开云体育

The wire may be non-resonate without your home brew matching network,but with the network it has to be.??? To be an efficient radiator, you haveto resonate.???? Kent WA5VJB

On Sun, Aug 16, 2020 at 11:04 AM, Jerry Gaffke wrote:

Measurements "at the antenna" can be important if we are trying to design a
better antenna.

But yes, for normal operating we are mostly concerned with
tuning the antenna system (including the coax) for minimum SWR as seen by the
transmitter.

Jerry, KE7ER

On Sun, Aug 16, 2020 at 10:52 AM, David Eckhardt wrote:

So, why the concern for 'at the antenna'
measurements?

I agree with you Jerry.

I suspect most purchasers of a NanoVNA are only using it as a low cost SWR analyzer. They connect it to the antenna feedline (usually coax) and then tune the antenna elements to get a low SWR for the transmitter. Those with a manual "antenna tuner" may adjust the dials for low SWR at several frequencies and record the readings for use when switching frequencies/bands.

Then there is the other group who use the other capabilities of the NanoVNA to do things like design and build antennas, measure component values, phase match cables etc. I have a ZS6BKW, which is a modified G5RV, that I have erected in an inverted V configuration. It works on multiple bands but the frequencies where the low SWR occurs depends on the overall doublet length AND the ladder line length. The two interact in a non intuitive way so it is useful to know the impedance at the doublet feedpoint in order to calculate the length of the ladder line which is acting as an impedance transformer. The NanoVNA and my RigExpert AA55 are useful in this regard.

So each of us uses the NanoVNA in a way that meets our needs.

Roger

How many hams actually match *at the feedpoint* to a 50-ohm coaxial line?
I'll bet very few. At VHF and UHF, yes, at the antenna match is practical
and usually accomplished, but not at HF.

That's why I use open wire feeders where SWR losses are *far* less than in
coaxial cable and the feedline is not stressed even with SWR at full
power. And......I use a single set of wires for 630 through 6-meters with
that system with home brew matching network.

Antennas do not require being resonant to do a good job of radiating. My
system is not resonant (± jX = 0.00, the definition of resonance) in any of
the HF ham bands, but it does very well, both in practice and in the 4NEC2
model. It's lowest 1/2-wavelength resonant frequency is 950 kHz, the lower
1/3 of the AM BC band. I seriously doubt I'd do any better with resonant
dipoles for all the individual bands.

Dave - W?LEV

--
*Dave - W?LEV*
*Just Let Darwin Work*

On Sun, Aug 16, 2020 at 10:31 AM, Jerry Gaffke wrote:

What kind of coax?
I bet it isn't RG174.
I have seen significant differences when checking out an HF antenna over 100
feet of RG8X.

It is 55 feet of RG58 as shown in the plots.

Not totally clear which side of the balun you consider to be the feedpoint of
the antenna
I assume this means you leave the balun attached to the antenna, and only
calibrate out the coax.

My current balun has two eye connections where the dipole is attached and an SO-239 connector on the bottom. For my purposes I have defined this as the "feedpoint" of the antenna. Not technically correct but close enough for me.

Yes I only calibrate out the coax. I do not think calibrating out the coax and the balun is a good idea. Even doing a good test of current balun performance with a NanoVNA requires some care in order to get it right.

Another comment - You need a good balun with decent CMRR or the outside surface of the shield will form part of the antenna and the results will be different than just the dipole itself.

Roger

Roger

Hello David,

Sunday, August 16, 2020

Without knowing what impedance the antenna itself presents, how do I
know the best way to match it to the coax impedance?


Best regards,
Chris mailto:chris@...


DE> What's important is the load the antenna plus feedline present to our
DE> modern transceivers. So, why the concern for 'at the antenna'
DE> measurements? Sure, its nice to know, but the coax contributes it's own
DE> impedance transforming properties. If you must, make the measurement at
DE> the shack end of the feedline, and use a tool such as SimSmith to take out
DE> the impact of the coaxial line.

DE> Again, what is important is the load presented to the transmitter, not
DE> necessarily what the antenna impedance is at the feedpoint. One must STILL
DE> consider the transmission line between the transmitter and the antenna
DE> feedpoint to obtain this result.

DE> Dave - W?LEV

DE> On Sun, Aug 16, 2020 at 5:31 PM Jerry Gaffke via groups.io <jgaffke=
[email protected]>> wrote:

Measurements "at the antenna" can be important if we are trying to design a better antenna.

But yes, for normal operating we are mostly concerned with
tuning the antenna system (including the coax) for minimum SWR as seen by the transmitter.

Jerry, KE7ER

What's important is the load the antenna plus feedline present to our
modern transceivers. So, why the concern for 'at the antenna'
measurements? Sure, its nice to know, but the coax contributes it's own
impedance transforming properties. If you must, make the measurement at
the shack end of the feedline, and use a tool such as SimSmith to take out
the impact of the coaxial line.

Again, what is important is the load presented to the transmitter, not
necessarily what the antenna impedance is at the feedpoint. One must STILL
consider the transmission line between the transmitter and the antenna
feedpoint to obtain this result.

Dave - W?LEV

On Sun, Aug 16, 2020 at 5:31 PM Jerry Gaffke via groups.io <jgaffke=
[email protected]> wrote:

Roger wrote:

You can clearly see that the two plots are nearly identical.

What kind of coax?
I bet it isn't RG174.
I have seen significant differences when checking out an HF antenna over
100 feet of RG8X.

I have a 20M dipole with a current balun attached to it which for this

test I consider to be the "feedpoint" of the antenna.

Not totally clear which side of the balun you consider to be the feedpoint
of the antenna
I assume this means you leave the balun attached to the antenna, and only
calibrate out the coax.

Jerry, KE7ER


On Sun, Aug 16, 2020 at 09:47 AM, Roger Need wrote:

I have a 20M dipole with a current balun attached to it which for this

test I

consider to be the "feedpoint" of the antenna. 55 feet of coax runs back

to

the shack. First measurement with NanoVNA Saver was made at the

feedpoint and

stored as an s1p file. The coax was then "calibrated out" and a second

Saver

measurement made back in the shack. The first s1p file was then loaded

and a

comparison was made. You can clearly see that the two plots are nearly
identical.

Roger

--
*Dave - W?LEV*
*Just Let Darwin Work*

Roger wrote:

You can clearly see that the two plots are nearly identical.

What kind of coax?
I bet it isn't RG174.
I have seen significant differences when checking out an HF antenna over 100 feet of RG8X.

I have a 20M dipole with a current balun attached to it which for this test I consider to be the "feedpoint" of the antenna.

Not totally clear which side of the balun you consider to be the feedpoint of the antenna
I assume this means you leave the balun attached to the antenna, and only calibrate out the coax.

Jerry, KE7ER


On Sun, Aug 16, 2020 at 09:47 AM, Roger Need wrote:

I have a 20M dipole with a current balun attached to it which for this test I
consider to be the "feedpoint" of the antenna. 55 feet of coax runs back to
the shack. First measurement with NanoVNA Saver was made at the feedpoint and
stored as an s1p file. The coax was then "calibrated out" and a second Saver
measurement made back in the shack. The first s1p file was then loaded and a
comparison was made. You can clearly see that the two plots are nearly
identical.

Roger

On Sun, Aug 16, 2020 at 08:12 AM, Jerry Gaffke wrote:

It will be interesting to compare readings of just the antenna vs
the antenna plus the coax and balun.

I have a 20M dipole with a current balun attached to it which for this test I consider to be the "feedpoint" of the antenna. 55 feet of coax runs back to the shack. First measurement with NanoVNA Saver was made at the feedpoint and stored as an s1p file. The coax was then "calibrated out" and a second Saver measurement made back in the shack. The first s1p file was then loaded and a comparison was made. You can clearly see that the two plots are nearly identical.

Roger

Which is better "N" or "SMA" ??

What a funny question !!

In the end you will all end up with SEVERAL versions of EACH.
It just depends on what you are doing for the next five minutes
I have two with SMA, two with BNC (not nano), and pretty soon one with N

What really matters is what on the inside and what your intentions are,
You may as well talk about what flavor cake and ice cream is BEST
The correct answer is "The one in front of you"

Thanks for the space,

NE7LS

On Sun, Aug 16, 2020 at 8:05 AM Jerry Gaffke via groups.io <jgaffke=
[email protected]> wrote:

Looking back at Hugen's original post:
/g/nanovna-users/message/15943
it seems he is simply giving buyers the option of having adapter cables
with the N-type connectors
already installed in the enclosure. An ideal solution.
And I'm fine with him making APC7 connectors available for an additional
$100 as an option
for those that want it.
I assume the nanoVNA board at the core of this still has the SMA's.

Jerry, KE7ER

On Sun, Aug 16, 2020 at 07:57 AM, Jerry Gaffke wrote:

Biastee wrote:

Technical merits aside, by adopting APC7, Hugen / Gabriel can

demonstrate

their impartiality and that they not siding with either N or SMA camp.

Yup, they'd be able to demonstrate they are siding with the APC7 camp.
So with those doing measurements using $10k+ worth of gear.
Which could be lucrative, but not their current target market.

SMA's strike me as the logical choice for a ridiculously small VNA
at a ridiculously low price. All of us should be using short adapter

cables

more or less permanently attached to the nanoVNA just to keep from
ripping up the nanoVNA's circuit board. We can have whatever connector
we want at the far end of that adapter cable at whatever price we wish

to pay.

I don't see a down side here.

Jerry, KE7ER

--




n_n

On Saturday 15 August 2020 02:35:01 pm Shirley Dulcey KE1L wrote:

R&L Electronics ( ) is a good US source of NanoVNA products. They sell the NanoVNA-H, NanoVNA-H4, and
NanoVNA SAA2. The price delivered to a US address is about the same as ordering it from Hugen's store and you get it faster. They also have
listings for the tinySA and the version of the SAA2 with N connectors, but those are currently out of stock.

So which is "best"? The NanoVNA-H and NanoVNA-H4 are limited to 1.5 GHz; the SAA2 goes to 3 GHz. The NanoVNA-H is the least expensive. The
NanoVNA-H4 is the only currently available option with a 4" screen (other than the NanoVNA-F, which Hugen does not sell). You can see all the NanoVNA
products here:


I'm partial to my H4 because of the larger screen and because I'm only currently working on HF projects; your priorities may be different. Here is the link:


I have no connection to R&L other than being a satisfied customer.

I too ordered an H4 from R&L, and am quite satisfied with their service. The unit came nicely packaged, and I also ordered SO239 and BNC adapters to go along with it. I've ordered other stuff from them in the past and have no hesitation in terms of recommending them.

On Sat, Aug 15, 2020 at 2:10 PM Stephen Thompson via groups.io
<v2positiverate@...> wrote:

There seem to be a million variations, prices and knock offs. Could someone please point me to the latest version at a good price from a
reputable outlet in the US of A.

I sure appreciate it, 73
Steve

--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin

On Fri, 14 Aug 2020 at 02:34, Jim Lux <jimlux@...> wrote:

On 8/13/20 5:52 PM, Glen Jenkins WB4KTF wrote:

The 50 OHM load (SMA-Male) that came with my nanoVNA-H4 measures 51.13+

ohms. Not a good start for calibration. Where is a good source for GOOD
loads that are accurate?

that's a VSWR of 1.023:1 and a S11 of -39dB

That’s true only if the reactance is zero.

How much are you willing to pay?

Pasternack has a SMA Male load (PE6002) for $38

Many RF engineers call them Pastercrap. I tested 10 Pastercrap loads and
not one met the specification.

For $1900 they'll sell you a 3.5mm load for a cal kit

up to 4GHz they claim S11 mag <-40dB

A Keysight load, with a better specification, would cost less, and would
actually meet its specification.

That's about the same as the load you have.

The chart in the data sheet actually shows better than -50dB at 1 GHz

That’s “typical”, but I don’t believe much from Pastercrap.

My company could supply decent loads, but I would need to charge nearly
twice what a NanoVNA would cost.


<>

--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales, company number 08914892.
Registered office:
Stokes Hall Lodge, Burnham Rd, Althorne, Chelmsford, Essex, CM3 6DT, United
Kingdom

Chris,
I think you had it right, just add a wire or a metal film resistor
across the screw terminals of the balun when doing the calibration.
An inch or two of wire won't make much difference at 14 mhz,
and I doubt there is a better way to do it. You would have about the
same problem if you took the VNA up there and somehow connected it
to the two wires that normally go into the balun.

It will be interesting to compare readings of just the antenna vs
the antenna plus the coax and balun.

Jerry, KE7ER

Looking back at Hugen's original post: /g/nanovna-users/message/15943
it seems he is simply giving buyers the option of having adapter cables with the N-type connectors
already installed in the enclosure. An ideal solution.
And I'm fine with him making APC7 connectors available for an additional $100 as an option
for those that want it.
I assume the nanoVNA board at the core of this still has the SMA's.

Jerry, KE7ER

Hello Jerry,

Sunday, August 16, 2020

Yes, when I measured using RG-58 test coax the results were very
different from measuring with LDF4-50 coax, That's the influence of
the coax itself. I want to see just the antenna at the shack end, not
the antenna and coax. Given enough length of poor coax I believe you
could see a good impedance match measuring without reference to the
antenna as the coax would swamp the reading, pretty much whatever the
antenna was like. So, how to do the O S L at a none coaxial
connection point remains the question? :)


Best regards,
Chris mailto:chris@...


JGvgi> I think Chris had it right, assuming he wants to measure his antenna
JGvgi> and does not want to measure the effect of adding all that coax and the balun
JGvgi> into the mix:

Now, my set up is transceiver - short LM400 flex pigtail to outside
shack window, 100 feet LDF4-50 coax, short LM400 pigtail around
rotator at the top of the tower, balun, antenna itself. The balun
(actually a choke in reality, toroid wound with a single length of
coax), has two 5mm stainless stud terminals with butterfly nuts. How do I do a O S L
on them, will just leaving them open shorting with heavy gauge bare
copper wire, and putting a 50 Ohm metal film precision resistor across
them be OK for up to 30MHz?

JGvgi> Doing the O-S-L calibration "at the same point where your transceiver will be connected"
JGvgi> is also correct, but you will then be measuring the reflection coefficient
JGvgi> of the entire system consisting of the antenna, balun, and the 100 feet of coax.
JGvgi> That will be a much different reading.

JGvgi> Jerry, KE7ER


JGvgi> On Sun, Aug 16, 2020 at 07:23 AM, Brian wrote:

The O-S-L should be done at the same point where your transceiver will be
connected - so, assuming you have a short (`12 inch) test jumper connected to
Port 0 on the NanoVNA - you would calibrate with your O, S and L at the end of
this test jumper - which then attaches to the same point where you intend on
connecting your radio. Done this way, the Nano VNA will test your balun/choke,
cable and antenna as a whole....

Brian - KI8KY

JGvgi>

Biastee wrote:

Technical merits aside, by adopting APC7, Hugen / Gabriel can demonstrate
their impartiality and that they not siding with either N or SMA camp.

Yup, they'd be able to demonstrate they are siding with the APC7 camp.
So with those doing measurements using $10k+ worth of gear.
Which could be lucrative, but not their current target market.

SMA's strike me as the logical choice for a ridiculously small VNA
at a ridiculously low price. All of us should be using short adapter cables
more or less permanently attached to the nanoVNA just to keep from
ripping up the nanoVNA's circuit board. We can have whatever connector
we want at the far end of that adapter cable at whatever price we wish to pay.
I don't see a down side here.

Jerry, KE7ER

I think Chris had it right, assuming he wants to measure his antenna
and does not want to measure the effect of adding all that coax and the balun
into the mix:

Now, my set up is transceiver - short LM400 flex pigtail to outside
shack window, 100 feet LDF4-50 coax, short LM400 pigtail around
rotator at the top of the tower, balun, antenna itself. The balun
(actually a choke in reality, toroid wound with a single length of
coax), has two 5mm stainless stud terminals with butterfly nuts. How do I do a O S L
on them, will just leaving them open shorting with heavy gauge bare
copper wire, and putting a 50 Ohm metal film precision resistor across
them be OK for up to 30MHz?

Doing the O-S-L calibration "at the same point where your transceiver will be connected"
is also correct, but you will then be measuring the reflection coefficient
of the entire system consisting of the antenna, balun, and the 100 feet of coax.
That will be a much different reading.

Jerry, KE7ER


On Sun, Aug 16, 2020 at 07:23 AM, Brian wrote:

The O-S-L should be done at the same point where your transceiver will be
connected - so, assuming you have a short (`12 inch) test jumper connected to
Port 0 on the NanoVNA - you would calibrate with your O, S and L at the end of
this test jumper - which then attaches to the same point where you intend on
connecting your radio. Done this way, the Nano VNA will test your balun/choke,
cable and antenna as a whole....

Brian - KI8KY

Although APC-7 connectors from China are now available, they still aren't
cheap. One connector from China costs about as much as the entire NanoVNA;
the US-made ones from Amphenol are double that. And you need two of them
for each NanoVNA, not to mention the connecting cables.

On Sun, Aug 16, 2020 at 10:28 AM biastee via groups.io <biastee=
[email protected]> wrote:

Two opposite camps have emerged in the connector debate. I suspect the
preference for either N or SMA is dictated by the application. The 1st camp
consists of those whose primary application is evaluating HF/VHF aerials
and so they chose N connector because it is more congruent with the thick
coax, e.g. RG-213, which they use to minimize loss. On the other hand, the
2nd camp comprises those who evaluate RF circuits - of course, they will
vote for SMA because of the near universal adoption of edge-mounted SMA
females in evaluation boards (one example:
).
In contrast, the N connector cannot be mounted on either 0.8mm or 1.5mm
thick PCB without some form of mechanical backing. Moreover, the size of an
N connector relative the PCB is akin to the proverbial tail wagging the
dog.

I am somewhat surprised and even dismayed that nobody has agitated for the
APC7 which I believe is the default connector on HP/Agilent/Keysight's sub
6 GHz VNAs. For those unfamiliar with VNAs, calibration is typically NOT
done using APC7 standards but with either N or 3.5 mm standards via
adapters. In VNA applications, this sex-less APC7 has two technical
advantages over either N or SMA:
1. Ordinarily, the THRU calibration results in one reference plane at a
male connector and the second reference plane at a female connector. So a
two port DUT will require the odd combination of one male connector and a
female one. Using APC-7 phase matched adapter, the "Swap equal adapters"
method enables a two port device with all female connectors to be
measured.
2. APC7 has lower reflection coefficient than either N or SMA in the sub 6
GHz range. Eg. APC-7 to SMA (m) adapter has a worse case SWR of 1.06. In
comparison, a N (m) to SMA (m) adapter has a worse case SWR of 1.25.

APC7's biggest setback is cost. However, China made APC7s are now
available. :-) Technical merits aside, by adopting APC7, Hugen / Gabriel
can demonstrate their impartiality and that they not siding with either N
or SMA camp.

Two opposite camps have emerged in the connector debate. I suspect the preference for either N or SMA is dictated by the application. The 1st camp consists of those whose primary application is evaluating HF/VHF aerials and so they chose N connector because it is more congruent with the thick coax, e.g. RG-213, which they use to minimize loss. On the other hand, the 2nd camp comprises those who evaluate RF circuits - of course, they will vote for SMA because of the near universal adoption of edge-mounted SMA females in evaluation boards (one example: ). In contrast, the N connector cannot be mounted on either 0.8mm or 1.5mm thick PCB without some form of mechanical backing. Moreover, the size of an N connector relative the PCB is akin to the proverbial tail wagging the dog.

I am somewhat surprised and even dismayed that nobody has agitated for the APC7 which I believe is the default connector on HP/Agilent/Keysight's sub 6 GHz VNAs. For those unfamiliar with VNAs, calibration is typically NOT done using APC7 standards but with either N or 3.5 mm standards via adapters. In VNA applications, this sex-less APC7 has two technical advantages over either N or SMA:
1. Ordinarily, the THRU calibration results in one reference plane at a male connector and the second reference plane at a female connector. So a two port DUT will require the odd combination of one male connector and a female one. Using APC-7 phase matched adapter, the "Swap equal adapters" method enables a two port device with all female connectors to be measured.
2. APC7 has lower reflection coefficient than either N or SMA in the sub 6 GHz range. Eg. APC-7 to SMA (m) adapter has a worse case SWR of 1.06. In comparison, a N (m) to SMA (m) adapter has a worse case SWR of 1.25.

APC7's biggest setback is cost. However, China made APC7s are now available. :-) Technical merits aside, by adopting APC7, Hugen / Gabriel can demonstrate their impartiality and that they not siding with either N or SMA camp.

The O-S-L should be done at the same point where your transceiver will be connected - so, assuming you have a short (`12 inch) test jumper connected to Port 0 on the NanoVNA - you would calibrate with your O, S and L at the end of this test jumper - which then attaches to the same point where you intend on connecting your radio. Done this way, the Nano VNA will test your balun/choke, cable and antenna as a whole....

Brian - KI8KY