Re: troubleshooting tesla coil, continued (fwd)

["Tesla list" <tesla@xxxxxxxxxx>]

---------- Forwarded message ----------
Date: Sun, 16 Sep 2007 22:06:17 -0700
From: Barton B. Anderson <bartb@xxxxxxxxxxxxxxxx>
To: Tesla list <tesla@xxxxxxxxxx>
Subject: Re: troubleshooting tesla coil, continued (fwd)

Hi Marko,

Yes, now that you identified 4 pairs of copper sheets and the 1" 
clearance using Mylar, 7nF is just about right! And yes, Mylar has a 
dielectric constant value of 3.2 and breakdown at 7,500 V/mil. But there 
is a problem here that you have identified in this particular post. The 
problem is Mylar (aren't you sick of people bringing up bad news?). 
Sorry, it's just that it sticks out like a sore thumb. You accounted for 
the volts/mil and dielectric constant, but you didn't account for how 
dielectrics behave with alternating currents such as RF (Tesla Coil 
frequencies).

Dielectrics have what is termed a "dissipation factor" identifying 
losses (always in the form of heat) when the dielectric is operated in 
an alternating current. The higher the frequency, the more losses that 
are generated. Some dielectrics have a high dissipation factor and 
others a low dissipation factor. What is desired for high frequency is a 
low dissipation factor. Mylar unfortunately has a high dissipation 
factor and makes them unsuitable for Tesla Coil tank caps. It's always 
best to go with a known dielectric which has a low dissipation factor 
such as polypropylene, polystyrene, polyethylene, etc.. just as the cap 
manufacturers themselves go with. Even with our MMC's, we account for 
the dissipation factor (tangent loss) as specified by the cap manufacturers.

There are a lot of things you'll be working on, but I just wanted to 
point out that Mylar is not a great choice for tank caps (even if the 
volts/mil is high and dielectric is in the medium range). Your cap being 
a plate cap would likely do better than a rolled mylar cap simply due to 
the thermal dissipation of the plates, so you may want to give it a go 
anyway and see how it reacts. Of course, if breakdown occurs (even a pin 
hole punch through), that will end the caps use. I personally used oil 
in my plate cap, mainly for the self healing ability.

If you are ever curious as to what constitutes a good dielectric or not, 
copper vs. steel in the primary, etc.., just do a search at 
www.pupman.com of the list archives. Cap dielectrics are one of those 
items where there is a lot of information available.

Take care,
Bart


Tesla list wrote:

>---------- Forwarded message ----------
>Date: Sun, 16 Sep 2007 16:43:56 -0400
>From: Marko Ruban <Marko@xxxxxxxx>
>To: Tesla list <tesla@xxxxxxxxxx>
>Subject: Re: troubleshooting tesla coil, continued (fwd)
>
>As I mentioned a bit earlier, I made an error in units, my cap was an
>estimated 7nF (not pF).  I finally got the multimeter, and that value was
>confirmed since initial post.
>
>Capacitor is constructed with 4 pairs of 10mil copper sheets (6" x 8")
>separated by 10mil Mylar sheets with 1" margin on the sides.  I picked
>Mylar because it has a high dielectric value and high breakdown voltage.
>10mil sheet is rated at 70kV.
>
>
>Tesla list wrote:
>
>---------- Forwarded message ----------
>Date: Fri, 14 Sep 2007 18:17:24 -0700
>From: Barton B. Anderson <bartb@xxxxxxxxxxxxxxxx>
>To: Tesla list <tesla@xxxxxxxxxx>
>Subject: Re: troubleshooting tesla coil, continued (fwd)
>
>Hi Marko,
>
>Your inserting 0.007uF into Javatc so the frequency's do match as they 
>should at that capacitance. However, your "saying" 7pF and that is way 
>off from both Javatc and what the plate cap really is. Judging by your 
>plate cap configuration, it is about 0.7nF (not 7nF) if your using glass 
>as an insulator. If air is your insulator, then it's down to 0.1nF.
>
>So you are way out of tune. The fix is the cap. For a plate cap using 
>glass with about 0.2" spacing, you would need a plate surface area of 
>900 square inches (30" x 30").
>
>
>If your going to use a plate cap, you might want to consider Castor oil 
>which has a high dielectric of 5. About 15 plates (8" x 11") with a 0.2" 
>spacing between plates would yield near 7nF (which is 0.007uF). Plate 
>caps are fun to play with, but when you start building to a capacitance 
>such as this, it can get bulky and messy. I of course recommend an MMC 
>for performance.
>
>Take care,
>Bart
>
>Tesla list wrote:
>
>  
>
>---------- Forwarded message ----------
>Date: Thu, 13 Sep 2007 19:19:45 -0400
>From: Marko Ruban <Marko@xxxxxxxx>
>To: Tesla list <tesla@xxxxxxxxxx>
>Subject: troubleshooting tesla coil, continued
>
>Hello guys, I'm back from a long summer vacation.  As per previous 
>suggestions I have acquired an old NST transformer (12kV, 60Hz, 60mA 
>output).  And the spark gap does run now with capacitor/primary in 
>circuit.  However, there's no visible effect on the topload, still.  
>Could it be that badly un-tuned?
>
>Please see the photo of my simple setup and make any suggestions...
>http://marko.dppl.com/TC_setup.jpg
>
>The flat rectangular thing in front, is the capacitor, which should be 
>about 7pF.  Can't make a more specific measurement since my new 
>multimeter is still in the mail.  Spark gap is on top of the NST, and 
>consists of two screws facing each other (about 0.2" adjustable 
>spacing).  Wire running in the back to the right is the ground.
>
>Also, here's a dump of JavaTC for my coil...
>http://marko.dppl.com/coil.txt
>
>Thanks in advance for any assistance.
>
>
>
>
>
>
> 
>
>    
>
>
>
>  
>
>
>
>
>
>
>
>
>  
>