Toroids

[Ian G Bennett]

Hi, a very simple question.
Is it possible to have a half turn on a ferrite toroidal core? I have always understood that a wire passed through the centre of a toroid constitutes a single turn.
Yet the diagram shown seems to contradict this.
What say you?

[GMB]

Well there is nothing like an experiment to check this kind of thing.
So I did nothing like an experiment (i.e. very crudely done).

Measuring the inductance of a raw loop and then with a ferrite clamp, then with one and two extra turns - what I get is that there is no half turn. Results consistent with 1, 2 and 3 turns i.e. ratio of about 1:4:9 after subtracting the initial open loop value.

If there is any "half turn" effect then it isn't with the wire going through but rather if the wire passes close by the core but doesn't thread it at all. That certainly has an effect and if very close could even be worth half a turn although in my quick test is was more like a quarter.

Perhaps someone with more time can verify this.

Update: Of course I am testing the wire in a loop.
I guess the result is different if the wire is straight and going "nowhere", e.g. like an aerial. Then there may well be a half turn effect.

I have always understood that a wire passed through the centre of a toroid constitutes a single turn

Which it is.

[Station X]

I don't think I'd trust that document. It refers to the toroid as being a choke, but surely the cable is part of the choke as well.

[GMB]

There is a classic problem with toroidal windings which many people get badly wrong and might relate to this idea of half turns.

The apparently neat thing that looks nice is to make your windings round the toroid and end up back at the start and then take the lead out wires from there.
The snag is that if you do this you create a "phantom turn" round the outside of the whole core, which radiates like hell and also makes the inductance a little more than you calculated. You have to cancel this turn out for proper results, which can look a bit ugly, but I have seen professionally made toroid chokes with this problem.

[Radio Wrangler]

The diagram shows that whoever did it hadn't a clue!

It's a classic mistake. Some professionals fall for it too.

However you relate the voltage of one end of that cable to the other always completes any incomplete turns.

The number of turns on a toroid is always an integer and is equal to the number of times the wire goes through the hole. Simple as that!

Two-hole beads are different, but can be treated as two toroids lying alongside each other.

The diagram is evidence that its originator and every editor and proof reader and publisher it's passed through don't know what they're doing.

David

[Tyso_Bl]

Lets have a laugh, tell us where the diagram came from, it needs a credit.

[AC/HL]

Looks like an installation manual, I doubt it's meant to be authoritative.
Without the last line no-one would have given it a second thought.

[John_BS]

Not only is it wrong in stating that each pass thro the centre = 1/2 turn, it then goes on to contradict its own definition by stating that two passes = 1.5 turns!

"That's not right. In fact, it's not even wrong"

Wolfgang Pauli

[kalee20]

Originally Posted by Ian G Bennett

Is it possible to have a half turn on a ferrite toroidal core

Yes, but I've never actually seen it done.

You would need a hole drilled half way through the core, to take the wire back, while only encompassing half the area. You would also need an additional figure-of-8 loop of wire (which may be completely isolated), to ensure flux balance.

The same would apply to a twin-hole bead, or even a conventional EI laminated core, with the wire going through one aperture and not returning through the other.

The diagram you put up, shows two turns - it's the number of passes through the centre.

When you measure inductance, you always complete a loop with the measuring instrument anyway. It doesn't matter if the loop is made really close to the core, or a few inches away, the result is the same. Though if you made a truly massive loop, inductance would start to increase simply due to the great area enclosed (this effect would be more noticeable with a low-permeability toroidal core).

You would need a hole drilled half way through the core

Or two toroids in 'series' with the half turn poking out twixt.

[kalee20]

Yes that would do. In which case forget the second toroid, for half a turn on the stacked pair! But for one and a half, two and a half, etc on the pair, it would work (with the flux balancing loop).

[Ian G Bennett]

The diagram is from Ofcom see shortcut

http://ofcom.org.uk/static/archive/r...n/cmchokes.htm

Shame on them. They are supposed to look after the radio spectrum.

[G8HQP Dave]

The old Radiocommunications Agency would not have made that sort of elementary error.