Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

Tesla Coil Build Thread






You will have yo add pictures when you finish it. Would this fix the issue I have of my laptop getting feedback through ground and disconnecting my midi when I play songs? I have to have the laptop floating and my battery has like 10 minutes of charge left.


Sigurthr said:
It is pretty simple, really. Capacitors pass AC while blocking DC. The size of the capacitor determines how much impedance it presents to a certain frequency of AC. The trick is to pick a value that presents little impedance to the RF from your TC, but a very high impedance to the mains AC power at 50 or 60Hz. The formula for capacitive reactance (impedance) is 1/(2piFC), where F = frequency in Hz, and C = capacitance in Farads.

In choosing a capacitor for an application there are a lot of things to consider, but foremost you have the voltage ratings, which must never be exceeded, and the polarization (if any). Polarized caps are to be used with DC, or where the DC offset exceeds the amplitude of the AC superimposed on the DC.

For this application, the capacitor will be directly connected to mains, so the DC voltage rating needs to exceed the peak voltage of the mains cycle, and ideally the capacitor's AC voltage rating should exceed the AC Vrms of the mains. To determine the peak voltage of a sine wave multiply the Vrms by 1.414. Now, they make capacitors designed specially for mains connections that are engineered to have failure modes that do not result in fires or explosions, these are Class-Y and Class-X rated. Class Y is for mains to earth or mains to mains, and Class X is for mains to earth only. We'll be using a Class-Y2 rated capacitor with 1.25kV DC and 300Vac ratings since I'm on 120Vrms service.

Here is the capacitor I selected/ordered:
Invalid Request

These caps are only 0.1uF, but Class-Y film caps are hard to find in large values, and quickly get expensive. At this capacitance they will present a 26Kohm impedance to the mains AC power. That means only 4.5mA of 60Hz power will seep through. At 200KHz, the f0 of my large TC, this 0.1uF is only a measly 8ohms, and while you wouldn't want 8ohms between the secondary return and true earth as this would seriously limit the flow of RF current to earth (due to the low voltage and high current output of the secondary return), it works marvelously for the high voltage and low current that the mains wiring in the walls pick up due to capacitive coupling from the toroid. At 750KHz, the frequency of most small form SSTCs with no topload, this capacitor presents a paltry 2ohm impedance. If I were to be running a SGTC or DRSSTC, which operate at much lower frequencies typically, I would want a larger capacitance, on the order of 0.68uF at minimum, perhaps up to 3.3uF or so for a 50KHz coil. *See note below about pulsed coils.

Implementation of the capacitor RF shunts is simple: Simply connect one capacitor between the HOT and Ground lines and another capacitor between the NEUTRAL and Ground lines. A third capacitor (X/Y or X rated) can be added between the HOT and NEUTRAL lines but it is not necessary from my experience. I'll be using a completely plastic project box to enclose the capacitors and connections. I'll use a heavy gauge three prong mains plug and a few inches of 12/3 or 14/3 extension cable for easy connecting and disconnecting from the outlet. Ideally you want one of these at the outlet the tesla coil is plugged in at, one at the outlet where any additional ground connection is made (sink, dedicated line, etc), and one at the outlet where any potentially sensitive equipment is located. You can add as many as you like with no detriment (other than to your wallet).

*Pulsed Coil operation (Staccato VTTC, DRSSTC, ISSTC, OLTC, SGTC) causes RF noise at the pulse frequency, its harmonics, the resonant frequency of the secondary (f0), and wideband noise if there is a spark gap. These lower frequency components will see a larger impedance than the f0 and will not be as effectively filtered out by the capacitor shunts. A typical spectrum for a SGTC will see peaks at the mains frequency and harmonics, the break rate, sever orders of harmonics at both even and odd intervals of the break rate, a continuum of white noise from DC to several GHz, the quarter-wave and half-wave frequencies of the wiring between the SG and primary tank, and the f0 of the coil. This LF section, numerically, would look like: 60Hz, 100Hz, 120Hz, 200Hz, 240Hz, 300Hz, 360Hz, 1KHz, 1.2KHz, etc. All of these noise components will find there way onto the mains but will not see a low impedance path to ground through our shunt capacitors.
Click to expand...
 
Fiddy said:
yeah? blown some up before?
Click to expand...

More than once... Luckily the shrapnel missed me each time. Mostly happened in trying class-e work where parasitic oscillations ramped current up way beyond the true Arms for the cap. Never had one overvolt but never used nonames in true HV work. Though I consider HV to be >1kV.

Speedy78 said:
You will have yo add pictures when you finish it. Would this fix the issue I have of my laptop getting feedback through ground and disconnecting my midi when I play songs? I have to have the laptop floating and my battery has like 10 minutes of charge left.
Click to expand...

It might, it might not. As I said with pulsed coils there is a lot of LF and harmonic content that you can't really filter. Floating the laptop or using a mains isolation transformer are your best options. Or... Faraday shielding the laptop.
 
NST showed up. MMC should be here tomorrow. Of the two I'm glad this cam first because I get to play!


Actually took a while to get the arc to travel up and not just sit at the bottom.
 
Should work now. I uploaded it from my phone so the setting are different I guess.
 
Looking good! The output from that guy is nonlethal (as long as NO CAPACITORS ARE HOOKED UP) as well. However, it hurts like you would not believe.
 
Regardless of weather its a MOT (definitely not), NST, secondary output of a TC I don't plan on putting any part of my body in it lol. I get shocked enough at work by stupid high voltage static charges and I don't need any more shocks.
 
Nice video :) ^^

* Happy Happy Happy *

Just converted my half bridge mini SSTC to a full bridge SSTC :D . Mosfets are 500V , 14 A rated but have low power dissipation as they are plastic 220 case . The DC blocking capacitor is 12uF metalized PP 900V . The smaller blue capacitors are the sunbbers for each half of the full bridge .

Output is nearly twice the size from the same secondary xD .

Here is the full bridge / DC blocking caps and snubbers and on the AL sheets are the recovery diodes . The bridge is built with 2 " half bridges " side by side then copper bar linking the +/- rails . Everything is soldered to copper tape stuck to a plastic sheet .


IMG_0162[1] by TwirlyWhirly555, on Flickr


IMG_0163[1] by TwirlyWhirly555, on Flickr


IMG_0164[1] by TwirlyWhirly555, on Flickr
 
Last edited:
Sigurthr said:
Excellent work! Is that a home made CT I see? Details please.
Click to expand...

Thanks :D ,

It is :P , its just a ferrite core ( same as a GDT core ) with 50 turns of enamelled copper wire on the secondary and one turn on the primary ( wire from the TC secondary base to earth )
 
Last edited:
You wouldn't want to stick a wet finger in there would you!


Nice work.
 
ionlaser555 said:
Thanks :D ,

It is :P , its just a ferrite core ( same as a GDT core ) with 50 turns of enamelled copper wire on the secondary and one turn on the primary ( wire from the TC secondary base to earth )
Click to expand...

What's the fres of your coil? I've been wanting to switch to CT feedback, but commercially available CTs are only available up to 200KHz or so.



Edit: My mouser caps came in the mail today so I made the first AC Line filter. It isn't really a filter, just a high-pass shunt. Forces RF to be carried on all three conductors evenly so that common mode chokes can filter RF out and so that devices don't see any RF on mains references to ground.

On the left is a 12ga x 3 mains cord, on the right is a 14ga x3 dedicated RF ground rod network.
r5rf.jpg


e6mp.jpg
 
Last edited:
You must log in or register to reply here.





Back
Top