Filtering on Neons
Subject: Filtering on Neons
From: richard.quick-at-slug-dot-org (Richard Quick)
Date: Sun, 28 Jan 1996 05:22:00 GMT
>Received: from uustar.starnet-dot-net (root-at-uustar.starnet-dot-net [220.127.116.11]) by uucp-1.csn-dot-net (8.6.12/8.6.12) with SMTP id AAA05964 for <tesla-at-grendel.objinc-dot-com>; Sun, 28 Jan 1996 00:45:40 -0700
* Carbons Sent to: usa-tesla-at-usa-dot-net
Quoting "Baddeley, Glenn" <GBaddele-at-vitgssw.telecom-dot-com.au>
> There has been recent discussion about using ferrite toroids as
> a core for filtering chokes on the secondary of power trans-
> formers. I have a number of ferrite rods (approx 12mm diam,
> 200mm long). Can they be pressed into service?
> Should I bundle more than one for a core? How many turns
> should be necessary? (12kv 60ma neon) (ie. What is the
> required inductance?)
OK, people left and right have been trying to pin me down on this
one for awhile. I would go out and measure mine, but I don't have
an inductance meter (next piece of diagnostic equipment to be
sure... Still would rather spend the money on copper and caps...)
The answers to this question may be found in Volume 15, #1 of the
TCBA publication NEWS. The article by Herman W. Flynn gives hard
data on the effectiveness of various chokes in Tesla coil
Basically he is showing that 30 and 44 turns of wire on a ferrite
core (Radio Shack #273-104) has an inductive reactance of 15,500
ohms and 24,200 ohms (respectively) at 700 kHz. The measured
inductance of these chokes are 3.52 mh and 5.50 mh.
With 80 ma of RF -at- 700 kHz flowing to his ground path Mr. Flynn
found that the typical air-wound choke coil as used by many
coilers over the last 80 years had no effect. This concurs with
my experimental results.
When he placed the 30 turn ferrite core choke coil in the ground
path the current flow dropped to 20 ma of RF.
The 44 turn ferrite core choke coil dropped the RF current to 8
ma of RF. Placing both choke coils in series dropped the RF
current flow to 5 ma.
Now, I have not yet gone out and looked at the Radio Shack part#
273-104. I don't think I really have to. The choke cores that I
am always talking about (2 - 4 inches OD) are going to be
considerably more massive than the stuff Radio Shack is stocking.
This is probably why I am hearing that people are having
difficulties in finding the bigger cores. Typically we are
operating at much lower frequencies at higher power. I would
respectivly submit that the larger ferrite toroid cores with
around 20 or so turns of wire on them are effective for our lower
frequencies and higher power levels.
I thought Chip's idea of using the 400 Hz variac cores for RF
choke cores was great.
> Also, what value of bypass capacitor is suggested? A 'few dozen
> picofarad' is a bit vague. Is there a rule that is more than
> just based on empirical experience?
Sorry, I can't give you a formula. I am not a math major, but I
can tell you what works... The upper end of values in the bypass
run about .001 microfarads, and the value used should be somewhat
lower. At the other end of the workable range of values is
literally "a few dozen picofarads". There are a lot of variables:
value and type of the chokes used, the frequency of the tank
circuit, and the dissipation factor of the bypass dielectric.
Enough variables that it is difficult to be precise.
I think the bottom line here is that you don't have to be
precise. It just has to work reasonably well, which it will.
Especially with a choke that is on the heavy side and high loss
dielectric in the bypass.
... If all else fails... Throw another megavolt across it!
___ Blue Wave/QWK v2.12