Good evening students and Elmers. I wanted to get part 8 posted on the
list so those of us in the USA can work on it during the long holiday
weekend. Please remember the founding fathers and their actions this
Saturday.

Ok, gather the following components:
C11 47pf
C12 150pf
C13 150pf
C14 150pf
C15 150pf
C104 .01 microfarad
R1 470 ohm
RFC1 22 microhenry
Y1 4 Mhz crystal
Y2 4 Mhz crystal
Y3 4 Mhz crystal

Install the above components into the board. Use caution while
installing the crystals Y1-Y3. If they are mounted flush to the board
the cases could short out the solder pads underneath. I mounted mine
about 1mm above the board surface using a temporary spacer while
soldering. I know that you can get insulators that fit over the bottom
of the crystals but I had none. Use whatever method you would like to
prevent the crystal cases from touching the solder pads on the top of
the board.

You can ground the cases of the crystals now by soldering a solid wire
across the top of the 3 crystals. DON'T OVERHEAT THE CRYSTAL CASES AS
DAMAGE CAN RESULT! Now attach the wire to the solder pad on the left
side of the crystals. It should be the only open solder hole on the
left side of the 3 crystals. When you are done you should see a single
wire going from the right most crystal, soldered across the top of all 3
crystals, then down to the solder pad on the left side. This grounds the
cases of the crystals preventing interferance from strong stations from
getting past the filter.

Ok, check your soldering, parts placement and parts value.

Got it all right? Good. Now apply power and check for smoke.
Place your scope probe or RF probe on the solder pad for U3 (not
installed yet). Key up the rig with your temporary jumper. You should
see a nice pretty 4Mhz sine wave. On my scope it was about 0.5 volts
peak to peak. That should be enough of a voltage to tickle those RF
probes. What happened to all those nasty mixer components that we had at
the output of U1? Well, thanks to the crystal filter they are all gone!

So now you have to ask the simple question "how does it work?"
For me, this part of the circuit is shrouded in mystery. Please try to
stay with me and lets see if we can get some discussion going on the
subject of crystal filters.

Remember, the basis of a superhet receiver is that it converts the
desired incoming signal to some "standard" frequency. In our radio that
frequency happens to be 4Mhz. This is done because it is easier to
design a filter that operates on one specific frequency than one that
operates over a range of frequencies. It also allows you to reject the
other sideband, unlike Direct conversion receivers. So the purpose of
the receive mixer is to convert the desired receive signal to the IF
frequency (4Mhz). The crystal filters job is to make sure that only the
IF frequency makes it from the receive mixer to the product detector.

In the last section we completed the receive mixer that was centered
around U1. This mixed our incoming signal from the antenna with the VFO
and presented the mixed signals at it's output. Let's pick some round
numbers.... Receive signal=7Mhz VFO=3Mhz
The output of the mixer will contain many mixer products, the major ones
being 4Mhz and 10Mhz. This circuit will filter out all but the 4Mhz
signal.

I did a bunch of reading on crystal filters and came away even more
confused then when I started. Most of the descriptions that I read
either were emperical designs, or the design was referenced as being
covered by one of the books in the bibliography. "Just build it this way
and it should work" was a common phrase in these writings.

Here is what I do know. The crystals must be closely matched, within 10
to 20 Hz of each other. Dave at Small Wonder Labs was kind enough to do
this for us. If the crystals are not matched the performance will be
poor. All of the capacitors are the same, 150pf. The value of the
capacitors has an effect on the bandwidth of the filter. It is important
to match the input and output impedance of the filter to minimize the
loss. I have no earthly idea on how to calculate this, but I do know
what parts in the circuit are performing the impedance matching. C11 and
RFC1 form an "L" filter to match the output impedance of U1 with the
input impedance of the filter. The filter is terminated into R1 (470
ohms) before reaching U3.

The crystal filter is the core of nearly every QRP rig out there. I
think it is important to understand what is happening here. Lets start a
discussion on the function and characteristics of this filter and see
what we can learn.