[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RF filter circuit (fwd)



---------- Forwarded message ----------
Date: Sun, 28 Oct 2007 15:49:36 -0500
From: David Dean <deano@xxxxxxxxxxxxxxxx>
To: Tesla list <tesla@xxxxxxxxxx>
Subject: Re: RF filter circuit (fwd)

Hi Kris

On Sunday 28 October 2007 10:40:16 am you wrote:
> ---------- Forwarded message ----------
> Date: Sun, 28 Oct 2007 07:09:05 -0700 (PDT)
> From: Kris Grillo <kristianisawesome@xxxxxxxxx>
> To: Tesla list <tesla@xxxxxxxxxx>
> Subject: Re: RF filter circuit (fwd)
>
> I had built a terry filter for my NST power supply, and later used it on
> my 2 MOT w/ half wave voltage doubler power supply.
> http://img98.imageshack.us/img98/8721/nstfilter2sk0.jpg Terry has
> mentioned that the filter is rated for .24 amps max. I noticed that the
> resistors get HOT quick using the two mots. There temp would exceed the
> capability of my IR temp gun (450F) in a about 60 seconds of run time, but
> without it I was frying diode after diode. I don't even dare try it on the
> 4 stack.  I am looking for the math involved in sizing the resistors to
> the power supply.

I am sure Terry used fancy math and computer simulations to come up with the 
values for the Terry filter, but the resistors should scale using Ohms law.
The Terry filter is rated .24 Amps, the NST .06. Four times the ampacity.
The MOT is .5 Amp so the filter should be rated 2 Amp to keep the same level 
of overengineering.
 
The 2 MOT level shifter can deliver about twice the power at half the voltage 
of a 15X60 NST so you are going to want to dissipate about twice as much 
power in the resistor, with four times the current going through it.
The resistor in the Terry filter is 1000 Ohms. The current is .24 Amp. 
Dissipation is I^2*R so 57.6 Watts. You want to dissipate twice that or 115 
Watts. You want the current to be for times so I=4*.24=.96 Amp.
So a 100 Ohm 200 Watt should do the trick.

> I have ordered two sets of power resistors, 100ohm 200W 
> and 50ohm 250W. 

Funny how that worked out. More comments below-

> These are going to be my trial and error sets. If they 
> don't work, I'll just keep trying different values until things stop
> burning up, I guess.
>
>
> Tesla list <tesla@xxxxxxxxxx> wrote:
> ---------- Forwarded message ----------
> Date: Sat, 27 Oct 2007 13:46:46 -0400
> From: Slurp812
> To: Tesla list
> Subject: Re: RF filter circuit (fwd)
>
> Heres what most around here talk about. Not sure who's page this is,
> but this it the "terry filter" Iv heard about. I am in the process of
> making one myself, minus the MOV's for now anyways...
>
> http://www.ozonejunkie.com/Images/nst%20filter.gif
>
> On 10/26/07, Tesla list  wrote:
> > ---------- Forwarded message ----------
> > Date: Fri, 26 Oct 2007 13:34:13 -0700 (PDT)
> > From: Kris Grillo
> > To: tesla@xxxxxxxxxx
> > Subject: RF filter circuit
> >
> > I have been trying to figure out how to properly size the components for
> > an rf filter to use with my 4 mot tesla coil power supply. Searching for
> > RC circuit, RF filter circuit, passive RF filter circuit, low pass filter
> > circuit, etc, has yielded many descriptions and schematics covering a
> > very broad range of applications. I'm not sure what I should be
> > specifically looking for.

The Terry filter is a simple RC low pass filter which acts as a voltage 
divider.  At some frequency Xc = R.  You want to keep that the same. If you 
have lowered the resistance by 10 then you need to lower the capacitive 
reactance by 10 to keep the cutoff frequency the same.

To lower the capacitive reactance by 10 at a constant frequency you need to 
increase the capacitance 10 times. In the Terry filter the caps are strings 
of six 3.3nf caps in series for 550pf. Means you need 5.6nf for the filter to 
work the same. 

As to the MOVs, I'd leave them out.  When they clamp, they switch off at the 
zero crossing. Does a MOT doubler circuit output ever cross zero?
Anyway if the MOV did clamp, that would take out your diodes.

Speaking of diodes-
After the Terry filter the current is limited to what can flow through the 
resistors in the filter. Your case probably a peak voltage of 12k Volts 
through 2 X 1k Ohms or  6 Amps. Changing that to 2 X 100 Ohm resistors would 
make that 60 Amps. A single string of 1N4007s would be good for the 6 Amps, 
as the maximum repetitive peak current is rated as 30 Amps, but it would take 
ten strings to be as robust and two strings just to meet the 60 Amp case.

> > Also, I do have some formal education in 
> > mathematics (college, 8 years ago, 3 semesters of algebra, 1 of pre
> > calc,) but I have no formal education in electronics and most of the math
> > descriptions contain symbols that I am unfamiliar with what they
> > represent. I was wondering if anyone would happen to know of any good web
> > pages I could check out that would help me in my quest to learn and
> > understand this circuit? Maybe something with mathematical examples
> > worked out, if anyone knows of any such sites.

Wikipedia, of course, and I find this interesting 

http://www.ibiblio.org/kuphaldt/electricCircuits/index.htm

later
deano

> >
> >  __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
>
>  __________________________________________________
> Do You Yahoo!?
> Tired of spam?  Yahoo! Mail has the best spam protection around
> http://mail.yahoo.com