User-agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:31.0) Gecko/20100101 Thunderbird/31.6.0
Bert (et al)-
Thanks much for the analysis. It's quite interesting that k drops so
markedly. Turns out that I'm planning to use 6-8 twisted strands of
doubly-varnished 20 awg solid wire for the primary (or primaries), since
I have a spool of it and the primary voltage is relatively low. With my
scheme, it happens that that will be divided into 2 actual conductors,
one conducting current CW and the other, CCW. From what you write, it
appears as if, with my 4 driving modules connected either in series or
in parallel and using one physical 12"-dia coil-layout, either of the
modes of connection will work. If between-turn k seems too high, I'll
just spread them out a tad.
Parallel module-connection will be easier since I'll not need to
choke-isolate the mains-power inputs, each from the others.
Right now, I'm laying out a new board. I still get too-high of a
ringing spike between MOSFET conductions, even after adding a decent 10
uF snubber capacitor across my big, fat 4000 uF supply-electrolytic,
which tells me that my present board layout is too spread out. (I'll
also use another of the snubber cap's for the actual snubber.) My
circuit-board manufacture consists merely of hand-grinding away
isolation-paths through the copper laminate rather than etching away
everything but conduction-paths. Easier, at least for me, for 1-up or
even 4-up fabrication.
And BTW, I specified a fractional turn since for series connection, such
is necessary, to reach either the drive module just preceding the
starting one or just following it. The individual primaries would each
have 3-4 turns, I suppose; will have to determine that after I fire up
the new board & crank it up to full power at the ~300VDC+ input. Will
expect to draw from the mains ~ 1/4 the current that will pop the breaker.
Also BTW...it goes slowly due to too much socialization of late not to
mention advanced age.
Ken
From:
Bert Hickman <bert@xxxxxxxxxxxxxxxxxxxxx>
Date:
4/16/2015 9:02 AM
Hi Ken,
It looks like you could also use coaxial stacked or coplanar single-turn
primaries in your system and still achieve the desired multiplication of
ampere-turns. It looks to be quite difficult to achieve coupling
coefficients much above 0.85 using single-turn air-core coils without
resorting to things like bifilar winding.
To quantify this, I used Antonio's Inca program to solve for the
individual winding inductance, mutual inductance, and coupling
coefficient between a pair of identical primary windings. Looking at the
photo of your coil's 24" x 24" base, I estimated that the diameter of
your primary is about 12". I also assumed #4AWG (about 0.20" diameter)
insulated primary wire with 1/16" thick insulation. I also assumed two
identical single-turn primary windings, one stacked above other the
other (configuration 1) and one tucked outside the other (configuration
2). Windings were assumed to be separated just by their two layers of
insulation (assumed to be a total of 1/8", for a center-to-center turn
separation of 0.325". All primaries were assumed to be about 6" above
the floor.
The coupling coefficient is relatively insensitive to primary diameter
at 12" or above. For example, doubling the primary winding to 24" only
increases k to ~0.79 for either configuration. If we reduce the
insulation thickness to about 30 mils (between 24" turns), k increases
to about ~0.85 - higher, but still below 0.90.
So, it looks like you have a wide degree of latitude for either stacking
or nesting primary turns while still achieving the desired ampere-turn
scaling. No need for using partial turns.
Bert
--
Bert Hickman
Stoneridge Engineering, LLC
http://www.capturedlightning.com
***********************************************************************
World's source for "Captured Lightning" Lichtenberg Figure sculptures,
magnetically "shrunken" coins, and scarce/out of print technical books
***********************************************************************
Ken Herrick wrote:
> Hi, Greg (et al)-
>
> Thanks for the comments. A question remains, though
[snip]
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla