Message-ID: <391DC6F3.A51@xtra.co.nz>
Date: Sun, 14 May 2000 09:19:47 +1200
From: "vernall" <vernall@xtra.co.nz>
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
Subject: Re: LF: Optimal width for LF?
References: <09e401bfbc0e$5e55ab60$2301a8c0@maly.cz.gmc.net> <391C6885.A4BB64A8@bellatlantic.net> <001c01bfbcbe$1b0fed60$a747b28f@w8k3f0>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Dick Rollema wrote:
> Narrow bandwidth filters do not have to ring provided  they are of the
> phase-linear type (constant group delay in the passband).
> 
> I use a home made passive phase-linear filter with a bandwidth of 30Hz  and
> centre frequency of 1000Hz. It can pass up to 24 wpm without trouble
> (unfortunately the operator can't ...) The filter has four tuned circuits
> with coils in large ferrite potcores.
> 
> Harry, PA0LQ, has designed a similar filter of the active type using 4
> opamps. If you are interested in his design write to:
> 
> H.A.A. Grimbergen, PA0LQ
> Lijtweg 1202
> 2341 HE Oegstgeest
> The Netherlands
> 
> 73, Dick, PA0SE

Are they a "Bessel" design?  All analogue filters have some trade-off
between skirt selectivity, phase response and ripple in the passband. 
Murphy's Law applies too, in that the phase linear types (least ringing)
also have least skirt selectivity.  Hence using 4 stages in your phase
linear filter?

Could you post a brief text summary of your design?  

73, Bob ZL2CA