Core lamination

[Tony]

Originally Posted by Juergen Knoop

the document says "equal size and temperature rise"

then that core must really be good........

[Juergen Knoop]

it allows 'better' iron and higher fluxes than EI cuts. Besides the costly iron, fitting the laminations into the bobbin is more difficult so it never entered mass production. Otherwise MD cores probably would have been in real competition with toroids.

[Juergen Knoop]

this is a chart from elrad 1/1988 concerning Ri of transformers with different cores:

[Juergen Knoop]

and some blurb on transformers at Jogis Röhrenbude:
Der Uebertrager, das unbekannte Wesen

Der erste große Vorteil ist, man kann genau die selben Spulenkörper verwenden wie beim M-Schnitt. Ein weiterer Vorteil ist, daß der Schnitt im Blech nicht in der magnetischen Hauptachse liegt sondern auf den weniger belasteten Seitenschenkeln. Das D steht für DIAGONAL so wie der Schnitt ausgeführt ist um das Stopfen des Kerns zu ermöglichen. Statt eines Luftspalts hat man hier nur einen Haarspalt.
Damit das Ergebnis optimal wird, wurden die Querschenkel um 50 % verbreitert, die Bleche nur in kornorientierter Form in der magnetischen Hauptachse sowie in dünner 0.35 mm starker Ausführung mit der bestmöglich Silizium-Eisen-Legierung eingesetzt. Damit bekommt man sehr streuarme und leistungsfähige Übertrager in kompakter Bauform. Wenn Netztrafos mit diesem Blechschnitt richtig berechnet und gewickelt werden, sind sie noch streuärmer als Ringkerntrafos.
Der einzige Nachteil der MD-Kerne ist, daß z.B. bei einem M 102 statt den Kernmaßen 102 mm x 102 mm beim MD 102 ein Kernmaß von 102 mm x 120 mm notwendig ist. Angesichts der vielen Vorteile können 18 mm mehr in einer Achse in Kauf genommen werden. Das Volumenmaß eines MD-Kerns ist auch nicht größer als ein Ringkern, denn auch hier gelten die physikalischen Regeln bezüglich des notwendigen Eisenquerschnitts und der notwendigen Windungszahl und der Drahtstärke.
Beim Ringkern hat man auch immer das Problem, was macht man mit den "runden Ecken"? Quadratische oder rechteckige Bauformen lassen sich immer bequem bei bester Raumausnutzung unterbringen. Und da bei richtig gerechneten Netztrafos mit MD-Kern die selbe Streuarmut wie bei Ringkernen erreicht werden kann, können diese problemlos eingesetzt werden.

[pieter t]

Originally Posted by Juergen Knoop

it allows 'better' iron and higher fluxes than EI cuts. Besides the costly iron, fitting the laminations into the bobbin is more difficult so it never entered mass production. Otherwise MD cores probably would have been in real competition with toroids.

MD laminations are one example of a series of laminates which differ from the lossless EI in order to improve certain parameters.
For all these alternative laminations there must be a logical application, for example when weight matters.
Mostly however it seems more economic to use EI, because transformer manufacturers have equipment to assemble the laminates into cores with the required stacking height (sometimes even with a welding option), and then they only have to buy their bulk quantity of EI laminates.
I don't believe in competition between MD and toroid because winding a toroid is a completely different procedure; when toroids hit the market it was actually done with the alternative "better" laminates, and nowadays the world wide capacity of producing toroids is enormous.
Right now Waasner in Germany is one of the few remaining manufacturers of laminates in Europe; the bulk is coming from the East. Maybe Waasner could still supply the special laminates, but cost wise it would not be able to compete with toroids.

[Juergen Knoop]

yes of course, toroids nowadays are cheap, but they weren't at the beginning. So in the past there may have been a window of opportunity for MD cuts gaining some market share.

[pieter t]

Agree.
Toroids have a big market share nowadays, and not all toroids are an improvement over one of the more classic types.
It's a pity that the better laminate types did not make it because of this economy of scale.
For my transformers I use exclusively what is in my opinion the best compromise between laminate types and toroids, that is c-cores.

[tommy1000]

Thank you for your comments, very helpfull.
Is there/was there larger sizes than MD 102 made?

[gorgon53]

In the "old days" toroids did not have ANY airgap. They where not wound grain orientated strips like nowdays. They where stanced out "discs" of unorientated steel. In strict ac application they where fine. The problem with the nowdays used "wound strip" toroid cores is that do to the fact that they are wound they can have a small airgap with fairly big local fluctations. This often unevenly distributed airgap and not particular rigid airgap can be the source of a whole lot of mechanical and electromagnetical noise especially with the high flux that grainorientated steel allows us to use nowdays to minimize weigth and cost.

[gorgon53]

A core with 2 magnetic paths has advantages over a core with only 1 magnetic path because a split magnetic path has smaller inner to outer magnetic path length difference at the same core-width. This givs the MD-core a advantage compared to the toroid. Furthermore the MD-core "window" coppercontent is in a totally different leage no industrial wound toroid could ever compete with.