All,
I guess it's time for me to put up my "common mode choke" test technique
for all to shoot at. I have a HP 4815A Vector Z meter, a N2PK VNA, and a HP
8753C VNA. I have all the probe adapters for the Z meter, and I have been
bitten by the small amount of stray C that you just can't eliminate (but
you should be able to calculate out). I have finally decided that my
standard test is based on the following:
I made up a board out of .062 FR-4 that is approximately 6" across (and
about 4" wide). I cut three 50 Ohm microstrip lines full width across the
6" dimension. I soldered an N connector on each end (6 connectors total).
So now I have 3 equal-length 50 Ohm runs spaced about an inch apart. One
run I left alone, and that's my THRU calibration path. The second run got a
0.1 inch (or so) gap cut in the middle; I soldered various chip resistors
across this gap (in series with the run), and I use this for reference
purposes. The third run is almost completely removed except for about 1/2"
pad right at each N connector. I soldered a 2" or so piece of wire with an
alligator clip to each pad, so that the tips of the alligator clips can be
just clipped together in the middle of the board ( making a thru circuit).
Now the way I use this is: I calibrate S21 with a 6 or 10 dB pad on each
end of the THRU run. Then I move the pads to the run with the resistor gap,
and I collect a family of S21 curves with different resistors soldered in
the gap - I chose 1k, 2.2k, 3.3k and 4.7k. I only have to do this once.
Then, I move the attenuators to the third run with the alligator clips. I
clip these on the CM choke I'm evaluating, measure S21, and compare the
resulting plot to the previously-taken family of resistor plots and try to
make a determination of where the measured choke falls, "resistance" wise.
I have been busy over the last few months making chokes for 3 pairs of
yagis - I'm putting up 4/4 on 20, 5/5 on 15 and 5/5 on 10 this summer - so
I have been taking a *LOT* of data. I am convinced that I can tell if a
choke is better on 15 than on 20, for instance, and I'm convinced that I
can tell if this number of turns on this many cores is better than the last
one I tried. I am not so convinced that I can put an absolute number on the
choking resistance of the units that I'm evaluating. I am pretty sure the
units I'm winding up with are pretty darned good, compared to some relative
measurements I have made on some commercial stuff.
I know I get the same results with the N2PK and the 8753C. 6 dB pads are
good enough - the data is identical with 10 dB pads. But, I also
consistently get lower measurements with the Z-meter; for instance, if I
conclude that a particular choke has approximately 3k choking impedance
based on VNA measurements, I would probably get 1.5k or so with the
Z-meter. (Again, it may be the strays in the test fixture thing).
The VNA technique has practical appeal to me and,as observed by W8JI (I
think), we often don't and can't know the CM Z in the circuit we're
choking. So I have concluded that, since this technique is stable and
repeatable, it's what I'm going to hang my hat on.
Comments, please?
73,
geo - n4ua
On Sat, Jul 14, 2012 at 3:56 PM, Richard (Rick) Karlquist <
richard@karlquist.com> wrote:
>
>
> On 7/14/2012 11:14 AM, pfizenmayer wrote:
>
> > Which leads me to ask - what is wrong with measuring the attenuation
> with a
> > two port VNA with the choke simply in series between port 1 and port 2
> ?(of
> > course calibrating out any cables if used) This is basically a "VI"
> > measurement and seems to me to give valid measure of the atteuation if
> the
>
> > Hank K7HP
>
> I have been doing this for over 30 years, and have taught many
> others to do it. However, what is "wrong" with the technique
> is that network analyzers are only specified in terms of an
> insertable coaxial DUT with defined 50 ohm reference planes.
> It is up to the user to fixture his actual DUT to look like
> this. The fixturing problem is the "final frontier" of
> measurement science. The experts don't always agree on what
> is the "correct" fixturing modality. Having said that, it
> is still plenty useful for ham radio.
>
> A minor issue is that the user has to be able to convert
> between the complex variables of S21 and Z. This doesn't
> seem to be built in to any NA's that I have seen. Sometimes,
> you can do user math and add the capability.
>
> The dual of this technique is useful for very low impedance
> DUT's. In this case, you shunt the through line with the DUT.
> Watch out for ground loops.
>
> Rick N6RK
>
> _______________________________________________
> UR RST IS ... ... ..9 QSB QSB - hw? BK
>
_______________________________________________
UR RST IS ... ... ..9 QSB QSB - hw? BK
|