474.2 kHz // U3S project

Sat to Sun. night  zero 474kHz TA decodes...   the end...
Friday to Sat  just 5 decodes

Thursday  to Friday  another poor propagation night hope that propagation will pick up soon.

The Beverages could stay another week the land is too wet

No more goodies fttb.
also  LA2XPA seems to have lost propagation no more exotic West USA spots.
  propagation shifted to south below  dream DX spot
2017-03-03 05:18  WH2XXP  0.475662  -26  0  DM33  50  DJ0ABR  JN68nt  9427  32

here WD2XSH/17   19 decodes  no better then -18  no exceptions from FN square

Wednesday -  Friday night:
29 decodes of WD2XSH/17 last night   strongest -7
 no other TA
Propagation faded here  by Aurora
Aurora enhanced propagation  like on 160 and 80m  can happen
is not seen on 630m yet  

U3S  continued

The output of the U3S   at 12 10 and 6 could not be improved with an extra parallel BS170
the 50MHz  dropped to 60mW
Adjusting the bias gained some more output
The BS170 is now allmost class a  84mA bias  100mA  with  max RF
The frequency stability is very poor  with the  27MHz  x-tal  at the Si5351A
Never had good results with standard X-tals  I prefer TcXo  or dil Xtal oscilators
Like the cheap 42MHz dil oscillator in use in the 70MHz / 28MHz xvtr  very stable

I followed the Jan Ove frequency improvement modifications  it is slightly better
but still  -4 at 28  and more on 50

The U3S 28MHz signal freezes the WSJT-x  program
Need to force shut down every time it tries to decode the U3S 28MHz signal.
The spectrum  on the K3 shows a lot IMD products  on 6 and 10m
To use the U3S as an automatic beacon in a tropical place  it will have to be unconditional stable.
The firmware hardware behaved stable during the tests sofar
But sometimes the clock did not synchronize after reboot. The U3S just keeps counting secondpulses from zero up while the QLG1  has a fix  and does send the NMEA  data
Is that a firmware  flaw?
Then with running hartbeat i need to reboot the U3S and it synchronizes direct after boot.
As if the running software  under certain condition does not check the clock sync state?
One more test with the 20meter extention cable for the external QLG1workes okay.
The first test i did with this long cable showed only the 1pps hartbeat the time did not synchronize
but stayed in the count mode.
First thought that 20m is too long for the serial data to the U3S,
but the boot sync problem fooled me.
After reboot with fixed GPS the time synchronized  immediate.

12M  signal is sounding very hummy  the 5V supply is clean
so it is not so easy to get it working good.


I think a stable TCXO  is needed
The heathsink soldered to the X-tal  I used  the cupper flange of  a broken IRF540,  of which  I have so many left from my experiments with the 474.2kHz QTX .
Plus foam isolation of the Si -PCB but it did not help.