[Elecraft] Mod for optimizing the crystal filter for narrower bandwidths

[Wolf-Ruediger Juergens]

Hello OM's,
I found an older email from Wayne describing a mod to get constant filter 
output even at the smallest bandwith. Also Peter, DL2FI, mentioned to make 
this mod.
This morning I wound 2 new toroids and applied this mod to my K2.
The result is ufb, even at 100Hz Filter setting the amplitude of the signal is 
nearly at a constant level.
At least for me this mod was very valueable. If you dont have the KSB2 and 
like to receive SSB with the CW filter, this mod is not recommended.

I've used the values which Roy, AB7CE suggested:
T7 : 3 turns primary, 22 turns secondary.
additional T? 22 turns primary, 10 turns secondary.

72 de Wolf, DL2WRJ

here is the original posting:
---------------------------------------------------------------------------
Subject: [Elecraft] optimizing the crystal filter for narrower bandwidths
From: Wayne Burdick <n6kr@elecraft.com>
Date: Tue, 13 Jul 1999 23:19:17 -0700

>Here's how I'd modify the filter impedance matching to favor very narrow
>bandwidths. Please let me know if you try this so we can post it as an
>application note. Keep in mind that the present matching is a compromise
>intended to allow the user to vary the bandwidth from narrow (about 300 Hz)
>to wide (about 1.5 kHz) while preserving respectable loss and ripple
>characteristics.
>
>If you don't mind, please do careful before/after loss, bandwidth, and
>ripple measurements. You could use the weak birdie at 7000 kHz or 3589 kHz
>and the VFO to sweep the filter manually. This is pretty coarse, but it's
>better than nothing.
>
>Perhaps you and Bob Tellefson, N6WG could try this at the same time -- he's
>been agitating for a narrower filter for a long time  ;)
>
>Procedure (untested--use at your own risk):
>
>1. Cut the trace from D29 to W2. DO NOT REMOVE W2, as inserting the
>transformer there would interfere with the SSB adapter installation.
>
>2. Make an input matching transformer using 20:10 turns #28 on an FT37-43
>core. This translates to a 4:1 impedance ratio, or about 150:38 ohms. The
>nominal I.F. impedance is 150 ohms, and 38 ohms is a good starting point
>for optimized coupling to a narrower filter. Try various filter settings at
>this turns ratio and see what the optimum setting is (for maximum signal
>strength--i.e. lowest loss.)
>
>3. The transformer should be mounted on the bottom of the board for a clean
>installation. Insert the transformer between D29 and W2. The side with 20
>turns (primary) goes to W2. One lead of primary and one lead of the
>secondary goes to ground -- use the closest available ground pad.
>
>4. At the other end of the filter, you'll now have to match 38 ohms to the
>MC1350 by changing the winding on T7. Start by assuming the MC1350 is about
>2K input Z. The original winding on T7 is 3:11, for an impedance ratio of
>about 13, thus stepping 150 ohms up to around 2K. Your new winding will
>have a ratio of sqr(2000/150) = 7.25, which for a 3-turn primary translates
>to a secondary of about 22 turns. This is a good starting point.
>
>5. Write down your initial results from the above procedure, then try
>adding a turn or removing a turn from each transformer to see if you can
>improve the matching. In general, the greater the transformer ratios, the
>lower the optimum crystal filter impedance, and the narrower the optimal
>bandwidth. Use CAL FIL each time to determine what the new filter's
>behaviour is like.
>

-- 
Nothing so needs reforming as other people's habits.
                -- Mark Twain, "Pudd'nhead Wilson's Calendar"