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

Re: Filter Choke Design

tesla-at-grendel.objinc-dot-com On Fri, 8 Mar 1996 09:01:10 +0700, you wrote:

>>From haba-at-snakemail.hut.fi Fri Mar  8 07:13 MST 1996
>>Received: from snake.hut.fi (root-at-snake.hut.fi [193.167.6.99]) by uucp-1.csn-dot-net (8.6.12/8.6.12) with ESMTP id DAA03274 for <tesla-at-grendel.objinc-dot-com>; Fri, 8 Mar 1996 03:34:50 -0700
>Date: Fri, 8 Mar 1996 12:34:20 +0200 (EET)
>From: "Harri \"Haba\" Suomalainen" <haba-at-snakemail.hut.fi>
>To: tesla-at-grendel.objinc-dot-com
>Subject: Re: Filter Choke Design
>Mime-Version: 1.0
>
>On Thu, 7 Mar 1996 tesla-at-grendel.objinc-dot-com wrote:
>> >Why is nobody using the setup often seen in filters ie. wind two
>> >windings to the same core. Properly phased they will cancel each
>> 	Great idea, but , the RF voltage from the TC primary will be
>> out of phase and cancel. Oh! put the windings on the common-mode-choke
>> out of phase from each other (switch on windings polarity) this will
>> make the RF and 60/50 Hz both common mode . Cancel that, it just makes
>> the core saturation problem worse.
>
>Nope. High frequency will be transformed in the" transformer" (common
>mode choke) but low frequency will be shunted through. Inductance of
>say 2mH is not much of impedance at the current levels there is at
>high voltage side. The few tens of mA going there should not be
>anything signifficant when compared to the spikes!
>
>Proof: (for single coil)
>E dt = NA dB => dB/dT = E/NA
>let voltage between the ends of the coil be low like it is for low
>freq. signals. Then magnetic flux will be low too. On the other hand,
>for high freq. E will be signifficant when the coil is ringing and
>dB/dt will lead to change in magnetic flux.
>
>Same idea can easily be applied to mutally coupled coils which are
>actually used in commercial filters! Anyone disagree or have I perhaps
>misunderstood something?
>--
Harri,
	I'm working on it.
	
	jim