[DSP-10] still no PLL locks

[Courtney Duncan]

Working with the provided "PLL Diagnostics" page has been very helpful.

I discovered that U108-9 was not connected to U107-14.  Repaired this.

Added 470 pf across U107-14 and U107-8 as suggested "just in case."

Now, after loading udiag2.exe, the voltages on U107-15 and -1 through -7 
and U108-15 and -1 through -7 are correct.  All are 0V except for 
U107-3, U107-7, and U108-2 which are 5V.  This is as expected.

Removed the slug from L102 and bent L103 and Q101 around a bit.  At this 
point, when I inject ground at the junction of C113 and R105, I get 
123.62 MHz.  When I inject 5V at the junction of C113 and R105, I get 
129.07 MHz.  This should be precisely the desired tunable range.

P106-2 is connected to an external 10 MHz reference and -3 to shield on 
the coax from that reference.  This is the TCXO used with the AMSAT 
kits, a Frequency Products model T 424.  Dan Schultz says I might need 
to use an attenuator on it or I might get "flakey results".  As you can 
see below, I am getting no results.

After loading udiag2.exe, going around U104 with an oscilliscope, I see 
the following:

-1  10.00001 MHz swinging from -0.5 to +3.5 V.  Very clean looking.
-2  4.97 VDC with 20 mV of noise.
-3  4.97 VDC
-4  4.96 VDC
-5  ground
-6  ground
-7  +/- 10 mV of noise centered at +10 mV.  Never 5V at any time.
-8  123.62 MHz swinging 1.3 to 2.0 V.  Clean
-13 ground
-14 a 5 V, 500 nsec wide pulse every 204-205 microseconds
     exactly as shown in the figure

There are no DC shorts between -2 and -3 or between -5 and -6.

-2, -5, -7, and -13 are not showing the expected signals.  Rechecked all 
  connections to these pins.  Since I am using an external reference, 
R103, C108, C109, X101, and C110 are absent.  All others appear to be OK.

Studying the National data sheet for this part, it looks like the LO 
phase counter is working fine and the reference phase counter is not 
working at all.  Also, the oscillator output is not working at all.

Is this a blown LMX1501A U104 part?

Is there something about the absence of one or more of the crystal 
oscillator circuit parts that is causing U104 to miss the 10 MHz input? 
  Like, does pin 2 need a pull up or pull down or something?  The 
datasheet just says "output".

If I build an attenuator, like 100 ohm + 10 ohm in series across the 10 
MHz part output, then tap off the 10 ohm for -20 dB, is there any chance 
that that will fix this?  The datasheet says you can drive pin 1 from 
TTL so I'm surprised that a 4 V swing across the Vcc/2 threshold doesn't 
work.