Notices
Results 1 to 9 of 9
Like Tree1Likes
  • 1 Post By jocular

Thread: Tesla Coil Problems :/ any help would be cool

  1. #1 Tesla Coil Problems :/ any help would be cool 
    New Member
    Join Date
    Apr 2013
    Posts
    1
    For a physics project, my group and I are building a 2 foot tesla coil. we've got a nice spark gap, and there's current going through our primary coil, however, there's nothing going through our secondary. Our main concern though, is that our spark gap quits sparking after just a few seconds, then our nst makes a low buzzing sound. didn't know what to think about that until we heard it bubbling on the inside and it started smoking. umm yeah. any help would be appreciated.


    Reply With Quote  
     

  2.  
     

  3. #2  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,641
    Quote Originally Posted by DBitty View Post
    For a physics project, my group and I are building a 2 foot tesla coil. we've got a nice spark gap, and there's current going through our primary coil, however, there's nothing going through our secondary. Our main concern though, is that our spark gap quits sparking after just a few seconds, then our nst makes a low buzzing sound. didn't know what to think about that until we heard it bubbling on the inside and it started smoking. umm yeah. any help would be appreciated.
    First, stop the smoking.

    You won't get anything out of the secondary unless it's tuned to the same frequency as the primary. And the primary, of course, must be loosely coupled to the secondary, otherwise it will be impossible to get two synchronous resonances. How are you measuring/tuning the resonances?


    Reply With Quote  
     

  4. #3  
    Forum Professor
    Join Date
    Jan 2013
    Posts
    1,773
    Quote Originally Posted by tk421 View Post
    Quote Originally Posted by DBitty View Post
    For a physics project, my group and I are building a 2 foot tesla coil. we've got a nice spark gap, and there's current going through our primary coil, however, there's nothing going through our secondary. Our main concern though, is that our spark gap quits sparking after just a few seconds, then our nst makes a low buzzing sound. didn't know what to think about that until we heard it bubbling on the inside and it started smoking. umm yeah. any help would be appreciated.
    First, stop the smoking.

    You won't get anything out of the secondary unless it's tuned to the same frequency as the primary. And the primary, of course, must be loosely coupled to the secondary, otherwise it will be impossible to get two synchronous resonances. How are you measuring/tuning the resonances?
    This may not be quite right. If the voltage source of the primary is 60 HZ, and the secondary resonates with it's capacitance at perhaps a million HZ, then obviously there is a vasr frequency difference.

    The idea is for the spark gap to interrupt primary current flow, like a switch, which allows time between "switchings" for the collapsing primaryy field to induce a voltage across the secondary coil many, many times higher, due to the turns ratio between the coils. Thus, the primary "kicks" energy into the secondary relatively slowly, compared to the resonating "tank" circuit formed by the secondary inductance and it's paralleled high-voltage capacitors, which resonate at extremely high frequency. These are often a trouble spot. I use 12 oz. beer bottles filled with salt water, an electrode reaching down into it, the bottles being wrapped with aluminum foil on the outside, about half-way up; the bottles are then stood upright in a plastic basin having about 5 inches of salt water in it. The top electrodes are connected together, while the outer "electrodes", the foil, are all connected via the salt water. A bit of light oil is floated on top of the water to eliminate evaporation, and insulate the conductive water from the varying amount of water vapor in the air, which affects overall capacitance.

    I'll talk more about my designs, if anyone asks for it; do not hesitate to do so! jocular

    Edit: (red) There is essentially no resonant circuit in the primary, as no significant capacity exists in it, only inter-electrode capacitance of the transformer used as a voltage source. Luminous tube transformers provide a good starting point for this purpose, as they are available with output voltage of 15,000, with primary voltage needed 120.
    Reply With Quote  
     

  5. #4  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,641
    Quote Originally Posted by jocular View Post
    Quote Originally Posted by tk421 View Post
    Quote Originally Posted by DBitty View Post
    For a physics project, my group and I are building a 2 foot tesla coil. we've got a nice spark gap, and there's current going through our primary coil, however, there's nothing going through our secondary. Our main concern though, is that our spark gap quits sparking after just a few seconds, then our nst makes a low buzzing sound. didn't know what to think about that until we heard it bubbling on the inside and it started smoking. umm yeah. any help would be appreciated.
    First, stop the smoking.

    You won't get anything out of the secondary unless it's tuned to the same frequency as the primary. And the primary, of course, must be loosely coupled to the secondary, otherwise it will be impossible to get two synchronous resonances. How are you measuring/tuning the resonances?
    This may not be quite right. If the voltage source of the primary is 60 HZ, and the secondary resonates with it's capacitance at perhaps a million HZ, then obviously there is a vasr frequency difference.

    The idea is for the spark gap to interrupt primary current flow, like a switch, which allows time between "switchings" for the collapsing primaryy field to induce a voltage across the secondary coil many, many times higher, due to the turns ratio between the coils. Thus, the primary "kicks" energy into the secondary relatively slowly, compared to the resonating "tank" circuit formed by the secondary inductance and it's paralleled high-voltage capacitors, which resonate at extremely high frequency. These are often a trouble spot. I use 12 oz. beer bottles filled with salt water, an electrode reaching down into it, the bottles being wrapped with aluminum foil on the outside, about half-way up; the bottles are then stood upright in a plastic basin having about 5 inches of salt water in it. The top electrodes are connected together, while the outer "electrodes", the foil, are all connected via the salt water. A bit of light oil is floated on top of the water to eliminate evaporation, and insulate the conductive water from the varying amount of water vapor in the air, which affects overall capacitance.

    I'll talk more about my designs, if anyone asks for it; do not hesitate to do so! jocular

    Edit: (red) There is essentially no resonant circuit in the primary, as no significant capacity exists in it, only inter-electrode capacitance of the transformer used as a voltage source. Luminous tube transformers provide a good starting point for this purpose, as they are available with output voltage of 15,000, with primary voltage needed 120.
    There are several misconceptions in your post, ranging from your definition of a Tesla coil, to a general ignorance of how coupled resonators work, and a neglect of what the spectrum of an interrupted "60Hz" drive signal looks like. Let's take a look at these one at a time.

    First, a Tesla coil's primary needs to be driven by a signal with spectral components at the same frequency as the resonance of the secondary. If you do not satisfy that requirement, it is not a Tesla coil. Period. It may be some other kind of coil, but it ain't a Tesla coil. I am using the definition provided by Mr. Tesla himself.

    Second, let's look at how the primary is driven. It is not driven by a pure 60Hz sinusoid, by your very own description. The "60Hz" signal is just the fundamental component, and is more or less just there for the ride. What does the actual work in a Tesla coil is not the 60Hz energy, but the higher-frequency components associated with the spark-gap-interrupted waveform. That was the key contribution by Tesla -- he showed how to convert low-voltage, low-frequency mains power to high voltage at a high frequency. Only the spectral component at the resonant frequency of the secondary does anything useful in a Tesla coil. All of the other components represent waste, essentially. That observation leads us to a third important point:

    Making the primary resonant at the same frequency as the secondary is important for efficient generation of the one spectral component that matters. Again, you don't have to resonate the primary, but then you don't have a Tesla coil, and efficiency will be poor.

    Finally, if you couple two resonant systems together, whether you end up with a synchronously tuned fourth-order system (which is what a Tesla coil is, to a lumped approximation) or not, depends very much on the coupling strength. If tightly coupled, the resonances are significantly perturbed, leading to a splitting of mode frequencies, and lousy Tesla coil operation. The need for loose coupling is why all photos of real Tesla coils (as built by Tesla, as described by Tesla, and as prescribed by Tesla) show a large spacing between the primary and the secondary.

    You don't have to take my word for this. You can derive for yourself the equations for the mode frequencies of coupled resonators. It's not hard, just a little tedious (but not so much so that my freshmen can't handle it with a little prodding). To make things simple, start with two identical resonators and derive the eigenmodes as a function of coupling. Either electrostatic or magnetic coupling will show what you want to show. You can make things easier on yourself by recognizing that you can readily identify one eignemode with a common-mode (even-mode) excitation, and the other with a differential-mode (odd-mode) excitation.
    Reply With Quote  
     

  6. #5  
    Forum Professor
    Join Date
    Jan 2013
    Posts
    1,773
    Quote Originally Posted by tk421 View Post
    [
    There are several misconceptions in your post, ranging from your definition of a Tesla coil, to a general ignorance of how coupled resonators work, and a neglect of what the spectrum of an interrupted "60Hz" drive signal looks like. Let's take a look at these one at a time.
    Okay, Okay, I relent! But please, explain to me if given a transformer, single-phase, 120 V primary, 15 KV secondary, regulation being such that the secondary will deliver harmlessly 50 MA while short-circuited (typical gaseous tube transformer for lighting use), and I place a spark gap in series with the secondary, such that it's short-circuit current flows back and forth through the gap at 60 HZ, what capacitance is present capable of producing harmonic frequencies of reasonably useful amplitude? jocular
    Neverfly likes this.
    Reply With Quote  
     

  7. #6  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,641
    Quote Originally Posted by jocular View Post
    Okay, Okay, I relent! But please, explain to me if given a transformer, single-phase, 120 V primary, 15 KV secondary, regulation being such that the secondary will deliver harmlessly 50 MA while short-circuited (typical gaseous tube transformer for lighting use), and I place a spark gap in series with the secondary, such that it's short-circuit current flows back and forth through the gap at 60 HZ, what capacitance is present capable of producing harmonic frequencies of reasonably useful amplitude? jocular
    Without knowing all the details of your configuration, all I can do is make some guesses. If you are using a standard NST, there is a significant amount of leakage inductance and capacitance associated with its own secondary windings, so there is a definite resonant frequency associated with the NST itself.

    The spark gap does interesting things to the spectral content of the current. It acts much like a modulator, actually, interrupting the 60Hz mains power at a rate determined by a combination of the breakdown voltage and the reactances connected to the spark gap (it's perhaps more intuitively appealing to say that the 60Hz mains power modulates the spark gap current). Since it sounds as if you are not using any tuning capacitors in the primary circuit of the Tesla coil, you're relying on adjusting the spark gap spacing to hit the resonant frequency of your fixed secondary. That can be made to work, after a fashion, but you'd get much more impressive results if you were to tune the primary explicitly, rather than relying on parasitic effects.
    Reply With Quote  
     

  8. #7  
    Forum Professor
    Join Date
    Jan 2013
    Posts
    1,773
    Quote Originally Posted by tk421 View Post
    Quote Originally Posted by jocular View Post
    Okay, Okay, I relent! But please, explain to me if given a transformer, single-phase, 120 V primary, 15 KV secondary, regulation being such that the secondary will deliver harmlessly 50 MA while short-circuited (typical gaseous tube transformer for lighting use), and I place a spark gap in series with the secondary, such that it's short-circuit current flows back and forth through the gap at 60 HZ, what capacitance is present capable of producing harmonic frequencies of reasonably useful amplitude? jocular
    Without knowing all the details of your configuration, all I can do is make some guesses. If you are using a standard NST, there is a significant amount of leakage inductance and capacitance associated with its own secondary windings, so there is a definite resonant frequency associated with the NST itself.

    The spark gap does interesting things to the spectral content of the current. It acts much like a modulator, actually, interrupting the 60Hz mains power at a rate determined by a combination of the breakdown voltage and the reactances connected to the spark gap (it's perhaps more intuitively appealing to say that the 60Hz mains power modulates the spark gap current). Since it sounds as if you are not using any tuning capacitors in the primary circuit of the Tesla coil, you're relying on adjusting the spark gap spacing to hit the resonant frequency of your fixed secondary. That can be made to work, after a fashion, but you'd get much more impressive results if you were to tune the primary explicitly, rather than relying on parasitic effects.
    Thank you for indulging me! NST: = neon sign transformer (?).

    If not disagreeable to you, I will reveal a bit more of my ignorance of deep understanding here. Given that the outut terminals of the transformer are spaced sufficiently apart to prevent arc initiation (they always are), a well-insulated probe (lucite-handled screwdriver) when slid in contact with one terminal and the tip brought gradually closer to the other, at some point of separation, a continuous arc will occur (call it arc initiation distance). Just previous to this, a slight visual and audible effect may be seen, light-purplish glow, but not an actual ionization of the air allowing full current flow (like St. Elmo's Fire).

    Now, the fixed spark gap we consider for use with magnetically-coupled coils must obviously be set at a distance equal to, or less than, the maximum arc initiation distance. Once the full arc begins, how can it's rate be anything other than 120 arcs per second (for 60 HZ)? No arc can be present as the voltage passes through the zero points of it's waveform, and arc occurs somewhere before the sine curve reaches peak value in both directions, does it not? This assumes parasitics must be small, which is likely the case, as this type of transformer in service is essentially firing an "arc" through it's load device (neon sign?) throughout it's lifetime; some operate for years without failure. One would think that parasitic oscillations of any magnitude in the secondary coil structure would limit good life expectancy. jocular
    Reply With Quote  
     

  9. #8  
    Forum Isotope
    Join Date
    Feb 2012
    Location
    Western US
    Posts
    2,641
    Quote Originally Posted by jocular View Post
    Thank you for indulging me! NST: = neon sign transformer (?).
    Yes, on NST. And thank you for the interesting questions!

    If not disagreeable to you, I will reveal a bit more of my ignorance of deep understanding here. Given that the outut terminals of the transformer are spaced sufficiently apart to prevent arc initiation (they always are), a well-insulated probe (lucite-handled screwdriver) when slid in contact with one terminal and the tip brought gradually closer to the other, at some point of separation, a continuous arc will occur (call it arc initiation distance). Just previous to this, a slight visual and audible effect may be seen, light-purplish glow, but not an actual ionization of the air allowing full current flow (like St. Elmo's Fire).

    Now, the fixed spark gap we consider for use with magnetically-coupled coils must obviously be set at a distance equal to, or less than, the maximum arc initiation distance. Once the full arc begins, how can it's rate be anything other than 120 arcs per second (for 60 HZ)? No arc can be present as the voltage passes through the zero points of it's waveform, and arc occurs somewhere before the sine curve reaches peak value in both directions, does it not? This assumes parasitics must be small, which is likely the case, as this type of transformer in service is essentially firing an "arc" through it's load device (neon sign?) throughout it's lifetime; some operate for years without failure. One would think that parasitic oscillations of any magnitude in the secondary coil structure would limit good life expectancy. jocular
    The "arc extinction frequency" is indeed 120 per second, exactly as you say. However, there's a bit more to it than that, and this is where the fun part comes from. If we look at the actual current waveform, we indeed see that it goes to zero (approximately) 120 times a second. But the particular way it gets to zero (or away from zero) is an abrupt transition; it's not a pure sinusoid. In the absence of any resonant behavior, you'd see sharp -- nearly discontinuous -- transitions from "off" to "on." Those fast-rising/falling edges are a signature of very high frequency content.

    If there are resonances, those sharp edges will excite them (you would essentially be looking at something closely related to the step or impulse response of the resonant circuit). What you would see is high-frequency ringing stimulated by each of those edges. The number of cycles of ringing before "significant" decay is approximately equal to the Q of the mode in question.

    Complicating all of this is that true arcs can present negative differential resistance, and that NSTs are designed to have saturable cores (partly out of economy, partly to provide "free" protective current-limiting). Consequently, simple circuit simulations rarely reproduce accurately the waveforms you actually measure, but they at least capture the basics somewhat well.

    Your questions have me happily strolling down memory lane, so thanks a bunch for that!
    Reply With Quote  
     

  10. #9  
    Forum Professor
    Join Date
    Jan 2013
    Posts
    1,773
    Quote Originally Posted by tk421 View Post
    The "arc extinction frequency" is indeed 120 per second, exactly as you say. However, there's a bit more to it than that, and this is where the fun part comes from. If we look at the actual current waveform, we indeed see that it goes to zero (approximately) 120 times a second. But the particular way it gets to zero (or away from zero) is an abrupt transition; it's not a pure sinusoid. In the absence of any resonant behavior, you'd see sharp -- nearly discontinuous -- transitions from "off" to "on." Those fast-rising/falling edges are a signature of very high frequency content.

    If there are resonances, those sharp edges will excite them (you would essentially be looking at something closely related to the step or impulse response of the resonant circuit). What you would see is high-frequency ringing stimulated by each of those edges. The number of cycles of ringing before "significant" decay is approximately equal to the Q of the mode in question.

    Complicating all of this is that true arcs can present negative differential resistance, and that NSTs are designed to have saturable cores (partly out of economy, partly to provide "free" protective current-limiting). Consequently, simple circuit simulations rarely reproduce accurately the waveforms you actually measure, but they at least capture the basics somewhat well.

    Your questions have me happily strolling down memory lane, so thanks a bunch for that!
    Fancy one's lack of deep technical understanding being able to do that! Ha! Seriously, my "delving depth" of the theory involved here is limited to a course in "Pulse Circuits", DeVry Technical Institute, 1963. Mucho long ago. By that time, I had already mastered the NST screwdriver trick, and made plenty of Jacob's Ladders!

    Let's assume our NST secondary has an open circuit voltage of 15KV, call that peak, so I can visualize the sine wave building up from zero amplitude towards it's peak value. Surely, if secondary current is as near zero as it can get (only current flow is via incomplete ionization of the air surrounding the output electrodes), then the voltage is close to being a pure sine wave, is it not?

    As we gradually bring two terminals closer and closer together (the sliding screwdriver trick), at some separation distance a visible arc will signify near short-circuit current flow, but not quite. We know from experience that once the arc has struck, we can begin separating the terminals, and it's length will increase, possibly to several times the initial "strike" distance, until it finally extinguishes. This is typified by the Jacob's Ladder experiment.

    Perhaps now I am beginning to see through my ignorance here: Because the secondary has very poor voltage regulation characteristics, once current begins to flow (the arc), the voltage does not continue on upward to 15KV, but rather "levels off" until time "catches up" with it's beginning to drop back toward the zero-amplitude part of the waveform. However, due to the already ionized air surrounding the arc, it will not extinguish until the voltage waveform reaches that level where it is no longer high enough to maintain the arc. I am unable to visualize what happens to the voltage level, though it was decreasing in an orderly fashion, at the instant the arc extinguishes.

    I'm being simplistic here due to my limited technical knowledge; am I close to visualizing what's happening? jocular
    Reply With Quote  
     

Similar Threads

  1. Tesla Coil
    By Pharz in forum Electrical and Electronics
    Replies: 2
    Last Post: July 18th, 2010, 10:57 AM
  2. Tesla Coil
    By Pharz in forum Physics
    Replies: 9
    Last Post: March 23rd, 2010, 07:21 PM
  3. Need Your Help Please About My Tesla Coil !!
    By tareef in forum Military Technology
    Replies: 9
    Last Post: August 19th, 2009, 10:45 AM
  4. Tesla COIL
    By kingjacob in forum Electrical and Electronics
    Replies: 5
    Last Post: May 6th, 2007, 11:28 PM
  5. tesla COIL
    By leopold in forum Site Feedback
    Replies: 3
    Last Post: April 20th, 2007, 09:24 PM
Tags for this Thread

View Tag Cloud

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •