Measuring chokes.../alligator test leads SOL

Richard Hill 2023/09/17 15:45

New NanoVNA user...I was planning to make test leads from SMA connecters to short 50 ohm coax to alligator clips, and do a short/open/50 ohm resister calibration.

I'm begining to think that may not be a good setup for HF choke measurements?

Thanks,

Rich, NU6T

William Hays 2023/09/18 02:52

I do OSL calibration using cable with alligator clips on frequencies under 30MHz all the time. Use quality 50 Ohm resistor and keep the fanned out part of the cable as short as possible.

In addition to cables with clipped ends, I have made prototype testing jig. I use a good 50 Ohm resistor and a short piece of copper for calibrations.

The example shows a simple 10Mhz pass and 20MHz stop mock up. I have successfully transferred several prototypes to permanent circuits using this jig.

William
AE4QL

Jig 2.jpg

Tom W8JI 2023/09/18 07:27

The setup depends on what you are trying to accomplish, the physical and
electrical component size, and what is around the choke in the test and
in the final location.

A physically small choke of modest impedance is not environmentally
sensitive. Clips with short leads to a connector can work just fine.

A larger choke, such as a high impedance high power RF choke where
resonances are involved, might need a test fixture and perhaps an
additional test. I use one of several test fixtures made from raw double
sided circuit board for vacuum tube amplifier RF plate chokes. I also
excite them with high voltage RF and look for hot spot, since heat is
not evenly distributed once near or above the lowest self resonance.

Problems testing some chokes are similar to problems when testing large
RF inductors, traps, or loading coils. It isn't so much the clips that
are the problem, rather it is the surrounding environment and precisely
what we are trying to learn. I bought a few nanoVNAs for RF tests but
often find them inadequate. For rough hobby stuff and modest impedances
I can get by with a miniVNA.

I'm just in the process of buying a used calibrated Keysight ENA. I
really wish the miniVNA would be suitable!

73 Tom

On 9/17/2023 6:45 PM, Richard Hill wrote:

> New NanoVNA user...I was planning to make test leads from SMA
> connecters to short 50 ohm coax to alligator clips, and do a
> short/open/50 ohm resister calibration.
>
> I'm begining to think that may not be a good setup for HF choke
> measurements?
>
> Thanks,
>
> Rich, NU6T
>

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

Leif M 2023/09/20 07:46

How about testing a high value resistor. That is, 1kilo-ohms or so. If the VNA shows that correctly, it works with other components, too?

I have measured components with wire leads and alligator clips. It works below 30MHz, but I had to shorten my test leads a bit. If the impedance of your DUT is much below or abowe 50 ohms, you will have problems.

Bob W0EG 2023/09/20 12:45

Keep in mind that a real resistor is also an inductor and a capacitor. It’s
impedance is a function of frequency. Wirewound power resistors are, of
course, quite inductive at frequency. “Non-inductive” resistors are still
inductive at higher frequencies depending on size and lead length.

On Wed, Sep 20, 2023 at 11:28 AM Leif M <leif.michaelsson@gmail.com> wrote:

- Show quoted text -

Tom W8JI 2023/09/20 13:39

Pure resistances change level of the bridge, not phase. Inductances and
capacitances shift phase, a very high reactance value compared to
resistance near 90-degrees and a low value near zero.

Pure resistances will not tell you much about ability to measure
reactances, since they change different things.

On 9/20/2023 10:46 AM, Leif M wrote:

> How about testing a high value resistor. That is, 1kilo-ohms or so. If
> the VNA shows that correctly, it works with other components, too?
>
> I have measured components with wire leads and alligator clips. It
> works below 30MHz, but I had to shorten my test leads a bit. If the
> impedance of your DUT is much below or abowe 50 ohms, you will have
> problems.
>

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

Leif M 2023/09/20 11:11

"Keep in mind that a real resistor is also an inductor and a capacitor. It’s impedance is a function of frequency." Sure. That's why I didn't suggest a larger resistor. But what should the impedance of a common resistor be. After that, measuring that with NanoVNA would help.

Richard Hill 2023/09/20 20:27

Tom, thanks. I am just beginning to learn what a nanoVNA can do.

My first project goal is to build several 1:1 chokes on 240-43 tortoids
with 16 AWG wire to learn how well they choke common mode currents for
primarily 10-15-20m. I'll likely build a set of chokes with 6, 12, &
then18 turns.

Rich, NU6T

On Mon, Sep 18, 2023, 8:59 AM Tom W8JI <w8ji@w8ji.com> wrote:

- Show quoted text -

Tom W8JI 2023/09/21 06:57

Hi Rich,

Feed line common mode is complex. A high choke impedance does not
necessarily mean common mode is reduced or suppressed by a certain
amount. It doesn't mean it has been suppressed at all!

The common mode system has standing waves. These standing waves are
caused by the surge impedance of the shield or balanced line. This surge
impedance varies with transmission line grounding, physical size, and
environment. This impedance varies from very low to vary high, depending
mostly on feedline grounding and routing.

In almost any case of coaxial cable I have modeled or measured if more
than 1000 ohms CM impedance is needed, the system has a design problem.
Some systems need nothing for suppression, some need a modest value of
suppression, and some cannot be suppressed with any amount of choking
impedance.

That aside I measure chokes and other physically large components in
test fixtures made from blank bare circuit board, although I sometimes
use copper sheeting. I put a connector on the fixture and calibrate to
that point. As a load I use a pair of small 100-ohm 1% surface mount
resistors in parallel on a small section of PC board. The board is cut
with a dremel to give ground, 50-ohm, and open positions for the clip
lead with the lead stretched close to test position.

Around the middle background of this box you might be able to see a
movable O,S,L test strip. I solder it where I need it. I test as close
to the real world environment as I can. 16ga wire might not represent
your final design at high frequencies. You need to verify the final
product in its final box.

The miniVNA unfortunately are unable to do impedance extremes
accurately, but they work adequately for what you are doing.

73 Tom

On 9/20/2023 11:27 PM, Richard Hill wrote:
> Tom, thanks. I am just beginning to learn what a nanoVNA can do.
>
> My first project goal is to build several 1:1 chokes on 240-43
> tortoids with 16 AWG wire to learn how well they choke common mode
> currents for primarily 10-15-20m. I'll likely build a set of chokes
> with 6, 12, & then18 turns.
>
> Rich, NU6T
>
> On Mon, Sep 18, 2023, 8:59 AM Tom W8JI <w8ji@w8ji.com> wrote:
>
> The setup depends on what you are trying to accomplish, the
> physical and electrical component size, and what is around the
> choke in the test and in the final location.
>
> A physically small choke of modest impedance is not
> environmentally sensitive. Clips with short leads to a connector
> can work just fine.
>
> A larger choke, such as a high impedance high power RF choke where
> resonances are involved, might need a test fixture and perhaps an
> additional test. I use one of several test fixtures made from raw
> double sided circuit board for vacuum tube amplifier RF plate
> chokes. I also excite them with high voltage RF and look for hot
> spot, since heat is not evenly distributed once near or above the
> lowest self resonance.
>
> Problems testing some chokes are similar to problems when testing
> large RF inductors, traps, or loading coils. It isn't so much the
> clips that are the problem, rather it is the surrounding
> environment and precisely what we are trying to learn. I bought a
> few nanoVNAs for RF tests but often find them inadequate. For
> rough hobby stuff and modest impedances I can get by with a miniVNA.
>
> I'm just in the process of buying a used calibrated Keysight ENA.
> I really wish the miniVNA would be suitable!
>
> 73 Tom
>
>
>
> On 9/17/2023 6:45 PM, Richard Hill wrote:
>
>> New NanoVNA user...I was planning to make test leads from SMA
>> connecters to short 50 ohm coax to alligator clips, and do a
>> short/open/50 ohm resister calibration.
>>
>> I'm begining to think that may not be a good setup for HF choke
>> measurements?
>>
>> Thanks,
>>
>> Rich, NU6T
>
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
> Virus-free.www.avg.com
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
>
>
> <#m_9184977105402679036_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

Small test box.jpg

Richard Hill 2023/09/21 09:10

Thanks, Tom. It is clear that I have a long journey to understand these
systems.

My goal has been to gain a basic understanding, and to build an appropriate
choke for my BuddiHex antenna, and perhaps understand whether a choke is
needed. That will take a bit of play to understand.

Thanks

Rich, NU6T

On Thu, Sep 21, 2023, 7:40 AM Tom W8JI <w8ji@w8ji.com> wrote:

- Show quoted text -

The Greene Family 2023/09/22 06:46

As W8JI said, these small VNAs are really great for a lot of applications, but they are limited once you get way outside the 50 Ohm impedance region. Lots of articles have been written about that. Even the very expensive VNAs are limited, but to a lesser degree. I guess you do get something for 20 dB or so more money.

And, as Tom has suggested, the actual performance of common mode chokes depends largely on the environment they are used in.

So, sticking with W8JI, have you considered using something like this to measure common mode currents on your feedline?

Current Meter ( https://www.w8ji.com/building_a_current_meter.htm )

Test your chokes in actual use.

73,

Clarke K1JX

W0LEV 2023/09/24 17:05

Since I use open wire for my transmission line to/from the antenna
feedpoint, I built a little box for the purpose of reading current on each
of the two conductors. Each of the two conductors is outfitted with a
current probe similar to that illustrated in the reference. I can switch
between the two current probes. I have tested and calibrated each using a
non-inductive resistor across the "output" to read equal values. In place
on the feedline, if both current probes read the same, I know my common
mode choke is doing the job.

Dave - WØLEV

On Fri, Sep 22, 2023 at 6:28 PM The Greene Family <cvgreene@snet.net> wrote:

> As W8JI said, these small VNAs are really great for a lot of applications,
> but they are limited once you get way outside the 50 Ohm impedance region.
> Lots of articles have been written about that. Even the very expensive
> VNAs are limited, but to a lesser degree. I guess you do get something for
> 20 dB or so more money.
>
> And, as Tom has suggested, the actual performance of common mode chokes
> depends largely on the environment they are used in.
>
> So, sticking with W8JI, have you considered using something like this to
> measure common mode currents on your feedline?
>
> Current Meter <https://www.w8ji.com/building_a_current_meter.htm>
>
> Test your chokes in actual use.
>
> 73,
>
> Clarke K1JX
>
>
>
>
>

--

*Dave - WØLEV*

Tom W8JI 2023/09/24 15:33

Hi Dave,

If the scalar current is compared, it will not confirm balance. It can
only tell you if it might possibly be balanced or can't be balanced. It
can only prove unbalance, it cannot prove balance.

Consider this example. If the lines were simply parallel and excited in
phase, 100% unbalance and all common mode, the scalar currents would be
exactly equal but the line would be all common mode and no differential
mode at all.

To confirm balance, the current transformers have to be matched and
connected in series at RF before detection. Then, by inverting phase of
one transformer by 180 degrees and reading the new sum, one phase will
give you the total common mode current and the reversal give the net
differential current. From that you sorta know balance. But you would
have to confirm this at two points separated by a large fraction of a
wavelength.

Any balance measurement must include phase.

As a matter of fact at any single point in the line, to be balanced, the
line must have:

1.) Equal and opposite phase currents
2.) Equal and opposite phase voltage to a neutral plane around the line

Attached is my line balanced test fixture. I use it with a dual channel
vector voltmeter or a four port vector network analyzer. It has two
current samples and two voltage samples and requires external
detection. This allows a single point measurement of balance.

73 Tom

On 9/24/2023 1:05 PM, W0LEV wrote:
> Since I use open wire for my transmission line to/from the antenna
> feedpoint, I built a little box for the purpose of reading current on
> each of the two conductors. Each of the two conductors is outfitted
> with a current probe similar to that illustrated in the reference. I
> can switch between the two current probes. I have tested and
> calibrated each using a non-inductive resistor across the "output" to
> read equal values. In place on the feedline, if both current probes
> read the same, I know my common mode choke is doing the job.
>
> Dave - WØLEV
>
> On Fri, Sep 22, 2023 at 6:28 PM The Greene Family <cvgreene@snet.net>
> wrote:
>
> As W8JI said, these small VNAs are really great for a lot of
> applications, but they are limited once you get way outside the 50
> Ohm impedance region. Lots of articles have been written about
> that. Even the very expensive VNAs are limited, but to a lesser
> degree. I guess you do get something for 20 dB or so more money.
>
> And, as Tom has suggested, the actual performance of common mode
> chokes depends largely on the environment they are used in.
>
> So, sticking with W8JI, have you considered using something like
> this to measure common mode currents on your feedline?
>
> Current Meter <https://www.w8ji.com/building_a_current_meter.htm>
>
> Test your chokes in actual use.
>
> 73,
>
> Clarke K1JX
>
>
>
> --
> *Dave - WØLEV
> *
>
>
>

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

testfixture1.jpg

W0LEV 2023/09/25 16:54

Tom, thanks for the "whack-on-the-side-of-the-head"! I knew this, but was
so motivated in building the little device that I failed to visit phase
relationships. Again, thank you!

Thinking further, if I had two identical current sensors (several turns of
conductor wound identically on identical cores and connected in parallel
should sum to zero if the phases and amplitudes are identical. However,
finding two toroidal cores of identical characteristics might be the devil
in the details as ferrites are typically controlled to only 20%. Thoughts?

Or, with one current sensor - same design - if I passed both leads through
the sensor, if phases were opposite and equal in amplitude, I should get
zero results. Non-zero results would indicate either :
1) phase relationships not 0 and 180-degrees
2) amplitudes not equal with proper phasing
3) lots of CM current.
Of course, any combination of the three above as well.

Of course, the two halves of my 450-foot doublet are certainly not truly
balanced by any means! That would take a monumental effort to prepare the
soil for a goodly area beneath the wires. Also assure the wires are
deployed in a perfectly identical manner. None of this is within the
efforts of the typical amateur, myself included. so.........

Gotta think this over a bit more in depth.😊🙂

Dave - WØLEV

On Mon, Sep 25, 2023 at 4:58 AM Tom W8JI <w8ji@w8ji.com> wrote:

> Hi Dave,
>
> If the scalar current is compared, it will not confirm balance. It can
> only tell you if it might possibly be balanced or can't be balanced. It can
> only prove unbalance, it cannot prove balance.
>
> Consider this example. If the lines were simply parallel and excited in
> phase, 100% unbalance and all common mode, the scalar currents would be
> exactly equal but the line would be all common mode and no differential
> mode at all.
>
> To confirm balance, the current transformers have to be matched and
> connected in series at RF before detection. Then, by inverting phase of one
> transformer by 180 degrees and reading the new sum, one phase will give you
> the total common mode current and the reversal give the net differential
> current. From that you sorta know balance. But you would have to confirm
> this at two points separated by a large fraction of a wavelength.
>
> Any balance measurement must include phase.
>
> As a matter of fact at any single point in the line, to be balanced, the
> line must have:
>
> 1.) Equal and opposite phase currents
> 2.) Equal and opposite phase voltage to a neutral plane around the line
>
> Attached is my line balanced test fixture. I use it with a dual channel
> vector voltmeter or a four port vector network analyzer. It has two current
> samples and two voltage samples and requires external detection. This
> allows a single point measurement of balance.
>
> 73 Tom
>
> On 9/24/2023 1:05 PM, W0LEV wrote:
>
> Since I use open wire for my transmission line to/from the antenna
> feedpoint, I built a little box for the purpose of reading current on each
> of the two conductors. Each of the two conductors is outfitted with a
> current probe similar to that illustrated in the reference. I can switch
> between the two current probes. I have tested and calibrated each using a
> non-inductive resistor across the "output" to read equal values. In place
> on the feedline, if both current probes read the same, I know my common
> mode choke is doing the job.
>
> Dave - WØLEV
>
> On Fri, Sep 22, 2023 at 6:28 PM The Greene Family <cvgreene@snet.net>
> wrote:
>
>> As W8JI said, these small VNAs are really great for a lot of
>> applications, but they are limited once you get way outside the 50 Ohm
>> impedance region. Lots of articles have been written about that. Even the
>> very expensive VNAs are limited, but to a lesser degree. I guess you do
>> get something for 20 dB or so more money.
>>
>> And, as Tom has suggested, the actual performance of common mode chokes
>> depends largely on the environment they are used in.
>>
>> So, sticking with W8JI, have you considered using something like this to
>> measure common mode currents on your feedline?
>>
>> Current Meter <https://www.w8ji.com/building_a_current_meter.htm>
>>
>> Test your chokes in actual use.
>>
>> 73,
>>
>> Clarke K1JX
>>
>>
>
>
>
> --
>
> *Dave - WØLEV *
>
>
>
>
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
> Virus-free.www.avg.com
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
> <#m_3738195604460637062_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>
>

--

*Dave - WØLEV*

Tom W8JI 2023/09/26 08:02

Hi Dave,

They would be very close even in production because the required load to
create a current transformer also swamps out small impedance
irregularities. Each is terminated, as they must be.

They must have some capacitance intervention between winding and
transmission line conductor, plus high voltage isolation if used live.

They aren't a problem.

Any common mode current will show as unbalance, but since the standing
wave points of common mode can be different than the differential, to
get a good idea at one point the voltage balance (including phase) also
must be checked.

I have a long video or article I started on balance and common mode. I
feel it is critically important because we are off on "a high impedance
choke fixes everything" trip that is largely false. I just lack time.

My best balanced antenna is a sloppy 75M low dipole near buildings and
other junk, different leg heights. Without any balun or choke it has
virtual no CMC. If a choke is added it actually becomes worse.
My worst balance dipole is 130ft high, in the clear perfectly, with coax
dropped straight down to ground. It horrible without a choke yet a
modest choke cleans it up about the same as a nose bleed choke.

The is mostly because of electrical length from source (antenna
terminals) to to feedline ground and resulting ability to drive common
mode onto the cable, not antenna symmetry.

73 Tom

On 9/25/2023 12:54 PM, W0LEV wrote:
> Tom, thanks for the "whack-on-the-side-of-the-head"! I knew this, but
> was so motivated in building the little device that I failed to visit
> phase relationships. Again, thank you!
>
> Thinking further, if I had two identical current sensors (several
> turns of conductor wound identically on identical cores and connected
> in parallel should sum to zero if the phases and amplitudes are
> identical. However, finding two toroidal cores of identical
> characteristics might be the devil in the details as ferrites are
> typically controlled to only 20%. Thoughts?
>
> Or, with one current sensor - same design - if I passed both leads
> through the sensor, if phases were opposite and equal in amplitude, I
> should get zero results. Non-zero results would indicate either :
> 1) phase relationships not 0 and 180-degrees
> 2) amplitudes not equal with proper phasing
> 3) lots of CM current.
> Of course, any combination of the three above as well.
>
> Of course, the two halves of my 450-foot doublet are certainly not
> truly balanced by any means! That would take a monumental effort to
> prepare the soil for a goodly area beneath the wires. Also assure the
> wires are deployed in a perfectly identical manner. None of this is
> within the efforts of the typical amateur, myself included. so.........
>
> Gotta think this over a bit more in depth.😊🙂
>
> Dave - WØLEV
>
>
> On Mon, Sep 25, 2023 at 4:58 AM Tom W8JI <w8ji@w8ji.com> wrote:
>
> Hi Dave,
>
> If the scalar current is compared, it will not confirm balance. It
> can only tell you if it might possibly be balanced or can't be
> balanced. It can only prove unbalance, it cannot prove balance.
>
> Consider this example. If the lines were simply parallel and
> excited in phase, 100% unbalance and all common mode, the scalar
> currents would be exactly equal but the line would be all common
> mode and no differential mode at all.
>
> To confirm balance, the current transformers have to be matched
> and connected in series at RF before detection. Then, by inverting
> phase of one transformer by 180 degrees and reading the new sum,
> one phase will give you the total common mode current and the
> reversal give the net differential current. From that you sorta
> know balance. But you would have to confirm this at two points
> separated by a large fraction of a wavelength.
>
> Any balance measurement must include phase.
>
> As a matter of fact at any single point in the line, to be
> balanced, the line must have:
>
> 1.) Equal and opposite phase currents
> 2.) Equal and opposite phase voltage to a neutral plane around
> the line
>
> Attached is my line balanced test fixture. I use it with a dual
> channel vector voltmeter or a four port vector network analyzer.
> It has two current samples and two voltage samples and requires
> external detection. This allows a single point measurement of
> balance.
>
> 73 Tom
>
> On 9/24/2023 1:05 PM, W0LEV wrote:
>> Since I use open wire for my transmission line to/from the
>> antenna feedpoint, I built a little box for the purpose of
>> reading current on each of the two conductors. Each of the two
>> conductors is outfitted with a current probe similar to that
>> illustrated in the reference. I can switch between the two
>> current probes. I have tested and calibrated each using a
>> non-inductive resistor across the "output" to read equal values.
>> In place on the feedline, if both current probes read the same, I
>> know my common mode choke is doing the job.
>>
>> Dave - WØLEV
>>
>> On Fri, Sep 22, 2023 at 6:28 PM The Greene Family
>> <cvgreene@snet.net> wrote:
>>
>> As W8JI said, these small VNAs are really great for a lot of
>> applications, but they are limited once you get way outside
>> the 50 Ohm impedance region. Lots of articles have been
>> written about that. Even the very expensive VNAs are
>> limited, but to a lesser degree. I guess you do get
>> something for 20 dB or so more money.
>>
>> And, as Tom has suggested, the actual performance of common
>> mode chokes depends largely on the environment they are used in.
>>
>> So, sticking with W8JI, have you considered using something
>> like this to measure common mode currents on your feedline?
>>
>> Current Meter <https://www.w8ji.com/building_a_current_meter.htm>
>>
>> Test your chokes in actual use.
>>
>> 73,
>>
>> Clarke K1JX
>>
>>
>>
>> --
>> *Dave - WØLEV
>> *
>>
>>
>
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
> Virus-free.www.avg.com
> <http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient>
>
>
> <#m_3738195604460637062_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
>
>
>
> --
> *Dave - WØLEV
> *
>
>
>

--
This email has been checked for viruses by AVG antivirus software.
www.avg.com

Beware of cheap underperforming clones

Measuring chokes.../alligator test leads SOL