[FCARC] RF sampler project at the KB8MDF station

[Adam T. Cately]

So -

      I'm sure you've all heard the discussions on the local repeater, 
about the "RF sampler" project that I've been researching and actually 
building, so that I can get a better/stronger/cleaner/etc. tapped signal 
into my oscilloscope and frequency counter on my Collins transmitter 
set-up.  I have been using the ubiquitous "isolated Tee" connection, but 
when you stub down the male pin to where it doesn't connect with the 
barrel connection that creates the isolated tap, it isn't consistent 
with frequency, nor is it consistent in signal strength, and a few other 
things that could be argued about - BUT, it does the job, if you AREN'T 
looking for a consistent measured signal tap.  I've been picking up some 
hum on my scope, and the counter doesn't register all the time, so I 
determined that I would try to find another type of signal source that I 
could use.

      Commercial RF samplers that are of any quality are also of greater 
expense, so I opted not to buy one, I assumed that I could build one 
that would work for what I want - I don't need measurement accuracy, I 
just need a tap so that I can watch the RF envelope on my scope and also 
have enough clean signal to drive my counter so that it reads 
consistently with my transmit carrier. Plus, I like to build stuff, so...

      After some soulful searching on the 'net, I found these links -

http://www.ad5x.com/images/Articles/Tap50dbRevA.pdf

which I opted not to use, as it has a direct connection to the signal 
source (your radio transmit RF power) and that isn't what I wanted

http://www.collinsradio.org/wp-content/uploads/2015/05/Build-a-Quality-RF-Power-Sampler-Jackson.pdf

which is more in line of what I was looking for, which led to this third 
link

http://www.n4ga.com/50db-hf-rf-sampler/

which is basically modeled after the second link above, and works with 
junk-box parts that you may have on hand.

      I have a LOT of junk-box parts...

      The following pics are what I put together over the last week or 
so, the first one is in line with my Collins transmitter and the second 
one will be installed at the desk inline with my solid-state rig (and 
up-and-coming amplifier) so that I can watch the wave-forms on SSB while 
transmitting on-the-air.  --- This does NOT give you the trapezoidal 
waveform you may be used to, if you run a commercially-built amateur 
'station monitor' and an amplifier, it only gives you a representation 
of the rf waveform on an oscilloscope, which is what I use ---



      (above)

      This used to be an isolator/circulator, stolen from a commercial 
transmit combiner - I found this after looking for a few hours for a 
project box and not finding any in my vast array of "stuff." - it's 
perfect for using as a sampler as it already has three connectors and 
space inside for adding the components. I adapted the N-connectors on 
the through ports to PL-259's and the tap to a BNC, but if you build 
your own you can use what you need.




      (above)

      This is a close-up of the inside layout of the components - I used 
a small length of RG-142 cable as my thru-connection, a toroid from my 
parts cabinets (no specific value of coil, 22 turns winding) one side of 
the toroid goes to ground and the output goes to a 40-ohm resistor to 
the tap, with a 10-ohm resistor to ground.  You can read up on why I 
used the RG-142 in the second link circuit descriptions of his build.  I 
kept a mostly-straight connection for the through ports as I didn't want 
to create right-angles on the feed-through signal, but you'll see later 
I changed that in my second build.



      (above)

      Here you can see the first complete rf sampler, I marked the 
thru-ports as IN and OUT, and the tapped port as RF TAP.  The tapped 
port resistor here is a 39-ohm, which I had plenty of in the parts 
drawer, which is also the value used in the parts-drawer contraption 
discussed in the last link, and the resistor to ground is the 10-ohms 
recommended value.  In the following you will see a different layout of 
the in and out ports - you can make these any way that works for you, 
other than there could be some unwanted coupling in the tapped port if 
you have the components too close to the toroid (as is stated in the 
descriptions one of the links above) - I plan to take either one of 
these back apart and see about utilizing some shielding between the 
toroid and the tapped port components, as I have seen some diagrams and 
pictures of this shielding done, using a more rectangular project box 
and setting the tapped port further away from the feed-through (see the 
below pics ...)  Your layout shouldn't be critical, as long as there is 
some separation between the toroid and the resistors - I used stacked 
pieces of heat shrink to hold the toroid in place over the coax, works 
just fine.





      (below)

      Here is my second attempt at the layout, keeping the in-out ports 
on the same side and using the third as the tapped port - I used the 
same toroid form as in the first, but wound it with the 32-turns 
recommended in the circuit descriptions.  Just MAKE SURE you label the 
ports so that you don't get confused (I caught myself AGAIN trying to 
tune into a grounded antenna switch, because I forgot to LABEL it and 
didn't remember it needed to be switched for my 10m Ringo, just today...)



      SO - after all of this nonsense, here's the scoop - on a service 
monitor, my contraptions measure about -50dbm +/- 1.5db at the tapped 
port with 0dbm through the in-out ports over a range of 1.8 - 54 Mhz - 
that's pretty flat and consistent - I shouldn't have to re-set my scope 
anymore when I change bands (which I had to do because the isolated Tee 
wasn't very consistent across band changes) and I have more solid 
readings on my counter, which is nice because the Collins rule-graph 
display is close, but I want to be closer.

      As I said before, I wasn't looking for lab-grade measurements, I 
just want to see the modulated RF waveform so that I can see where 100% 
is at on AM, and check the SSB output for over-driven audio if it 
happens - plus, it's nice to have stuff to watch while you're talking on 
the air about all the talking you want to do while you are talking about 
talking.  The 4' x 6' "ON-THE-AIR" sign is the next project . . .


      KB8MDF

-- 
- - - - - - - - - - -
     Adam T. Cately
  - - - - - - - - - - -