开云体育

OOOOoops,........That's W?LEV, not WLEV. For some reason I have to wait a
few seconds to release the "ALT" button for the "zero" in ASCCI code to
register. Who knows.......it's a digital world........

Dave - W?LEV

On Thu, Aug 4, 2022 at 10:57 PM David Eckhardt <davearea51a@...>
wrote:

Jim, the NANOVNAs I contend are highly useful for us amateurs and "top
level" professional applications and CERTAINLY for education. However, I
use mine where I can not carry the (100-lb.) 8753C with S-Parameter Test
Set. That and the price point is where the NANOVNAs shine for the amateur
and professional. Of course, Keysite has offered a portable VNA as does
R&S, but the price point(s) keep us amateurs away. Fortunately, at one
time (pre-Carley), HP encouraged dumpster dipping. For those of us who
practiced our skills on the hobby as well as on the job, that was quite a
priviledge!!!

Dave - WLEV

On Thu, Aug 4, 2022 at 5:50 PM Jim Lux <jimlux@...> wrote:

On 8/3/22 4:38 PM, 0root via groups.io wrote:

Hey all, first post here.

I currently own and love the nanovna-h and h4, I was sure to buy them

from the alibaba store zeenko as this is apparantly hugens chosen
manufacturer.

However, if I wanted to get more precision - specifically in the areas

of Return Loss / VSWR and attenuation readings - what would be the next
device up to get - within the region of $100 - $2000 and no higher...

I see alot of siglent stuff, but the reviews are not too appetizing

One other thing..
The "raw measurement" that the NanoVNA makes is essentially a filtered
sum of 48 samples. The raw adc samples are probably about 1 part in 10^4
(call it 1E-4 fractional uncertainty) so 48 samples will be sqrt(N)
better, or 1.4E-5. (ignoring arithmetic precision).

That's voltage, not power, relative to full scale.

In reality, the input signal isn't full scale, it's more like 0.1 full
scale or 0.05 full scale, so the uncertainty is about 0.1%

The reflection coefficient is calculated as the algebraic combination of
two of those measurements, so the uncertainty is roughly doubled (0.2%)

So that really sets your accuracy - especially with a big signal (e.g.
the reflection of a short or open) - with a good load, the accuracy is
less, because reflection power is less, so the uncertainty of that
measurement is poorer. If the reflected power is -40 dB, then the
signal to noise of the measurement is 1/100th, so instead of a 0.1%
uncertainty, it's more like 10%



More expensive analyzers will have better SNR, from a larger stimulus
signal, a lower input noise floor, and a narrower detection bandwidth
(which reduces the noise).

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

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


--
Dave - W?LEV

Jim, the NANOVNAs I contend are highly useful for us amateurs and "top
level" professional applications and CERTAINLY for education. However, I
use mine where I can not carry the (100-lb.) 8753C with S-Parameter Test
Set. That and the price point is where the NANOVNAs shine for the amateur
and professional. Of course, Keysite has offered a portable VNA as does
R&S, but the price point(s) keep us amateurs away. Fortunately, at one
time (pre-Carley), HP encouraged dumpster dipping. For those of us who
practiced our skills on the hobby as well as on the job, that was quite a
priviledge!!!

Dave - WLEV

--
Dave - W?LEV

Yet another reason to become familiar with the complex representation of
the impedance! Oh,.....yes,......AND the Smith Chart.

Dave - W?LEV

On Thu, Aug 4, 2022 at 8:43 PM Maynard Wright, P. E., W6PAP <w6pap@...>
wrote:

This constant SWR circle is certainly a useful exposition of the concept
and is very
nicely done. Having the Smith Chart display on the nanoVNA can be very
helpful even
if you retrieve data numerically. It can validate the way that you
interpret the data.

It's important to note that SWR = Zt / Zo is correct only when Xt and Xo
both equal
zero. For most cases of interest to amateur radio Xo will always be zero,
but Xt will
most often not be zero, so the simplified equation is of limited use.

73,

Maynard
W6PAP



On Thursday, August 04, 2022 09:15:26 AM Roger Need via groups.io wrote:

On Wed, Aug 3, 2022 at 06:52 PM, Kenneth Hendrickson wrote:

On both charts, 25+j0 and 100+j0 need to be swapped. They are wrong as
labeled.

Thanks for the correction. I put these together quickly and should have
proofread them for errors. Here are the corrected versions showing

several

complex impedance values that all result in an SWR of 2.

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


--
Dave - W?LEV

On Thu, Aug 4, 2022 at 01:43 PM, Maynard Wright, P. E., W6PAP wrote:

It's important to note that SWR = Zt / Zo is correct only when Xt and Xo both > equal
zero. For most cases of interest to amateur radio Xo will always be zero, but > Xt will
most often not be zero, so the simplified equation is of limited use.

I agree that the simple formula is of little use in real world use and it cause more confusion than it is worth.

Also....
SWR = Zt/Zo for Zt >= Zo and Xt and Xo both = 0 Example: Zt = 100 + j0 and Z0 = 50 + j0 then SWR = 100/50 = 2
SWR = Zo/Zt for Zt <= Zo and Xt and Xo both = 0 Example: Zt = 25 + j0 and Z0 = 50 + j0 then SWR = 50/25 = 2

Roger

This constant SWR circle is certainly a useful exposition of the concept and is very
nicely done. Having the Smith Chart display on the nanoVNA can be very helpful even
if you retrieve data numerically. It can validate the way that you interpret the data.

It's important to note that SWR = Zt / Zo is correct only when Xt and Xo both equal
zero. For most cases of interest to amateur radio Xo will always be zero, but Xt will
most often not be zero, so the simplified equation is of limited use.

73,

Maynard
W6PAP

On 8/4/22 10:24 AM, 0root via groups.io wrote:

What do you guys make of these calibration boards?

Accurate enough to give a result with <3% differences?

As noted elsewhere, the basic measurement uncertainty is around 1%, so if your cal standard uncertainty is 1%, then you're in the right ballpark.

But it *really* depends on what you're measuring.

Consider measuring a 40 dB attenuator, and you care about S21 accuracy. The NanoVNA can easily do <1% measurement uncertainty of a 0dB attenuator, but the uncertainty of a signal that is 1/100th the amplitude will be bigger. The signal is smaller, but the noise is the same.

If you look at uncertainty curves for Keysight VNAs, you can see how the uncertainty varies as a function of the return loss (S11) or attenuation (S21)


See page 15

You can see that for S21, at lower frequencies, the uncertainty in magnitude is 0.1 dB (about 1% for linear magnitude) for "low loss" components, but as the loss gets bigger (-70dB) the uncertainty starts to climb. That's what the NanoVNA can do, but it will start to climb at lower attenuations.

It's even more obvious in the S11 (Reflection Uncertainty). For a big mismatch (== lots of reflected power), the uncertainty is low (fractions of a dB), but as the load gets to be higher quality, the uncertainty climbs. A load with a -40dB S11 has an uncertainty of 3-4 dB (40%).

That's with -15dBm stimulus (comparable to the NanoVNA) 10 Hz IF bandwidth (500 times narrower than NanoVNA - that improves the SNR by 27 dB). So if the FieldFox does pretty good to -50dB S21, then the NanoVNA can do almost as well up to -25 dB, but then it will start to degrade.


The FieldFox also probably has a *better* calibration algorithm (more terms, etc), so that helps too.



keysight even offers a calculator for determining the uncertainties - it might be usable with other VNAs

On 8/3/22 4:38 PM, 0root via groups.io wrote:

Hey all, first post here.
I currently own and love the nanovna-h and h4, I was sure to buy them from the alibaba store zeenko as this is apparantly hugens chosen manufacturer.
However, if I wanted to get more precision - specifically in the areas of Return Loss / VSWR and attenuation readings - what would be the next device up to get - within the region of $100 - $2000 and no higher...
I see alot of siglent stuff, but the reviews are not too appetizing

One other thing..
The "raw measurement" that the NanoVNA makes is essentially a filtered sum of 48 samples. The raw adc samples are probably about 1 part in 10^4 (call it 1E-4 fractional uncertainty) so 48 samples will be sqrt(N) better, or 1.4E-5. (ignoring arithmetic precision).

That's voltage, not power, relative to full scale.

In reality, the input signal isn't full scale, it's more like 0.1 full scale or 0.05 full scale, so the uncertainty is about 0.1%

The reflection coefficient is calculated as the algebraic combination of two of those measurements, so the uncertainty is roughly doubled (0.2%)

So that really sets your accuracy - especially with a big signal (e.g. the reflection of a short or open) - with a good load, the accuracy is less, because reflection power is less, so the uncertainty of that measurement is poorer. If the reflected power is -40 dB, then the signal to noise of the measurement is 1/100th, so instead of a 0.1% uncertainty, it's more like 10%



More expensive analyzers will have better SNR, from a larger stimulus signal, a lower input noise floor, and a narrower detection bandwidth (which reduces the noise).

What do you guys make of these calibration boards?



Accurate enough to give a result with <3% differences?

On 8/4/22 9:39 AM, 0root via groups.io wrote:

Siegfried Jackstien - Thanks for the recommendation, I will research these.
John Gord - Re maintaining a consistent length between O, S, L - the three now provided with the nanoVNA are just RPSMA caps... therefore they should meet this criteria?
Thanks for the details on the article, Should be able to find it.
Donald S Brant Jr - Thanks for the recommendation, I'll research LibreVNA
Also, you're right re making your own...a matter of confidence vs doubt... I guess buying multiple kits then testing them all to see the alignment may help put faith in the chosen kit?

The problem is you're in the "observer with two watches, which one is correct" scenario.

Unless you have access to a (much) better measurement tool, you won't know if it's the cal standards or something else.

Another thing to remember is that it's "knowledge" of the cal standard that's important. Your cal standard for a load might range from 45 to 55 ohms with some reactive components. As long as you KNOW what the impedance is at each frequency, that can be incorporated into the calibration process (not inside the NanoVNA, but in some other software).

The basic instrument assumes perfect standards, and that simplifies the calculation of the calibration terms used. But that's not strictly necessary.

W0LEV - Thats good to hear... I have heard that its not worth upgrading as the nanoVNAs are good enough... I guess this is looking more like finding a better calibration kit to use with the N.V

Siegfried Jackstien - Thanks for the recommendation, I will research these.

John Gord - Re maintaining a consistent length between O, S, L - the three now provided with the nanoVNA are just RPSMA caps... therefore they should meet this criteria?

Thanks for the details on the article, Should be able to find it.

Donald S Brant Jr - Thanks for the recommendation, I'll research LibreVNA

Also, you're right re making your own...a matter of confidence vs doubt... I guess buying multiple kits then testing them all to see the alignment may help put faith in the chosen kit?

Don't downplay the accuracy of the NANOVNAs. I have compared several
offerings from the NANOVNAs with the HP 8753C with the associated
S-Parameter test set using HP cal. standards. The agreement, within each
other's limitations, is astounding. Sure, the cal. standards that come
with the NANOs are not the quality of those from HP, but not bad at all,
especially in the HF to low VHF range. I seriously doubt you will find
anything more "accurate" within your targeted price range. If you want to
increase your upper price limit to $10 or $12 killobucks, then, yes, you
can do better. You "might" find a used 8753C, but beware. I have a
friend who went that route for $4k, and the display does not behave
properly.

Dave - W?LEV

On Thu, Aug 4, 2022 at 4:30 AM Siegfried Jackstien <
siegfried.jackstien@...> wrote:

KC901
Different range (and so price) available
Between 1000 and 2000 bucks or so
Much more professional and accurate as all the Nanos... And still portable
use (with a bit bigger display)
So i recommend those if you really want to spent more for a better unit
Dg9bfc sigi

Am 04.08.2022 03:32 schrieb "0root via groups.io" <hntpro@...

:

Hi John

Good guess at both my range of interest(indeed below 900MHz) and the
education side... I have very limited techniques as im new to it all.

Im going to try and find that article you mention, if not il try to buy

an

old hard copy of the mag on ebay.

Re the loads, you are referring to the three calibration connectors
provided correct?
Would making my own finer tuned ones be a realistic task? as in
components, pricing, and skill?
If not if you know anywhere to buy some finer tuned I would appreciate

the

link

Thanks for the help!

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


--
Dave - W?LEV

On Thu, Aug 4, 2022 at 05:32 AM, Arie Kleingeld PA3A wrote:

Lets put it simple without too many formulas. There has been a lot of
math to come to the SWR formulas but we can think about it.

Lots of things to tell about that subject. The following might be
interesting for you:

Arie,

Thank you for posting your paper. I suggest that you correct the text below to state that this formula is only true when the reactance X is 0. I have seen on many occasions people thinking that the magnitude of impedance |Z| can be used in this formula. For example if the complex impedance is 40 + j30 then |Z| =50 and the SWR is 2.0 (when the reference impedance is 50 +j0 ohms). Someone misusing the formula would think |Z|/50 = 1.0 for the SWR which is incorrect.

Roger

On Wed, Aug 3, 2022 at 06:52 PM, Kenneth Hendrickson wrote:

On both charts, 25+j0 and 100+j0 need to be swapped. They are wrong as
labeled.

Thanks for the correction. I put these together quickly and should have proofread them for errors.
Here are the corrected versions showing several complex impedance values that all result in an SWR of 2.

Hi Russ,

Lets put it simple without too many formulas. There has been a lot of math to come to the SWR formulas but we can think about it.

The reflection coefficient is the percentage of reflected voltage from a load. From that and the still going forward voltage we get those standing waves. At some points along the transmissionline the voltages add-up, at other points the substract. If we measure the ratio of the substracted voltages and the added voltages then we get the SWR.

So: From voltages as you well stated.

In the general case the reflectioncoefficient can be a percentage plus an phase shift. That is what the Smith chart shows.

Lots of things to tell about that subject. The following might be interesting for you:




the first ten pages of this presentation will tell you what you want to know.


73

Arie PA3A

Op 3-8-2022 om 17:52 schreef Russ:

You can certainly make a homemade short or open, but characterizing it accurately is another thing entirely. Without accurate characterization a standard is useless.
73, Don N2VGU

I am happy with my LibreVNA, but I also bought a better calibration kit than was provided. The quality of the kit and accuracy of the standards' characterization will directly affect the accuracy of the results. I use it in my professional practice, most recently to characterize a MIL SATCOM IF processing unit I designed for a client for a client. Jan's software support is excellent and he is making ongoing improvements.
73, Don N2VGU

Oroot,
At frequencies up to 1GHz or so, it is certainly practical to make your own Open, Short, and Load. Ideally, the three standards should be the same length, which becomes more critical at higher frequencies because the wavelength gets shorter.
Search for something like "homemade 50 ohm load vna" to get some ideas on what others have done.
The QEX article is available online to ARRL members. Perhaps a nearby friend who is a member could print the article for you. Failing that, the author of the article is Gary Cobb, and his email is g3tmg@.... He might be willing to email you a copy of the article. The title is "Improved Low-loss Measurements with a NanoVNA".
--John Gord

Correction
Between 1400 and 4000
But that one goes to 20 gig hz!!!


Greetz sigi dg9bfc

Am 04.08.2022 06:30 schrieb Siegfried Jackstien <siegfried.jackstien@...>:

KC901
Different range (and so price) available
Between 1000 and 2000 bucks or so
Much more professional and accurate as all the Nanos... And still portable use (with a bit bigger display)
So i recommend those if you really want to spent more for a better unit
Dg9bfc sigi

Am 04.08.2022 03:32 schrieb "0root via groups.io" <hntpro@...>: