EMRFD Message Archive 11506

Message Date From Subject
11506 2015-08-20 12:32:57 swift_glen multi-hole ferrite transformers

I mostly understand a single-hole toroid inductor or transformer and how RF currents generate flux paths in the core, and how flux linkage passes RF power from one winding to another.

I can even imagine flux paths in a two-hole ferrite transformer, especially when they're wound as in EMRFD, with one "turn" going up one hole and down the other. But I see SOME two-hole transformers wound otherwise, where one turn goes up one hole, around the outside, and the next turn goes up the same hole.


Then there's the six-hole ferrites, and the mind boggles at imagining flux paths of say, a trifilar mixer transformer. Is there any reason that you don't see these cores used in this application?

-Glen

11507 2015-08-20 13:15:54 Graham / KE9H Re: multi-hole ferrite transformers
As a general rule, there is no flux coupling from one hole to the other hole in a multi-hole ferrite core.  Each hole should be considered a separate transformer core.  They are only a mechanical convenience.  You will get identical performance from separate ferrite single hole beads with the same hole size/length/wall thickness.
--- Graham / KE9H

==

11508 2015-08-20 13:31:38 in3otd Re: multi-hole ferrite transformers
the coupling between two windings made "on the outside" of a binocular transformer, like this http://rclindia.tripod.com/balun.gif

 

is not very high, I think I measured a coupling coefficient around 0.2 for two windings made in this way
11509 2015-08-20 16:54:27 swift_glen Re: multi-hole ferrite transformers
"Experimental methods" is in our namesake, so I grabbed a 6-hole core and measured inductance.
I think these 6-hole cores follow mostly the same construction, where three holes are offset from the
other three. Unlike EMRFD, I'm counting one "turn" as a wire passing through a single hole:
1 turn: 2.4uH
2 turn: 4.6uH
3 turn: 8.6 uH
4 turn: 12.3 uH
5 turn: 17 uH
6 turn: 22 uH
There seems to be SOME flux coupling, because inductance increases more than linearly as you've
suggested . But there's very little flux coupling because inductance doesn't nearly increase with turns
squared as it would for complete flux coupling.
-Glen


---In emrfd@yahoogroups.com, wrote :

As a general rule, there is no flux coupling from one hole to the other hole in a multi-hole ferrite core.  Each hole should be considered a separate transformer core.  They are only a mechanical convenience.  You will get identical performance from separate ferrite single hole beads with the same hole size/length/wall thickness.
--- Graham / KE9H

11510 2015-08-20 17:22:49 w7zoi Re: multi-hole ferrite transformers
Hi Glen,

Bravo.   You did the experiment!

73, Wes
w7zoi

11511 2015-08-20 17:27:18 Bill Carver Re: multi-hole ferrite transformers
Gasp....NOW what do we do? Coupling wasn't 0.9999 but it's not 0.0000
either. Damn reality, it's so.....inconvenient.
W7AAZ
11512 2015-08-20 22:49:02 Dana Myers Re: multi-hole ferrite transformers
11513 2015-08-21 00:15:38 iam74@rocketmail.... Re: multi-hole ferrite transformers
LOL.
Wait until you get to RF.

This is probably best explained by overlapping parallel and series inductance...but how to predict how much of each...?  I would guess in this case empiricism is certainly the best approach. As Wes says, kudos for the experiment.

Somehow I am reminded of power transformers with U-shaped laminated cores and a bar across the ends.

john
AD5YE

---In emrfd@yahoogroups.com, wrote :

Gasp....NOW what do we do? Coupling wasn't 0.9999 but it's not 0.0000
either. Damn reality, it's so.....inconvenient.
W7AAZ
11514 2015-08-21 04:24:08 Graham / KE9H Re: multi-hole ferrite transformers
I suspect that most of the coupling you are seeing is external to the ferrite, where the wires pass close to each other in the air.  The ferrite itself is a pretty good magnetic shield.
--- Graham / KE9H

==

11515 2015-08-21 08:30:45 n2msqrp Re: multi-hole ferrite transformers
Glen,

Thanks for running the test.

I have a question for the group. Glen got these results with a 6-hole core:

1 turn: 2.4uH
2 turn: 4.6uH
3 turn: 8.6 uH
4 turn: 12.3 uH
5 turn: 17 uH
6 turn: 22 uH

If we had a similar core with just 1 hole we would expect the inductance increase with turns squared:

1 turn: 2.4uH
2 turn: 9.6uH
3 turn: 21.6 uH
4 turn: 38.4 uH
5 turn: 60 uH
6 turn: 86.4 uH

So this begs the question? What is the application of those 6-hole cores versus 1 hole with multiple turns? I've seen them used as filters on power busses. Could it be that, for a given inductance, this configuration does not saturate with a given DC current versus the 1 hole multiple turn version?

Mike N2MS


Glen