Post subject: How to test a directional couplers?
Posted: Tue Jun 16, 2009 12:31 pm
Tue Jun 16, 2009 12:25 pm
I have got few directional
couplers pretty old one manufactured by company named anzac electronics
its not in existence was closed in 1960 itself...I dont have the data
So I need to find the specifications for those directional
couplers and need to test them whether they are working properly or
not...Can any body please suggest me some experiments to test a directional
coupler using Vector network analyzer..
Please do reply
Thanks a lot...
Post subject: Re: How to test a directional couplers?Posted:
Wed Jun 17, 2009 11:38 am
Joined: Thu Sep
25, 2003 1:19 am
Basically, performing return loss measurments on all ports and insertion
loss measurements between all ports will tell you if the couplers are
performing well and what the frequency range and coupling value is.
You could also determine directivity which is Coupling - Isolation.
Is one of the ports terminated?
I'm sure RFCafe has appnotes,
if not, checkout Macom and Minicrcuits for some tutorial.
Post subject: Re: How to test a
directional couplers?Posted: Tue Jun 23, 2009 11:55 am
Joined: Tue Jun 16, 2009 12:25 pm
port is terminated internally...
how do i perform the test? I
have a tow port VNA...
should i just give input and check whether
i am get output for the desired frequency range? I did check rf cafe
and mini circuits they dont provide how to test DC for unknown configurations...
Post subject: Re: How to
test a directional couplers? Posted: Tue Jun 23, 2009 12:43 pm
Joined: Sun Aug 03, 2003 2:02 pm
Location: Erie, PA
a page I have on the basics of directional couplers, but it does not
have explicit instructions for how to test a directional coupler.
So, here is a brief description of how to determine the
fundamental parameters of a directional coupler. Since your couplers
are configured with the reverse port (Port 4 in the drawing) internally
terminated, that makes the process a bit simpler. Physically, the layout
of the directional coupler is similar to the drawing, with the mainline
path along the long dimension of the body, and the coupled port(s) perpendicular
For clarity when referring to port numbers, network
analyzer ports will be referred to as Port1NA and Port2NA, and directional
coupler ports will be referred to as Port1DC, Port2DC, Port3DC.
1) Calibrate your network analyzer (NA) across the frequency range
that you will be testing. If you do not know the frequency range, then
cal across the entire bandwidth of the NA.
2) The first test
will be the mainline loss. Connect Port1NA to Port1DC (input), and Port2NA
to Port2DC (output). Place a calibrated 50-ohm termination on Port3DC
(coupled). The insertion loss within the operational bandwidth of the
coupler will typically be less than 1 dB, so if you did not know the
BW to begin with, the 1dB points will be a good ballpark number. You
can note the S-parameters as needed in one or both directions (S11,
S21, S12, S22).
Note: If you cannot obtain an insertion loss value
of less than 1 dB anywhere in the band, then either your network analyzer
does not cover the DC's bandwidth, the assumed DC port configuration
is not correct (try other port combinations), or the coupler is broken.
3) The next test will be the coupling factor. Connect Port1NA to
Port1DC (input), and Port2NA to Port3DC (coupled). Place a calibrated
50-ohm termination on Port2DC (output). The measured S21 value in the
bandwidth determined in the first measurement is the coupling value
wrt to the input. Again, you can note the S-parameters as needed in
one or both directions (S11, S31, S13, S33). If you measure -20 dB in
the middle of the band, then you have a 20 dB directional coupler.
4) If you need to know the isolation between the directional coupler
output port and the coupled port, then connect Port1NA to Port2DC (output),
and Port2NA to Port3DC (coupled). Place a calibrated 50-ohm termination
on Port1DC (input). The measured value is the coupled port isolation.
Again, you can note the S-parameters as needed in one or both directions
(S22, S32, S23, S33).
Directivity is the numerical difference
between the value measured here and the value measured in the previous
step. Directivity should be greater than 20 dB, typically 30 dB or better.
- Kirt Blattenberger
RF Cafe Progenitor &