[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: AC Resistance of wires - was 8 kHz Tesla Coil
- To: tesla@xxxxxxxxxx
- Subject: Re: AC Resistance of wires - was 8 kHz Tesla Coil
- From: "Tesla list" <tesla@xxxxxxxxxx>
- Date: Thu, 06 Oct 2005 09:06:54 -0600
- Delivered-to: testla@pupman.com
- Delivered-to: tesla@pupman.com
- Old-return-path: <vardin@twfpowerelectronics.com>
- Resent-date: Thu, 6 Oct 2005 09:07:44 -0600 (MDT)
- Resent-from: tesla@xxxxxxxxxx
- Resent-message-id: <hlOOe.A.f6F.-2TRDB@poodle>
- Resent-sender: tesla-request@xxxxxxxxxx
Original poster: boris petkovic <petkovic7@xxxxxxxxx>
Hi Bart,
-----
>My comment was aimed relative to
> the predictions
> mentioned. I expect Gerry's measurements to be far
> below the
> predictions. Doesn't mean the predictions are wrong
> as losses
> elsewhere cannot be easily separated (the age old
> problem). But as we
> tackle one loss at a time, eventually we will be
> able to break them
> all down, well, if were persistent.
------
IMHO,the best and probably the "only" way we can
separate the losses and distinguish which is which,
would be some combination of coil Q msms and accurate
enough simulation of losses due to overall H-field
(corresponding current joule heating losses )
presence.
That means engaging adequate software and hardware for
big numbers crunching to model the thing:).
Hence,my proposal of using relatively large
counterpoise ,at least as long as we are interested in
exploring the highest Q coils operated
outdoors,against the ground plane.
Once the computer simulation data are obtained ,we
should be able to extract from Q msms in-situ,how much
goes to dielectric losses.
Of course ,there are also other ,purely
experimental,approaches.
Say to build somehow formless/naked secondary and
compare it with the identical one wound on form and
having wire insulation.
Not an easy task for my money.Crazy idea isn't it?
Take care,
Boris
__________________________________
Yahoo! Mail - PC Magazine Editors' Choice 2005
http://mail.yahoo.com