relief. I hope you don't mind. – Transistor Feb 01 '16 at 18:05
  • I like it! :) Not sure I can make it the answer, but it definitely gave me a chuckle :) If you want you can put in the quote from the comment above and then I'll accept. No good reason a right answer can't be funny :) – scanny Feb 01 '16 at 18:33
  • 4
    Thanks for the like. Don't accept it as the answer. Some humourless high-ranker will go nuts. I've had two down-votes! It's a sad world. – Transistor Feb 01 '16 at 18:41
  • @scanny: It's been converted to a comment by Nick. Imagine if you'd accepted it. We'd both be in trouble! – Transistor Feb 01 '16 at 19:49
  • Why doesn't anybody ask "What is a DC transformer?" in the comments? Seriously, I don't understand this "DC transformer" term here. Can anyone please explain it? – hkBattousai Feb 16 '16 at 10:24
  • @hkBattousai -- As Adam mentions below, it is an ideal modeling construct useful in the analysis of switching converters. It has no physical counterpart; the math just all works out very neatly to an exact DC analog to AC transformer behaviors, like scaling voltages by turns ratio, reflecting impedances and so on. – scanny Feb 16 '16 at 18:04

1 Answers1

14

As you (and others) have said, it's the symbol for a DC transformer, which is an ideal model used to transform impedances for calculations involving switching DC-DC converters. The only source I can find for the symbol is Robert W. Erickson's book Fundamentals of Power Electronics, so it seems likely that Erickson himself invented it. I took his Coursera class a couple years ago, and there wasn't any special meaning associated with the bar. The bar is visually distinct from the (many) other transformer symbols and solid horizontal lines are often associated with DC.

UPDATE: As mentioned in his comment, Scanny tracked down Professor Erickson on Coursera and asked him about the symbol. Erickson's response:

This symbol was invented by Profs. Middlebrook and Ćuk in their classic papers on converter modeling from the 1970's. It denotes that the transformer is not a physical magnetic transformer, but rather is an ideal DC transformer. Many of us in the power electronics field have adopted this symbol.

The earliest Ćuk paper on record is "A General Unified Approach to Modeling Switching-Converter Power Stages" (Middlebrook and Ćuk 1976), an IEEE Power Electronics Specialists Conference presentation from June of 1976. This paper introduces the idea of using idealized transformers to model switching action. The IEEE copy of the paper (linked above) uses plain transformer symbols without the horizontal lines. The reprint at Scanny's link (possibly from the International Journal of Electronics in 1977, Vol. 42, Issue 6) uses both a straight line and a wavy line together -- more on that in a moment.

At the next PESC in June of 1977, Ćuk and Middlebrook presented a follow-up paper covering the model's application to discontinuous conduction mode. This paper seems to be the first use of the horizontal line by itself. It's behind the silly IEEE paywall, so I've reproduced the part that introduces the symbols here:

Excerpt from (Ćuk and Middlebrook 1977)

So the straight line (bar) indicates a DC transformer, the wavy line (sinusoid) indicates an AC transformer, and both together indicate an AC+DC transformer.

Fun fact: At that same conference, Ćuk and Middlebrook also introduced the topology known to modern engineers as the Ćuk converter.

Adam Haun
  • 21,331
  • 4
  • 50
  • 91
  • 1
    Some additional citations from Prof. Erickson: https://www.coursera.org/learn/power-electronics/discussions/l-XdOMkYEeWmXg4iQyfEyQ *This symbol was invented by Profs. Middlebrook and Cuk in their classic papers on converter modeling from the 1970's. It denotes that the transformer is not a physical magnetic transformer, but rather is an ideal DC transformer. Many of us in the power electronics field have adopted this symbol.* Perhaps first appearing here: http://www.ee.bgu.ac.il/~kushnero/temp/guamicuk.pdf. Might be worth adding to your answer. I'm accepting yours :) – scanny Feb 02 '16 at 19:47
  • Interesting! I've updated my answer with that information and more. – Adam Haun Feb 02 '16 at 23:15
  • Awesome find Adam! That reference is definitely straight "from the horse's mouth" :) I can see @SpehroPefhany 's point about the graphical disadvantages of that notation; it's difficult to make out when superimposed/overprinted like that. I expect that's why the black bar became heavier, to make it stand out visually. Nice to know the original source and to have a connection back to Middlebrook's work. I've admired his contributions in control loop compensation for a while now :) – scanny Feb 02 '16 at 23:25