From: "Gary - G4WGT" <g4wgt@tiscali.co.uk>
To: <rsgb_lf_group@blacksheep.org>
References: <003001c9d8c4$719e7710$6401a8c0@asus> <A86DF6F539EE4CD98CF0727513118FC8@JimPC>
Date: Wed, 20 May 2009 12:16:52 +0100
Message-ID: <001501c9d93c$7a2bb640$6401a8c0@asus>
MIME-Version: 1.0
In-Reply-To: <A86DF6F539EE4CD98CF0727513118FC8@JimPC>
Thread-Index: AcnY5vt7nVRsEzq2S+qRwWqzcBOrsgAUyckg
Subject: LF: RE: Re: 518khz - Notch Filter.
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


LF,

Thanks for all the replies, comments & suggestions.

I thought I could do the filtering within the SpecLab software but all my
attempts have had no effect at all, I am not really aware how to manipulate
the filter controls to do this, if anyone can help with that it could be the
answer. I will post a query to the SpecLab group.

Jim, that is correct regarding the LO=512KHz & I do have a couple of
alternatives but unsure until I check if they will resolve the problem.

Rik, I guessed that the Q of the circuit needs to be improved but unsure how
to go about it in that sort of arrangement, I will look at your ideas.

Andy, I will look at the link you provided soon.

Alan, thanks for the link & your comments.

Graham, the 518KHz signal is so strong at times I think attenuator would be
so large it would degrade the rest of the band, thanks for your comment.

73

Gary - G4WGT.





-----Original Message-----
From: owner-rsgb_lf_group@blacksheep.org
[mailto:owner-rsgb_lf_group@blacksheep.org] On Behalf Of James Moritz
Sent: 20 May 2009 02:01
To: rsgb_lf_group@blacksheep.org
Subject: LF: Re: 518khz - Notch Filter.

Dear Gary, LF Group,

An image of 518kHz at 506kHz implies LO = 512kHz? You could change the LO 
frequency to 500kHz, although then you would not be able to receive signals 
on that spot frequency. You could also try optimising the amplitude/phase 
"tweaks" in the software to give maximum rejection of the unwanted sideband 
at 518kHz.

A simple notch filter with high rejection will inevitably have significant 
loss at nearby frequencies - there is a trade-off between Q, depth of notch,

loss of wanted signal and frequency difference between wanted signal and 
notch frequencies. This can be improved by adding aditional components, e.g.

adding a series capacitor to a parallel-tuned trap can "tune out" the loss 
at a frequency below the notch frequency. It is feasible to get something 
like a few dB loss at 504kHz with 20 - 25dB at 518kHz. More complicated 
designs like Andy's have a much nicer response shape, but may require 
unachieveably high inductor Q to get the design response.

Cheers, Jim Moritz
73 de M0BMU