Using Overtone Crystals For Filters

---------

From: Kevin Purcell (xenolith@halcyon.com)
Date: Mon Oct 10 1994 - 03:57:25 EDT


This was posted to homebrew newsgroup, but I though people here (the
inet-riggers?) might be interested.

I have a very strong bias for single conversion receivers. Having them
convert to a VHF IF gets rid of the image problems associated with lower
IFs. The problem is getting a tight xtal filter in the low VHF that you can
afford or build yourself.

One suggestion was using overtone xtals in a Cohn filter topology. Ray,
WD5IFS has tried it and it works and you can get the xtals cheaply too!

This technique is also used in the Index Labs QRP rig (another reason its a
nice rig!).

These are Ray's results:

>Date: 6 Oct 94 22:49:01 GMT
>From: mack@mails.imed.COM
>Subject: Using Overtone Crystals For Filters
>
>
> >Date: 20 Sep 1994 06:34:18 GMT
> >From:
> >ihnp4.ucsd.edu!munnari.oz.au!newsroom.utas.edu.au!news@network.ucsd.e
> du
> >Subject: Overtone Crystal Filter
> >Ian wrote:
>
> >Electronics Australia published an SSB receiver with an IF of 8MHZ
> >using 3 8MHZ computer xtals in a ladder network configuration. All
> >the capacitors were 100pF. The article said using a ladder network
> >enabled the xtals to be all resonant at the same frequency.
>
> >Can the same idea be used with overtone xtals? For example, I would
> >like to build a 48MHz IF filter.
>
> >Thanks for any info, Ian
>
> Well, I took this as a personal challenge to find out what's what. I
> hooked up 3 48MHz microprocessor crystals to the network analyser.
> These crystals were CTS model MP-480. I measured them at 16 MHz and
> at 48 MHz for their equivalent circuits. At 16 MHz they showed very
> good response with about 3 ohms equivalent series resistance and Q
> over 100,000. At 48 MHz 2 showed good Q of about 160,000 and 15 Ohms
> equivalent resistance. The third had much better series R of only 9
> Ohms, but the Q was considerably less and it was also not a good match
> for frequency.
>
> I used the design equations from the May 1982 QST article (Hayward)
> for 2, 3, and 5 element filters. They all showed reasonable
> calculated values as long as the bandwidth was between 1000 Hz and
> 3000 Hz. Outside these limits either the capacitors got really small
> or the load impedance for the filter got really small. In the 2500 to
> 3000 Hz range the load impedance is high enough that there is minimal
> insetion loss.
>
> The 2 element filter I built with the 2 good crystals from my parts
> bin showed good agreement with the formulas when measured on the
> network analyser.
>
> Digi-Key sells microprocessor crystals for about US $1.35 in this
> frequecy range or 10 for US $11.00. This is cheap enough to buy a
> bunch and select for matching.
>
> I hope this gives you encouragement to try this, Ian. Here is a short
> bibliography of articles which should help if you try to build a
> filter. I have arranged them in order of the amount of help they were
> to me:
>
> QST May 1982 pp 21-27
> Communications Quarterly Winter 1993 pp 11-18
> QST Nov 1980 pp 20-23
> QST Jul 1987 pp 24-29
>
> I seem to recall one in an old issue of RF Design but I haven't had
> time to look for it.
>
> Ray
> WD5IFS
> mack@mails.imed.com

Kevin Purcell N7WIM / G8UDP xenolith@halcyon.com 206/649-6489
Seattle dBug Mac Developers SIG organiser kevinpu@atm.com


Search QRP-L Archives

[ QRP-L Archive | ]
[ 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 ]

---------

This archive was generated by hypermail 2b29 on Fri Jun 02 2000 - 11:27:02 EDT

kd4ab@kd4ab.org