Speaking for now only of 50/60Hz toroidal power transformers, would there be any electrical difference between a transformer made from a core of rolled up strip (typical) or one made from stacked stamped laminations?

THIS (except a round version):

versus THIS:

???

I realize it would be a lot easier to make one from a rolled strip, but assuming you didn't have a roll of electrical steel to wind up, and for some reason you DID have a stack of (apparently nonexistent) circular laminations, would your stacked transformer core perform any differently than a rolled-up one of the exact same dimensions?

 

The coiled version is how Toroidal are constructed.
More efficient than the EI core winding.
Max.

 

The coiled version is how Toroidal are constructed.
More efficient than the EI core winding.
Max.

I know that, but I suspect it's just because it's easier and much more efficient than stamping out hundreds of donut slices recycling the donut holes. I'm trying to confirm if that's the only reason, or if the coil design is actually superior in electrical/magnetic performance than one made from stacked laminations would be..

 

Very interesting point. The eddy currents due to the flux change should be the same. The rolled strip core has a better flux distribution at the edges.

For nanocristalline cores (e.g. "Vitroperm") of higher power, they make U-cores (and E-cores) from rolled strip. Same with amorphous core material (e.g. "Metglas"). The new JNEX-cores (in the category "iron cores"), which I have been playing around a littel bit, provide both options. See sone pictures and design here: http://www.jfe-steel.co.jp/en/products/electrical/catalog/f1e-002.pdf

I think perfomance would be equal but I have no idea what would be easier so manufacture and/or assemble (assuming a high flexibility in the final design).

 

I just happened to have been looking into this question myself. It appears to me that the two situations are different. With large circular laminations the flux linkages induced by the winding current can travel a full circle through each lamination of the toroid without ever having to leave the lamination. On the other hand, with the continuous strip, no flux linkage can go completely around the toroid without having to jump an air gap from one turn of the spiral to the next. It would appear that the permeability of the two should be different. Comments?

 

The stamped E I (or C I) laminations are much lower assembled cost, and pretty good performance because of grain-oriented steel. A wound torroid might have better performance and no magnetic discontinuities, but the assembly costs are much higher and have more limitations on what can be done because there is no way to open the core and insert a coil.

ak

 

but the assembly costs are much higher and have more limitations on what can be done because there is no way to open the core and insert a coil.
ak

Although they are Much easier to modify by winding on a small overwind for 5v ~ 24v for e.g.
Max.

 

Hey, I'm glad someone revived this thread. It remains unanswered in my mind.

To be clear, this question is NOT about EI transformers Vs. toroidal transformers, or even square-toroid (as pictured first) vs. round-toroid transformers.
The question is regarding only toroidal tranformers, and the two different possibilities of orientation of the lamination(s) in toroidal transformers.

I have taken the opportunity to illustrate in Sketchup and I will provide my conclusion (which I am not 100% sure of).

I created two toroids of identical dimensions and identical windings; one constructed of continuous wound strip, and one constructed of stacked laminations:



Now, let's zoom in on each:







and zoom in a little bit more....





In the last two images, can you tell which is which?
I can't. And I don't think the transformers can, either.

in either case, you have a wire wrapped around a core. As the wire wraps around the core, it sees this : long flat side ("with the grain") ... side made of layers ("against the grain") ... long flat side ... side made of layers ... long flat side ...repeat.
("grain" reference to wood grain, not to "grain" as in "grain oriented steel")

In either case, it sees an equal amount of "against the grain" and "with the grain."

I do not think there is any difference. Please correct me if I'm wrong...

 

I do not think there is any difference. Please correct me if I'm wrong...

I don't see any reason for any significant difference, the amount of copper over each side of the lamination's is usually roughly equal.
Max.

 

A core material that has a permeability that varies with orientation can be optimized for flux moving primarily in one direction, such as around circle of a torroidal core. Thus, a wound torroidal core is more efficient than one made from a stack of stampings. In the stamped core, flux orientation with respect to grain changes 4 times per circuit around the core, two with the grain and two against. In a wound core it travels with the grain all the time.

http://en.wikipedia.org/wiki/Electrical_steel

ak

 

A core material that has a permeability that varies with orientation can be optimized for flux moving primarily in one direction, such as around circle of a torroidal core. Thus, a wound torroidal core is more efficient than one made from a stack of stampings. In the stamped core, flux orientation with respect to grain changes 4 times per circuit around the core, two with the grain and two against. In a wound core it travels with the grain all the time.

http://en.wikipedia.org/wiki/Electrical_steel

ak

I read the wiki page and I read your reply about 5 times. Tried to visualize what you're saying, but I don't see it in my mind's eye.
So that I might catch up to your understanding, let me ask this:
If the "stacked" core, instead of a stack of individual sheets, was a continuous flat helix (like a spiral-cut ham), then would it be the same as the rolled core?

edit: and part two of that question, if the spiral wound core, instead of being one continuous strip, were made of layers of seamless concentric tubes, would it be the same as the stacked core?

 

oh, shoot. I think I get it now. the stampings are made from a sheet. A sheet which is grain-oriented in a single direction before it's cut. Therefore after it's cut, each lamination will be grain-oriented only in one direction. The wire windings of the stacked toroid are NOT oriented only in one direction. They are oriented in EVERY direction (being wrapped around the core 360 degrees), meaning only a small percentage of them will be oriented in the grain direction. But with the rolled toroid, it is rolled in the direction of the grain orientation, therefore all of the wire windings are oriented with the grain.

Is that correct?

 

If you straightened out the toroids in your drawings into solenoids, they would obviously be identical since they differ only by a 90° rotation.

But since they are not straight, one thing I wonder about is the asymmetry between the inner surface of the donut versus the outer surface. The conventionally wound toroid layers do not see or connect both the inner and outer surfaces, whereas your stacked plates would.

 

If you straightened out the toroids in your drawings into solenoids, they would obviously be identical since they differ only by a 90° rotation.
.

That's a great way to express it. I should have thought of that in post #1. Well, I'll use it now, to explain what I think AnalogKid is saying.




Looks like exactly the same thing, right?

But they're both made of grain-oriented steel, oriented in one direction.

(arrows represent grain orientation)



when the left side is unrolled, it reveals that the grains are not all oriented in the optimal direction.

 

Are you drawing that in SketchUp? Nicely done.

Good point about the grain. I may be mistaken, but I don't think all transformer cores use grain-oriented steel for the laminations.

 

oh, shoot. I think I get it now. the stampings are made from a sheet. A sheet which is grain-oriented in a single direction before it's cut. Therefore after it's cut, each lamination will be grain-oriented only in one direction. The wire windings of the stacked toroid are NOT oriented only in one direction. They are oriented in EVERY direction (being wrapped around the core 360 degrees), meaning only a small percentage of them will be oriented in the grain direction. But with the rolled toroid, it is rolled in the direction of the grain orientation, therefore all of the wire windings are oriented with the grain.

Is that correct?

Almost, and your later drawing is a good illustration. The magnetic flux always enters the core at 90 degrees from the orientation of the winding. Once it is moving within the core, the advantage of grain orientation kicks in as shown in your drawings. But the orientation of the windings with respect to the grain is a minor contributor.

And, yes, many transformers do not use CRGO because NGRCO is cheaper. It also is less efficient, meaning more weight for the same power transfer, but for things like control transformers this is not a design driver and does not outweigh the cost savings.

ak

 

Are you drawing that in SketchUp? Nicely done.

Good point about the grain. I may be mistaken, but I don't think all transformer cores use grain-oriented steel for the laminations.

True, which brings me to the idea that spawned the question...

I believe that motor stators are typically NOT made of grain oriented steel. I've been looking for a source of beefy transformer cores for my ungodly spot welder project. Microwave transformers are becoming harder to find as microwaves move on to inverter technology. So where are we mad scientists to get our transformer cores? I've got a bin full of old 3phase induction motors in my garage, and I believe I might be sitting on top of a goldmine of multi-KVA toroidal transformer cores.

The stators aren't the optimal shape at all; winding a tube is going to be a lot harder than winding a legit toroid, but a legit toroid in those dimensions costs more than a spot welder, and then you still have to wind it, and build a welder around it.

The winding slots will likely be a pain to navigate around, but once the windings are in place, the slots should act as cooling fins. A well-placed muffin fan could blow through the center of the "statoformer" or "transformator" and keep it nice and happy, taking full advantage of the beast's awkwardly long internal surface area.

I've looked for a long time for evidence of anyone ever having tried this, and found near nil. That's both exciting and depressing for me. Usually I know when I've had a good idea, because I Google it and find 100 other people who beat me to it at least 100yrs ago. I did find one scrap of hope though; somebody on YouTube posted a video of a 1/3hp single phase stator wound as a transformer and I commented on it. I think I will give it a try when I get time. Like, next year maybe

Edit; yes, it's SketchUp

 

Get ready for sticker shock when buying raw transformer cores. Used to be able to find burned out pole pigs at the junkyard but now they are recycling not only the copper from them plus the core laminations too. Begged the local line repair depot for one but no go.

 

Get ready for sticker shock when buying raw transformer cores.

Exactly. Hence the search for unlikely candidates. Any prediction on the outcome of using a 1HP+ polyphase motor stator as toroidal transformer core?

 

Are you thinking of replacing the rotors with a wound core?
As in a wound rotor motor principle? (obviously not revolving)!
Max.