[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Space winding question



Original poster: "Barton B. Anderson by way of Terry Fritz <twftesla-at-uswest-dot-net>" <tesla123-at-pacbell-dot-net>

Hi John C.,

Tesla list wrote:

> Original poster: "John H. Couture by way of Terry Fritz
<twftesla-at-uswest-dot-net>" <couturejh-at-worldnet.att-dot-net>
>
> Bart -
>
> For #30 AWG - 1 mil insulation - closewound - 19" long
>
> The JHCTES Program
>   Bare wire dia          .01"
>   Insul Thick            .001"  (1 mil)
>   Overall dia = .01 + 2(.001) = .012" dia
>   Wire c/c    = .012"
>   Spacing     = 0  (closewound)
>   TPI = 1/.012 = 83.3 Turns per inch
>   Turns = TPI x length = 83.3 x 19 = 1582.7 Total coil turns
>
> To get this same result with your Java TC Designer Program I used .01 for
> the bare wire, .002 for the insulation (to get the same result), and 0 for
> the spacing because of coil being closewound.

Ok John, I now understand what you were getting at. Your right, JavaTC
should have included 2 * the insulation
factor. The reason it didn't was a typo mistake on my part when writing the
code. It is modeled after my Excel
version of JavaTC which does include each side of the insulation. Thank you
for pointing out the mistake. All
of my calc's have "always" included the insulation factor correctly and I
thought I had it in there. This is
why I didn't understand what the problem was. Anyway, I have updated JavaTC
to version 7.05 to correct the
insulation error. It now calcs the "above" parameters correctly. Take a
look if you wouldn't mind.

>
> Note that the overall dia includes TWO thicknesses of insulation per wire
> c/c or TPI = 1/Overall dia (as above). A spacing factor should not be used
> because that is only a wild quess. The spacing is used when
>
>        TPI x Overall dia = 1   (closewound)
>        TPI x Overall dia < 1   (spaced windings)
>        TPI x Overall dia > 1   (TPI incorrect)
>
>  If there is any doubt the turns should be actually measured. Your program
> calcs do not appear to do what you said below.
>

John, regardless of the insulation error the "method" is unchanged. Let me
restate since it appears to me that
my method is unclear to you.

1) If winding a closewound coil, use zero for the spacing.
2) If winding a spacewound coil, use the spacing factor.
3) Following actually winding the coil, measure TPI, then adjust the
spacing factor to match the calculated TPI
to the actual measured TPI regardess if it's a spacewound coil.

This is exactly what I'm trying to say below. If you thought the insulation
factor I was using in version 7.04
was designed in, please understand it was not. Just a simple mistake. Thank
you for pointing it out. My mouth
dropped open when I took a look at the code and compared it to my Excel
version and found I'd left out the
2*INS.

Do you have a problem with my method described above?

Bart A.

>
> ----------------------------------
>
> -----Original Message-----
> From: Tesla list [mailto:tesla-at-pupman-dot-com]
> Sent: Friday, March 16, 2001 7:07 AM
> To: tesla-at-pupman-dot-com
> Subject: Re: Space winding question
>
> Original poster: "Barton B. Anderson by way of Terry Fritz
> <twftesla-at-uswest-dot-net>" <tesla123-at-pacbell-dot-net>
>
> Hi John C, Malcolm,
>
> Sorry to take so long to reply.
>
> Tesla list wrote:
>
> > Original poster: "Malcolm Watts by way of Terry Fritz
> <twftesla-at-uswest-dot-net>" <m.j.watts-at-massey.ac.nz>
> >
> > On 15 Mar 01, at 7:53, Tesla list wrote:
> >
> > > Original poster: "John H. Couture by way of Terry Fritz
> > > <twftesla-at-uswest-dot-net>" <couturejh-at-worldnet.att-dot-net>
> > >
> > > Malcolm -
> > >
> > > The TPI must be coordinated
> > > with the gauge and the insulation thickness to give an overall
> > > thickness of
> > >
> > >       TPI = Turns / Overall thickness
>
> Yes, I agree. Three factors affect TPI.
> Insulation thickness, bare wire diameter, and any spacing between adjacent
> insulation. This is exactly
> what JavaTC performs  TPI = 1/(bare wire diam. + spacing + insulation). All
> the variables are taken into
> account. The problem is that one of these variables was assumed. You told
> us you used 2 mils instead of 1
> mil in the program. Why? The wire table says 1 mil (or near - actually
> 0.0009 inch for my American Wire
> Table). Why did you enter zero for the spacing? Are we assuming that a
> closewound coil has zero spacing?
> This isn't true for most real world coils. We all wind differently and to
> different degrees of accuracy.
> That's exactly the reason JavaTC has a spacing factor. This is where the
> extra mil should have been
> inserted.
>
> I've found that "actual" TPI is a little lower than calculated. I once
> measured TPI at different points
> along the coils length. After counting each turn "one-by-one", I ended up
> with less turns than calc'd.
> Why, because there is inherent spacing between turns even on a nicely wound
> (attention to detail, etc..)
> coil. This inherent spacing builds with turns. There is "no way" for a
> program to calc this factor. It
> must be measured and entered after the coil is wound. If a coiler is trying
> to wind with zero spacing,
> then zero should be used initially.
>
> I wish I had Nates coil in front of me. I'd count the turns (yep, I'm that
> boring - but ambitious!). I'd
> bet a lunch TPI is actually lower. When TPI is lower, the spacing factor
> must be higher. Inductance
> decreases, Fr goes up. The primary inductance must then decrease and tune
> at fewer turns.
>
> JavaTC was designed to enter a coils design dimensions like the rest, but
> this is only the "first step".
> Once the coil is built, the coiler can measure TPI at various points along
> the coil and average them or
> count turns (not a lot of fun, but accurate). This info is then fed back
> into the design for finer
> detailing especially in the tuning portion. Unfortunately for Nate, 30awg
> is very small and difficult to
> count. The way this is done is by inputing the basics: height, diameter,
> bare wire diameter, insulation
> thickness, and then changing the "spacing" input until TPI calc'd equals
> the coilers measured value.
> There is no other way to accomodate this "real life" factor in any program.
> But is real, and the program
> accomodates it.
>
> It also works well for space wound coils obviously. Good program for those
> arm chair cowboy days
> wondering what some of the differences are between 1/2" spacing or 1/4",
> etc..
>
> > > I believe that this is a problem with Bart's great program as I
> > > mentioned in another post. I haven't checked this out completely but
> > > was waiting for Bart's comments. The rest of his program gives the
> > > same answers as mine but Bart's has more information.
> > >
> > > The coil length is always equal to the TPI times the number of turns.
> >
> > I think not. Not in a spacewound design.
>
> Malcolm, John is right. TPI is the number of turns per inch (spacing
> included).
>
> Take care,
> Bart A.