Barkenhausen Effect In Laminations

[CJ]

Came across an interesting phenom while working with current transformers.

There are clumps of magnetic domains in core steel used to stamp lams.
Depending on the quality, there could be some rather large clumps.

Trick is, they can get hung up, they might need a little more juice to get them to switch.
This only shows up if you move up the B-H curve Very slooowly.

What happens is, say you inject a current into the primary.
The voltage across the burden resistor on the output of the CT will not form a straight line when plotted against input current.

So when the clumps flip around, you get irregularities on your graph, something that has been driving me nuts for years until now.

They say to use a pen when doing lab work, now I know why.
If you "cheat" your data points closer to being linear, with a pencil, you just erase.
And you erase the Barkenhausen effect that you are trying to dismiss as sloopy lab equipment or operator error.

With a pen, when you discover this effect, you can go back and see your that chicken scratched graph points that got shoved over a little bit were really right in the first place. So always use pen, and cross stuff out.

Here is a pic of what I am talking about. If you slowly increase the primary current, you can see the effect as an un smooth B H curve, the irregularities due to Mr. FunkenBarker

How this has anything to do with audio is up to you people to figure out.
Don't shoot the messenger!

cj

Kind of like discrete electron orbits, only a lot closer together.
This is a good reason to not be wasteful and slam the core with close to what it will take in the way of B max.
That way, you muscle thru this stuff with enuff horsepower to negate the effect, whatever that effect might be.

 

[bcarso]

I like to mention Barkhausen noise to people who think transformers are noiseless.

 

[CJ]

I think I know what you mean, Mr. JBL.
Brad,who do you think was smarter of the Lansing Bros,
, James, or Altec
:razz:

Hey, do you know that freak from FGlorida, the loud speaker guy who was full of more S than me, from florida. I want t0o say Biull Roberstosn, but it has been a while.

He was alwyas getting into beefs abiout mastering credits with rthat aged mmotown freak bob geldof , of b9ib ludwig or some crak f5r4eak who watched ove eDDIE bIONGO jAMERSON

Butnthe guy knew his stuff.
That big, long, console thast JBLK produs=ced, he had All the specs.
Foleded horn mahogany, 1500 back the = ferd falcon...

 

[Nele]

How does this relate to people saying you need to over-rate your transformers? Especially in (diy-)hifi circles it's in vogue to for instance use a 300VA transformer for driving two 5534's....

 

[CJ]

I really can"t relate to that question right now, as it is worded in a , hence werth, un givemiavble syntax. (really weird question)

But you need to check this. Unless you can come up with stuff like this in the studio at moments notice, than it really does not matter what equipment you build:


http://www.youtube.com/watch?v=hEjcK9OKSmU

 

[Nele]

weird question? me? ask weird questions? waddayamean, buddy! dang homes.. hehe.

I was just thinking, if you use a hugely overqualified transformer, capable of delivering about 125A, and you draw only 50mA, doesn't this give a big Barkenhausen effect?

Or am I missing the point completamento?

Peace out,
c

 

[Nele]

Ah yes, ye olde brethren Van Halen..

 

[SSLtech]

I met Bill Roberts in Skycraft in 2004; I had my LA-2a with me; he liked it a lot... He's an interesting guy, -offered to master a brass band album for me. He's moved since then, but I did repair an AKG-D12 for his buddy since.

Google Bill Roberts Precision Mastering, I reckon you'll find him somewhere.

There's a great guy in the central CA Valley (can't say his name) who you'd also dig greatly, if you're into that sort of thing. He builds microphones, and to date the BEST-SOUNDING set of horn-loaded speakers that I've EVER heard.

EVER.

-And I am not a big fan of horn-loaded speakers, so that's some testament. -Not 100% certain that I'd still not prefer a soft-dome setup for my own critical listening purposes, but of all that have tried, this guy has come the closest. -He has a nice, wicked sense of humour, also... that helps. -I think that the first time I ever met him was about ten years ago, when I slid around a corner of the building sideways in a cabrio, in a cloud of tire smoke, with a screaming girl in the back who was convinced that we were all going to die. -After it scraped to a halt, I hopped out and insulted him, and we've been firm friends ever since.

I think Mitch was the smartest Lansing Brother.

Keef

 

[NewYorkDave]

My favorite brother was Groucho. And the name is Barkhausen... Heinrich Barkhausen.

 

[pucho812]

I always liked Harpo. Oh yeah and curly. :thumb:

 

[Guest]

[quote author="bcarso"]I like to mention Barkhausen noise to people who think transformers are noiseless.[/quote]

It is kind of modulation by noise...

 

[beatpoet]

[quote author="CJ"]Came across an interesting phenom while working with current transformers.

There are clumps of magnetic domains in core steel used to stamp lams.
Depending on the quality, there could be some rather large clumps.

Trick is, they can get hung up, they might need a little more juice to get them to switch.
This only shows up if you move up the B-H curve Very slooowly.
[/quote]

I couldn't find a lot of info on this...

The magnetic permeability affects flux density; but only for certain amplitudes?

 

[Guest]

CJ, you've opened a box of Pandora; now wait for digital simulation boom of this effect to get "Neve Sound" :green: :green: :green:

 

[CJ]

funny how things time out, we are getting laminatons from China, do you think they know about this?

I think not.Interesting to compare chinese steel with American steel, Barkenhausen wise.



\

 

[beatpoet]

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

http://jnaudin.free.fr/spgen/barkhausen.htm

http://pil.phys.uniroma1.it/~zapperi/bark.pdf

http://www.exo.net/~pauld/activities/magnetism/audiblemagneticdomains.html

It appears that the noise created by the realignment of the magnetic domains is not dependent on any controlled input signal.

I wonder how one would create a controlled test for determining the properties of these magnetic domain shifts between laminations of differing transformers?

CJ, are you using a slow DC sweep in both directions to measure this effect? I'm wondering if the same Barkhausen noise increases in only frequency [of occurance/audibility of the same Barkhausen noise] upon subjecting the coil in question to increasingly higher frequencies [Hz of signal?]

I would like to see a full shootout between some vintage lams and the modern China lams.

A side note: If one heated nickel-iron steel to the melting point and flash-quenched it in cool water, perhaps one could create a near-perfect crystalline structure in the lam steel. This process usually creates micro-abnormalities near the surfaces and edges of a steel structure, so some mild stock removal would be needed to create the 'perfect' lamination.

It would be interesting to hand-create a set of laminations in this manner; I would think that the manufacture costs would be far too high to even find a production transformer created to this level of perfection.

However, I can see problems with even a perfectly aligned magnetic domain in a lamination; that is the magnetic field is not linear and the magnetic domains would be.

 

[mikep]

interesting stuff guys.

anyone know how this applies (or doesnt apply) to amorphous core materials?

mike p

 

[beatpoet]

I'm wondering how they manufacture those cores (amorphous). Does the crystalline structure follow the curvature of the core, or does only form a cross-section?

I wonder if anyone has done a laminated version?

Doesn't the crystalline structure of steel adhere to magnetic flux lines if subjected to a field while quenching?

A real manufacturing challenge to get right, I'm sure.

 

[CJ]

lundahl puts out a new amorphous design every year, at least.


the lams a really thin, so the K factor goes to hell, a lot of useless air, but i guess it is worth it.

i can nevcer get a straight answer out of Keven at K and K"

So Kevin, do these amprphous core really sound better?"

Kevin: "Well, the B max does not get re intoduced into the zobel tweakers cause the domains will not allow it"

"OK, Kevin, thanks a lot, guy!"

:shock:

I guess if you sell them, you are supposed to non comment.

Remeber thi stuff is almost a hundred yeras old, Barken Burger and all that.

Read the Sowter article, Soft Mag Materials at there site.

 

[beatpoet]

[quote author="CJ"]lundahl puts out a new amorphous design every year, at least.


the lams a really thin, so the K factor goes to hell, a lot of useless air, but i guess it is worth it.
[/quote]

Everything I think of has been developed and refined 100x over.

Sometimes I wonder how I survive with such meat-and-potatoes knowledge of what's out there.

 

[beatpoet]

I like TP.

Sure, sometimes his technicals are agonizingly drab, but I really do appreciate the way he uses philosiphy to excite the emotional tendrils of the cow-like album-buying radio-listening masses.