Hi Alex,44 TPI sounds about perfect for #24 (it's what I always end up at). So many things can Fres to be other than we calc (objects nearby, humidity, measurement apparatus, etc.). Javatc tries to account for the height of objects to ground, walls, ceiling, etc.., but difficult to go beyond that in a model. Your resistance value looks pretty good. The wire adds a small change in diameter since the model actually looks at the center of the wire from one side to the other (so say a 2.26" radius). This adds about an ohm. So yes, certainly a 4.5" form radius.
Regarding the 140 Vac input: That is just the variacs ability to increase the voltage a little. My variac actually hits about 150 Vac full throttle. No need to up the cap value as it's not critical. It's just an average LTR value for static gaps your looking at. Using a variac for increased voltage basically gets you about 3" increase in spark length.
Your coil should do well. One thing I personally would like to see at some point is to increase your toroid size. I use a 6.25" x 21" toroid and breakout is no problem (and it's a very smooth toroid). Your 3" x 12" is a little small. It will work fine, but I think you would have improvement with a toroid nearer to the 6" x 20" range. Not only does this allow more energy to be stored in the top load at breakout, but also lowers the frequency to about 225 kHz, so say about 8 turns on the primary. Heck, if you had my toroid, we would have nearly identical coils! Cool! If gap is worthy (and only time will tell) the coil is good for 405" spark lengths on average at 120V and say 45" at 140V.
Javatc is certainly a great tool for design purposes. But when it comes to coil performance, experience always rules the day. I use Javatc religiously for design, but for the tricks to getting a coil to perform it's best, I tinker with tuning, making sure the toroid is sized well, more power is usually good, and ensure the gap is designed for the power applied. And the gap is one of the most important in my book. Tuning is actually least of my concerns. Depending on coil performance, tuning will need to be adjusted from time to time. I usually run about 5% high on primary inductance for spark loading. I do have a larger coil that I actually run very high on inductance (nearer to 10% high on Lpri). Javatc does not adjust for this. You have to do that on your own.
Basically, once I have a coil adjusted to specs, I'll then adjust for how the coil "runs". I want the coil to run smooth and consistent. On my little 4.5" coil (similar to yours), I run about 5% high on L. This allows me to keep the coupling as high as possible and have the coil run smooth. I ran the coil a year ago or less for a long run (between 20 and 30 minutes non-stopped). The coil ran fine the entire time with no degradation to spark lengths, no racing sparks, etc. I simply put on some music, let it go, and stood back and enjoyed the show.
Take care, Bart b alex pettit jr wrote:
Sorry ! Without Attachments ...Hi All,Thanks for the details.I agree, the ResFreq should be closer to 295. The coil dia is 4.5", and wire resistance is a total of 29.7 ohms. I did a quick turn countand it is right at 44 TPI.I will check this HP 5315B counter agains the scope today. It would be ideal to have the JAVATC model match the hardware.As I said, I built many TCs over the decades- most not too impressive in their performance. My last venture wasin the 1980s in duplicating one from a book of the 1915 era. With stained and varnished wooden cases for the transformer, glass plate cap, rotary spark gap and coil base, it looksquite authentic for that time. It is not an optimal design, but I'd rather leave it for its antique appearance and build another. The referenced coil has an 8" dia tube, 16" long and wound with string spaced enameled wire. It gives ~ 12" sparks.SO, I am excited about this coil performing a bit more impressively !Bart, from your JAVATC 'demo' it appears you overdrive your NST to 140 volts. I was considering this - is that ok ? Will it gain me anything if I do not increase the pri cap from 0.0188 to 0.0199 uF as specified ? A wider spark gap would be easy.Thanks again for the guidance, Alex PPS:I just retired from industry and began teaching a senior college level mechanical eng course and lab in mechanicaldynamics, vibration and modal analysis. The parallel between that and these electrical systems is AMAZING ! ********************************************************Hi Alex, I'm showing your coil should be nearer to 295 kHz. Your 323 kHz seems a little high to me. I know it's a small amount, but in this configuration, Javatc is always much closer. Usually in situations like this, the inputs to Javatc are not exact. For example, secondary is slightly smaller than stated (like 4.25") or things of that nature. Turns affect this also, but your 920 turn count seems about right and the secondary diameter would affect this frequency quite a bit (20 kHz just for a 1/4" difference in diameter). On the primary side, cap tolerance will affect how much primary is needed as well as the primary itself. Good to measure the cap to be sure of it's total C. In any case, 1 full turn will affect about 44 kHz on the primary side for your coil. Thus, if the tuning is close to begin with, the ideal tap will be within inches. I suspect your coil will perform well at about turn 6. When you tune the coil in at low power, you also get small sparks. As you increase power, spark lengths increase. Those sparks add a small amount of loading changing the frequency of the secondary slightly. So, a small amount of increased inductance is often added to match streamer loading. Say your coil tunes in at low power at 5.6 turns. When full power is applied (and if good streamers are forming), you may want to increase the inductance a few inches (but that is something you have to play with). Take care, Bart_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla
_______________________________________________ Tesla mailing list Tesla@xxxxxxxxxxxxxx http://www.pupman.com/mailman/listinfo/tesla