DChrist
Harmless
Posts: 1
Registered: 29-1-2008
Member Is Offline
Mood: No Mood
|
|
High Voltage
Has anyone built a solid state tesla coil. I want to build one about 4 feet high and wondered how many turns and what gauge should be on the primary
coil versus the number of turns and gauge on the secondary coil.
Also should a neon sign transformer or Ignition coil be strong enough to power a larger tesla coil
|
|
|
Twospoons
International Hazard
Posts: 1282
Registered: 26-7-2004
Location: Middle Earth
Member Is Offline
Mood: A trace of hope...
|
|
There are many websites dealing with the design process in great detail. Recent analyses show the tesla coil to operate less like a transformer, and
more like an unterminated transmission line driven at its 1/4 wave resonant frequency.
Google is your friend.
Incidentally, a solidstate coil (search "SSTC") will not be driven from a neon transformer, but direct from the mains.
Helicopter: "helico" -> spiral, "pter" -> with wings
|
|
|
microcosmicus
Hazard to Others
Posts: 287
Registered: 31-12-2007
Member Is Offline
Mood: spin up
|
|
| Quote: |
Also should a neon sign transformer or Ignition coil be strong enough to power a
larger tesla coil
|
The beauty of the Tesla coil is that it is a resonant device. If it is carefully built to
minimize losses, tuned to the peak of resonance, and impedances matched, a
surprisingly weak driving circuit will suffice to energize it fully. Of course, the lower
the input power, the longer it will take to build up the energy to the maximum value.
| Quote: |
Recent analyses show the tesla coil to operate less like a transformer, and more like
an unterminated transmission line driven at its 1/4 wave resonant frequency.
|
Tesla, Kelvin, and Helmholtz understood this point quite well in their day. The
problem was with the duller contemporaries and popularizers who misconstrued
his device as a lumped LC circuit.. Unfortunately, this misinformation persisted
for more than a generation and is only now being rectified. To get a
reasonable approximation using lumped components, one rather should
treat each turn as a separate component.
Also, because of this, the critical quantity here is not the number of
turns but the length of the secondary wire, because that is what determines the
time which it takes for a signal to travel across the transmission line (the resonant
frequency being the reciprocal of this time). To some extent, this is also influenced
by the capacitance on the end, but the main function of that capacitor is to terminate
the transmission line at an antinode.
The advantage of using a solid-state driving circuit as opposed to an
old-fashioned spark-gap oscillator or some other device is that a solid state
circuit can give a far better approximation to a perfect sine wave. Even
far better than that would be a crystal oscillator. This means that the limiting
factor for performance is going to be the quality factor of the resonator, not
the coherence of the driver.
For an excellent introduction to the theory and practice of Tesla coils,
I refer you to Richie Burnett's webpage:
http://www.richieburnett.co.uk/tesla.shtml
[Edited on 30-1-2008 by microcosmicus]
|
|
|
12AX7
Post Harlot
Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline
Mood: informative
|
|
Tesla was a smart man, but he missed one important factor: loss. If there were no loss, for example, the voltage would climb to infinity on even the
smallest resonant circuits -- one microwatt could generate megavolts! This obviously isn't the case, and the reason is because there IS loss in the
coil, due to resistance, dielectric loss and radiation, and loss in the load, due to the prodigious plasma discharges produced -- it takes a whole lot
of energy to ionize that much air!
A lumped constant model for a coil isn't perfectly accurate, but it's quite effective at the fundamental frequency, which is, after all, the frequency
used in a Tesla coil. In fact, the response curve is within decibels of the real thing, from DC to maybe a half octave or so past the fundamental
frequency. (Real resonant stubs support harmonics, which are best modeled with a time-delay functional block, which is a good description of a line
anyway. For simulation purposes, a line can be modeled with arbitrary precision by an arbitrary number of lumped constants. This follows from
calculus principles.)
| Quote: | Originally posted by DChrist
Has anyone built a solid state tesla coil. I want to build one about 4 feet high and wondered how many turns and what gauge should be on the primary
coil versus the number of turns and gauge on the secondary coil. |
That depends on power, but 18AWG for the secondary and 1/4" copper tubing for the primary are typically used.
Incidentially, the highest Q (lowest loss) for the secondary comes when it's about as tall as it is wide. I don't know why the long solenoid
persists; it seems a bad idea. It does allow more turns to be placed though.
| Quote: | | Also should a neon sign transformer or Ignition coil be strong enough to power a larger tesla coil |
I'm guessing you'll want something in the range of a few kilowatts; a large NST will suffice. NSTs go with spark gap excitation. As 30kV rated
transistors are hard to come by, there's no point in getting a NST for a SSTC. Even a VTTC would use only a few kV (MOT scale), if that. For an
SSTC, use an isolation or "control" transformer, to get something in the range of 120 to 480VAC from whatever outlet you have.
Tim
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
