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

Large Flat Coil



Original poster: "Barton B. Anderson" <bartb@xxxxxxxxxxxxxxxx>

Hi Everyone,

Today (Sunday) I took some frequency measurements of a large flat coil I've been working on (slowly). There have been some mishaps with wire purchases (beware of eBay). Anyway, I just thought I'd share for anyone interested. If no interest, hit the delete key. BTW, this is a 1/2 scale replica of Tesla's famous flat coil. I just didn't have the space for a full size replica.

Flat Coil Specs:
ID:  3.625 (hv end)
OD: 50.625 (rim grounded end)
Turns: 95
Wire: 18 awg magnet wire

Here's a couple photo's of the wound coil.
http://www.classictesla.com/flat/Img_3462_fast.jpg
http://www.classictesla.com/flat/Img_3461_fast.jpg

and other misc photo's:
http://www.classictesla.com/flat/

Measurement:
Coil set on a wicker stand providing a base height of 31.25" (coil is parallel to ground plane). Below the stand, a 36" flat metal plate was used as RF ground. I used a planewave antenna for the scope measurements which was about 5 feet from the edge of the coil. Because the flat coil is in a horizontal position, the planewave antenna at the coils height would have negligible extC. Hooked low impedance amplifier to the scope at it's base (which is the outer rim end of the winding). Nothing of course is connected to the inner winding end. Freq generator fed the z-amp. All this was measured out in the backyard. Besides the house wall, there is a cement brick wall along the entire yard about 10 feet high.

Fres measured 438.5 kHz. I tried to measure f2, but the generator just won't output a high enough frequency (a thorn in my side a long time now). Anyway, Javatc measured 455 kHz using measured radius values from center of coil to the nearest wall. I ended up having to set the wall radius value to 39" which resulted in a predicted frequency of 438 kHz. I checked it by inserting the planewave 6" below the center of the coil (providing an effective 10.2" grounded disc equivalent just below the coil). This measured 406 kHz and Javatc predicted 407 kHz.

I'll have to retest this again at another time as the backyard is narrow and there is also a good sized above ground pool within 10 feet of the coil (giant object). Front yard would have been better (hind sight). I'll be off the list for 7 days (AIT Convention in San Antonio). Because today was nice out, I wanted to at least start measuring before I decided to go further with the coil.

The coil will end up with 91 turns (as originally designed). A primary ribbon is to replace the outer 2 turns. Javatc is doing ok on flat coils it appears. This particular rim grounded configuration cannot be ran with the online version however. I ran this on an updated version I'm working on offline. The new version will allow more of these kind of geometry's, but there is also an update with primary to secondary spacing that is still being worked out (the actual proximity where both distance and height are accounted [primary inside or outside with any geometry]). The online version can of course do flat coils, but the grounded end cannot be interchanged. This seemingly simple output is actually complex, so it's taking some time to code.

Take care,
Bart