RE: double wound secondary (fwd)

["Tesla list" <tesla@xxxxxxxxxx>]

---------- Forwarded message ----------
Date: Thu, 4 Oct 2007 09:30:11 +0100
From: Colin Dancer <colind@xxxxxxxxxxxxxx>
To: 'Tesla list' <tesla@xxxxxxxxxx>
Subject: RE: double wound secondary (fwd)

My initial reaction to this thread was that x4 inductance was wrong, because
overwound parallel inductors sharing the same flux have an inductance close
to that of a single coil.  However, careful consideration shows that this
might not be the case once you consider the detailed geometry and
assumptions made.

The key question is whether you get the same number of turns on the
"overwound" coil as with the "bifilar" coil?

For a given wire diameter, the number of turns you get on an "overwound"
coil is L/d, as each layer is close wound.  For a bifilar coil made using
the same wire but with two strands twisted together, there is ever reason to
believe that the spacing will be closer to L/2d, as at least in some places
the two strands will be next to each other along the length of the coil.

As a result of this geometry, the standard formulae for inductance (with its
n^2 term) means that a close "overwound" coil will have 2x the number of
turns and 4x the inductance of a coil wound by twisting together two strands
of the same wire diameter.

Obviously if the bifilar wire has strands of 1/2 the diameter of that used
in the overwound coil then the spacing and number of turns would be the same
and hence the inductance would be the same.  The conductor area, however,
will only a 1/4 of the overwound coil so the DC resistance will be 4x higher
(assuming skin depth can be ignored) which in turn may help improve the Q of
the coil.

Without more thought or physically trying it out, I wouldn't like to comment
on the pros or cons of the technique as relating to insulation of
self-capacitance.

Colin.

-----Original Message-----
From: Tesla list [mailto:tesla@xxxxxxxxxx] 
Sent: 03 October 2007 21:46
To: tesla@xxxxxxxxxx
Subject: Re: double wound secondary (fwd)


---------- Forwarded message ----------
Date: Thu, 4 Oct 2007 03:42:36 +0800
From: Peter Terren <pterren@xxxxxxxxxxxx>
To: Tesla list <tesla@xxxxxxxxxx>
Subject: Re: double wound secondary (fwd)

Something intuitively seems wrong here with a double wound coil having 4
times the inductance.  I don't know the math but I thought that a wide
conductor had a lower inductance. This surely would be the extreme case of n
parallel inductors. Don't forget that it is not the same current that passes
through both coils in parallel (as in a series arrangement when by the
square of twice the number of turns then you would have 4 times the L).
My guess is that inductance does not significantly change with two parallel
closewound coils connected in parallel. Otherwise Litz wire would have a
phenomenal increase in inductance.

Peter
http://tesladownunder.com

> ---------- Forwarded message ----------
> Date: Tue, 02 Oct 2007 01:18:56 -0700
> From: Barton B. Anderson <bartb@xxxxxxxxxxxxxxxx>
> To: Tesla list <tesla@xxxxxxxxxx>
> Subject: Re: double wound secondary (fwd)
>
> Hi Dave,
>
> The Q isn't as neat (in my little book) as the inductance. The only 
> way possible I can account for this is if the two windings at the 
> bottom and top are in parallel and very close proximity. We know 
> inductors in parallel add. But, when current is flowing and if the two 
> inductors are in "that" type of proximity (and position) winding for 
> winding, then we now have the currents in both inductors doubling up 
> on the flux (the density is double on both windings). This would 
> certainly result in a 4x (or about) inductance. I've never done that 
> with a coil, but from an arm chair physics view, it makes perfect sense
that you would have 4X the L.
>
> One of the goofy things not always realized with the multiturn coil 
> formula's (air core or otherwise) is that that a series connection is 
> assumed. You must have 2 parallel wires in this type of close 
> proximity to achieve this. Very cool!
>
> Take care,
> Bart
>
>>---------- Forwarded message ----------
>>Date: Tue, 02 Oct 2007 01:32:13 +0000
>>From: sparktron01@xxxxxxxxxxx
>>To: Tesla list <tesla@xxxxxxxxxx>
>>Subject: RE: double wound secondary (fwd)
>>
>>Antonio
>>
>>Gary Weaver has posted several times (and I have built) a "multiple layer"
>>coil that is not specifically a bifilar wound coil.  Wind one layer 
>>_close wound_ then wind another layer on top of and in "groves" 
>>between adjacent turns of lower winding layer.
>>
>>By careful arrangement of individual wire entrance and exit into 
>>windings, maximum winding error of only +/- 1 turn (much less is 
>>practical) will occur.
>>
>>I have wound such a coil with two layers, and have noticed a MUCH 
>>higher Q then a typical "bifilar" wound coil.
>>Inductance is ~4X higher to boot.  In this case, it is equivalent of 
>>two coils close wound in parallel, R would approach
>>R/2 (proximity effects will make it larger, but still significantly 
>>less R then a single winding coil).
>>
>>Band pass testing with two winding coil revealed a bandpass so narrow, 
>>I could not fine tune VFO to maximum response, it would "jump" either 
>>side of response peak.  Gary noticed large improvement with two 
>>windings in parallel, less improvement from 2 to 3 layers in parallel.
>>
>>My coil was used on a VTTC powering a CO2 laser.
>>
>>Regards
>>Dave Sharpe, TCBOR/HEAS
>>Chesterfield, VA. USA
>>
>> -------------- Original message ----------------------
>>From: "Tesla list" <tesla@xxxxxxxxxx>
>>
>>
>>>---------- Forwarded message ----------
>>>Date: Mon, 1 Oct 2007 10:06:33 -0400
>>>From: Scott Bogard <teslas-intern@xxxxxxxxxxx>
>>>To: Tesla list <tesla@xxxxxxxxxx>
>>>Subject: RE: double wound secondary (fwd)
>>>
>>>
>>>Thanks guys,
>>>     I understand much better now.
>>>Scott Bogard.
>>>
>>>
>>>
>>>>Date: Sun, 30 Sep 2007 08:35:57 -0600
>>>>From: tesla@xxxxxxxxxx
>>>>To: tesla@xxxxxxxxxx
>>>>Subject: Re: double wound secondary (fwd)
>>>>
>>>>
>>>>---------- Forwarded message ----------
>>>>Date: Sun, 30 Sep 2007 11:06:50 -0300
>>>>From: Antonio Carlos M. de Queiroz <acmdq@xxxxxxxxxx>
>>>>To: Tesla list <tesla@xxxxxxxxxx>
>>>>Subject: Re: double wound secondary (fwd)
>>>>
>>>>Tesla list wrote:
>>>>
>>>>
>>>>>---------- Forwarded message ----------
>>>>>Date: Sat, 29 Sep 2007 18:24:54 -0400
>>>>>From: Scott Bogard <teslas-intern@xxxxxxxxxxx>
>>>>>To: tesla@xxxxxxxxxx
>>>>>Subject: double wound secondary
>>>>>
>>>>>
>>>>>Hey everybody,
>>>>>     I know, this topic has been covered an awful lot in the 
>>>>>archives, I looked; but there seems to be some disagreement, 
>>>>>weather it is good or bad.  It seems lately, everybody now thinks 
>>>>>it is not such a bad idea, as it decreases the resistance, 
>>>>>therefore increasing output.  But, from what I know of formulas and 
>>>>>such, two inductors (since a coil is essentially an
>>>>>inductor) in parallel decreases the inductance, which should 
>>>>>decrease voltage out?  Does this situation not apply with a 
>>>>>transformer, or does the resistance decrease make that much of a
>>>>>
>>>>>
>>difference to make up for it?
>>
>>
>>>>>Or, does nobody actually know why it works so well?  Just curious 
>>>>>(and considering double winding my 6-in secondary). Scott Bogard.
>>>>>
>>>>>_________________________________________________________________
>>>>>
>>>>>
>>>>Two coupled inductors in parallel, with bifilar winding and high 
>>>>coupling (M=sqrt(L1*L2)), act as a single inductor with L=L1=L2=M.
>>>>The proof is easy.
>>>>It remains to be seen if the loss is smaller than when using a 
>>>>single wire with the same area of the two wires used in the bifilar
>>>>
>>>>
>>winding.
>>
>>
>>>>You can probably use a simulator as Javatc, that calculates wire 
>>>>losses, considering a single coil with the total number of turns 
>>>>with the two wires, and then divide the resulting resistance by 4 to 
>>>>obtain the resistance of the bifilar coil. Compare then with what 
>>>>happens with a wire with doubled area and single winding. In 
>>>>principle, I would expect similar results, or advantage for the 
>>>>single wire winding because there would be space between the turns,
>>>>
>>>>
>>reducing the proximity effect, if the winding length is kept.
>>
>>
>>>>Note that using the same winding length you can use a wire with 
>>>>twice the diameter, and so four times the area. The losses must be 
>>>>smaller in this case.
>>>>
>>>>Antonio Carlos M. de Queiroz
>>>>
>>>>
>>>>
>>>>
>>>>
>>>_________________________________________________________________
>>>Explore the seven wonders of the world 
>>>http://search.msn.com/results.aspx?q=7+wonders+world&mkt=en-US&form=Q
>>>B
>>>RE
>>>
>>>
>>>
>>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
>