Re: Variable Mutual Inductance Primary Tuning (VMIPT Sorry :o)))

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Malcolm Watts by way of Terry Fritz <twftesla-at-qwest-dot-net>" <m.j.watts-at-massey.ac.nz>

HI Terry,

On 26 Jul 2002, at 15:48, Tesla list wrote:

> Original poster: "Terry Fritz" <twftesla-at-qwest-dot-net>
> 
> 
> Hi All,
> 
> I don't know if this has been done, thought of, or discussed here before but.
> 
> My Off-Line coil project will have a pretty fixed primary caps size, fixed
> primary inductor size, fixed secondary inductance, and reasonably fixed
> secondary capacitance.  So there is a problem with tuning.
> 
> Here is the idea.
> 
> There will be two primary coils (actually two BIG single turn loops of
> copper).  They will be movable so that the two loops are either very close
> together or maybe up to 6 inches away from each other.  They will be wired
> in parallel.
> 
> When they are close together, they will have high mutual inductance between
> each other so the "total" inductance will tend to increase.  If one roughly
> assumes the inductance is proportional to N^2, I would get about 4X the
> inductance of a single loop when they are close together.

Only if they are wired in series. Consider that if k = 1 and they are 
wired in parallel, they are effectively a 2-strand piece of Litz 
wire. 
 
> Now if I move them apart so the mutual inductance is low, they will tend to
> act as two magnetically separate inductors wired in parallel and the
> inductance will be roughly 1/2 that of a single loop.
> 
> Theoretically.  I could get a tuning range of 1/2 to 4 X that of a single
> loop in this way.
> 
> Does that seem at all reasonable??  I really would not need a great range
> of tuning so if it just worked a little it would be fine.  I ran a test
> case of this on MandK, realizing that it was never made to do this type of
> stuff, but I got the results attached at the end of this post.
> 
> These coils have to be at least 1.5 inches apart where k=0.27 (M=0.24uH).
> At say 6 inches, K=0.0755 (M=0.067uH).  The coils naturally have an
> inductance of 1.23uH (12 inches diameter copper tube 1 inch thick).  The
> program gives the secondary (identical to the primary in this case) an
> inductance of 0.65uH.  
> 
> I suppose this would be like having two size full primary coils in which
> you vary the distance to tune them, which I think has been discussed
> before.  But with just single turn loops, it may work better.
> 
> Of course, I would have to make up and test something like this before I
> got too carried away, but I thought I would bring it up for ideas and
comments.
> 
> ----------
> Everything else is going very well.  I ordered up parts for the AC control
> cabinet which in this case will be pretty hollow and weigh about two pounds
> ;-)  Basically just big switches rectifiers lights and fuses.  But it
> interfaces the coil directly to the 240VAC line so it has to be pretty
> good.  It only needs to handle about 5 amps so it's not to bad at all.
> Still very simple compared to a normal TC controller.  Anyone who does not
> like hauling heavy stuff will love these Off-Line coils :o))  There just
> isn't any iron in them!  
> 
> I figure the IGBTs will dissipate a total of 20 watts and the anti-parallel
> diodes about 10 watts.  So a 780 watt coil loosing only 30 watts to the
> gap!  Trying for 95% efficiency ?)  I am using maybe ten IRG4PF50WD IGBTs
> that do 900 volts and 204 amps peak for only $7 each.  I'll get ten of them
> from DigiKey and see how it goes.
> 
> http://www.irf-dot-com/product-info/datasheets/data/irg4pf50wd.pdf
> 
> I am trying to locate ten Cornell-Dubilier 940C6W4P7K   4.7uF 600V film
> caps.  Can't find them yet and stuff like this is hard for a "little guy"
> (even with a bottomless credit card) to get in tiny quantities.  Any ideas?
>  Chris B. is trying at his end (hope hope ?)  Allied had the higher
> voltages, but not these.  I think the factory has like 200 in stock so
> maybe they sell sample quantities direct.
> 
> http://www.cornell-dubilier-dot-com/film/940600.htm
> 
> Even though they are "just" metalized 940C types, they will work fine here.
>  Ten in parallel will handle 168 amps RMS and 4920 amps peak!  Easily
> enough for our needs.  The dV/dT, inductance and series resistance is
> trivial.  Unfortunately, they may be like $20 each ?  Maybe someday, we can
> bulk buy them ?
> 
> Not sure about the secondary yet until I get the primary inductance thing
> figured out.  I have worked some on the zero crossing detection and control
> stuff but that should all be very conventional "bunch of ICs" stuff in a
> heavily shielded box.
> 
> In general, the Off-Line coil is physically very simple with little real
> hardware to it.  Lots of "completely different" ideas to it, but the goal
> is all the same.  Very strange not dealing with transformers or high
> voltages, but just really high currents and very low resistances.  Aside
> from the problem finding the caps, all the parts are easy off-the-shelf
> stuff.  Anyone should be able to make one.  You can't be quite as creative
> in just using any old parts, but you don't need many.  The primary coils
> and circuits will probably be about 10 pounds if I use a heavy frame.  The
> cost is say $50 for the secondary and top terminal, $200 for caps, $150 for
> IGBTs and control stuff.  So about $400 right now.  There will be all kinds
> of opportunities for improvements, but I am trying to keep is super simple,
> cheap.
> 
> Cheers,
> 
> 	Terry

I'm following your progess with interest. 

Regards,
malcolm