EMRFD Message Archive 2753
Message Date From Subject 2753 2009-03-03 12:41:33 Art winding binocular cores
This is an awesome group, thanks to all who maintain it and keep it spam free.
I have a question about winding binocular cores.
My winding instructions say to wind the secondaries bifilar, on top
of the primary winding...so, my core will have 3 windings, all with
the same number of turns in each winding execpt that 2 of the
windings are bifiliar and connected to make a single winding after
the core is done.
Is it 'proper' or generally accepted to wind XX turns of trifilar
(all at the same time) in order to make the winding easier and less
time consuming???
-------------------------------------------
As a follow up, I'm directed to wind 20 turns of bifilar to make a
single winding with a center tap. Then, I am directed to wind another
40 turns on top of the bifilar winding for the other side of the
transformer. In this example, is it just as proper to wind 4
quadfilar coils all at the same time (and connect two of the 20 turn
windings to make an equivelant 40 turn winding, leaving the center
tap to float).
Yes, I understand the turns have to be cross connected so the phase
is correct, so this is not my question.
It seems to me I end up with the same end result in both cases above,
although the electrical performance will be somewhat different due to
parasitic capacitances between the windings.
Is winding toroids in this manner acceptable/generally accepted as ok?
I have a large amount of these transformers to wind, hence my
interest in winding the turns all at once rather than the slower
method suggested by the original designer.
Thanks,
Art2754 2009-03-03 13:35:15 GrahamH@austin.rr... Re: winding binocular cores
With respect to both of your questions, if there is a difference
from the core wound per the instructions, the difference will
be at the high end of the frequency range.
Based on reading some papers that Chris Trask has written on
magnetics, I would say that trifilar is a legitimate way
to maintain tight coupling between the coils, and equal
currents in each, so your thought to wind as trifilar,
then connect as appropriate in your first proposal is
good.
Quadfilar does not maintain the same coupling between
the different windings as does bifilar or trifilar, so
in the case of your second proposal, it may or may not
yield the result you would want, and if there is a
problem, it will be at the high frequency end, so
check coupling loss and impedance there.
If not acceptable, I would wind it like the instructions.
--- Graham / KE9H
==
-2755 2009-03-03 13:38:38 Wes Hayward Re: winding binocular cores
Generally yes, it's just fine to start with a trifilar or even
quadrafilar winding, and then wire them to create the desired
transformer afterward. I came upon some surplus transformers that
consisted of 12 quadrafilar turns on toroids with permeability a bit
higher than -43 material. The windings were with a four wire colors,
so the parts were especially handy to use. I built all kinds of
stuff with them until they were gone and I had to go back to winding.
There may be some situations where you do not want the transmission
line properties that you would get with a "primary" winding that
tightly coupled to a secondary. I suspect that this would be a rare
situation.
The arguments apply equally to toroids and binocular cores. The
balun (binocular) cores differ mainly in having holes that are so
small that the wire almost fills the hole. This tends to push the
series leakage L down, far below the parallel inductance, as shown in
the model of EMRFD Fig 3.74.
Our colleague in this group, Chris Trask, has written a lot of great
stuff about RF transformers. Perhaps he will have a comment2756 2009-03-03 14:25:35 Chris Trask Re: winding binocular cores
> My winding instructions say to wind the secondaries bifilar, on top
> of the primary winding...so, my core will have 3 windings, all with
> the same number of turns in each winding execpt that 2 of the
> windings are bifiliar and connected to make a single winding after
> the core is done.
>
> Is it 'proper' or generally accepted to wind XX turns of trifilar
> (all at the same time) in order to make the winding easier and less
> time consuming???
>
option. Trifilar can also be used to make 3:1 and 3:2 unun transformers and
3:1 balbal transformers with good performance.
>You should never use quadrifilar twisted wire as the coupling between
> As a follow up, I'm directed to wind 20 turns of bifilar to make a
> single winding with a center tap. Then, I am directed to wind another
> 40 turns on top of the bifilar winding for the other side of the
> transformer. In this example, is it just as proper to wind 4
> quadfilar coils all at the same time (and connect two of the 20 turn
> windings to make an equivelant 40 turn winding, leaving the center
> tap to float).
>
the individual wires is not even. It's better to make two windings of
bifilar wire along the outside of the core and then cross-connect them so
that the currents and voltages are better equalized.
>It depends. If you're using small gauge twisted wire that will conform
> Is winding toroids in this manner acceptable/generally accepted as ok?
>
to the shape of the toroid, then that's okay. If the wire is too large to
be pressed against the toroid, then you're better off using Sevick's
parallel wire method. He also makes the distinction between parallel and
twisted wires in his book(s).
Chris
,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_2757 2009-03-03 20:16:33 Art Re: winding binocular cores
the toroidal and binocular cores.
It's a much more complicated issue than it appears on the surface,
and I appreciate the direction.
I have one more question, I should have asked before I think, but
thought the answer was pretty obvious, so I didn't ask::> Needless to
say, I was WRONG!
In the case of a 78 turn toroidal inductor, 1 single strand layer,
(specified by the designer)......... The same inductor wound with 39
turns of bifilar did work better, but the higher frequency rejection
was down some. Would this inductor be better wound with 36 turns of
trifilar?????
Again, thanks all.
Art2758 2009-03-04 06:04:38 Chris Trask Re: winding binocular cores
>In the case of a 78 turn toroidal inductor, 1 single strand layer,
>(specified by the designer)......... The same inductor wound with 39
>turns of bifilar did work better, but the higher frequency rejection
>was down some. Would this inductor be better wound with 36 turns of
>trifilar?????
>
You may want to use an inductance meter or rig up a test set with a signal generator and scope to determine what the inductance is.
Chris
,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_