开云体育

From a course in EMC / RFI I once taught (assuming ideal components):

[image: image.png]

[image: image.png]




*RESISTANCE* *REACTANCE*

Frequency independent Frequency dependent
(Sits on the real horizontal axis (Sits anywhere on the
Smith
of the Smith Chart) Chart but the
horiz real axis)

Can sink power (as in a dummy load) Can store power (as in charge in
a dielectric -
a
capacitor - or magnetic energy in a magnetic

material - a filter choke)

Can not radiate RF energy Can radiate RF energy


Resonance is *defined* where X(L) = X(C)


Dave - W?LEV

<>
Virus-free.
www.avast.com
<>
<#m_6434894888749294306_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Mar 30, 2020 at 7:20 PM David Eckhardt via Groups.Io <davearea51a=
[email protected]> wrote:

You can learn a whole bunch and pretty much teach yourself by playing with
SimSmith. I use it extensively. It's free and can be accessed at the
following URL. It also offers some excellent tutorials, but one must first
understand and internalize what a complex impedance truly represents.

Tutorials:

Download:

RFCAFE.com also offers a wealth of good information on the subject in
addition to all things RF.

Dave - W?LEV


<

Virus-free.
www.avast.com
<

<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Mar 30, 2020 at 6:26 PM Jerry Gaffke via Groups.Io <jgaffke=
[email protected]> wrote:

Paul,

I got it that you had some familiarity with Smith Charts, and were just
stumped on
why the tutorial was incomplete in identifying the frequencies involved.
I wrote the explanation such that others in the forum could also figure

it

out.

I'm in about the same boat as you.
Slept through EE classes on using the Smith Chart some 45 years ago,
made a career in digital design. Now getting back to my ham radio roots.
It took a few weeks of going over material on the web to get comfortable
with Smith Charts.

I don't feel I have touched bottom unless I can perform the computations.
Python is an excellent choice, complex numbers are implemented in the

core

language.
For example, in Jeff's tutorial

his first example using numbers can be summarized as follows:
Gamma = (Zload - Zo) / (Zload + Zo); If Zload = 50 + j200, then
Gamma = 0.8 + j0.4

Here's a complete python session to duplicate that computation:

def Gamma(Zload, Zo):

... return((Zload - Zo)/(Zload + Zo))
...

Gamma(50+200j, 50)

(0.8+0.4j)

The above example will run the same under Python 3.x as it does under 2.x
Both will give an answer of (0.8 + 0.4j)
But I recommend you stick with Python 3.x, as 2.x is now ancient history.

I'm thinking of someday creating a Python exercise to do the math
(including the plots)
of Jeff's tutorial. Start it out with an introduction to exactly what
complex numbers are.

I agree, a good explanation written for somebody starting from scratch
on such a "complex" subject matter as Smith Charts is not easy to cook

up.

But writing it is a good exercise, you soon realize there are nooks and
crannies
of the subject you haven't quite figured out as well as you thought.

Jerry, KE7ER


On Mon, Mar 30, 2020 at 09:42 AM, Nbridgema wrote:

Jerry, thanks; it was enough to point out that I really hadn't missed

some

piece of information. I understand what you pointed out although I

hadn't

followed it through as thoroughly as you. You were hearing my

frustration of

not being able to follow the statement verbatim.

Now retired, I've been re-engaging with RF electronics, which was a

childhood

fantasy. So I've been looking for tutorials on any number of subjects,

like

the Smith chart (as a refresher), and Python (about which I know less

than

zero, and lack any programming knowledge).

I observe a common denominator: the one skilled in a subject forgets

how

much

they already know that will be required of the listener to understand

their

explanation. And I can't throw stones!! I now know I must have done the

same

thing explaining some aspect of propagation in optical fiber, and saw

the

listener's eyes glaze over!

-Paul
I_B_Nbridgema

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

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

<>
Virus-free.
www.avast.com
<>
<#m_6434894888749294306_DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Mar 30, 2020 at 11:34 AM, DiSlord wrote:

Fix screensot (shold be)
Set minimum freq as 10kHz
Remove DFU menu
====================================================================
See attached screenshot:

1. "capture" console command working again. Thanks!
2. Minimum frequency settable to 10 kHz. Thanks!
3. DFU menu removed.

DISlord, thanks for the quick changes and the 0.8 beta firmware. This firmware version makes the NanoVNA-H4 my preferred device for everyday use.

You can learn a whole bunch and pretty much teach yourself by playing with
SimSmith. I use it extensively. It's free and can be accessed at the
following URL. It also offers some excellent tutorials, but one must first
understand and internalize what a complex impedance truly represents.

Tutorials:

Download:

RFCAFE.com also offers a wealth of good information on the subject in
addition to all things RF.

Dave - W?LEV


<>
Virus-free.
www.avast.com
<>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>

On Mon, Mar 30, 2020 at 6:26 PM Jerry Gaffke via Groups.Io <jgaffke=
[email protected]> wrote:

Paul,

I got it that you had some familiarity with Smith Charts, and were just
stumped on
why the tutorial was incomplete in identifying the frequencies involved.
I wrote the explanation such that others in the forum could also figure it
out.

I'm in about the same boat as you.
Slept through EE classes on using the Smith Chart some 45 years ago,
made a career in digital design. Now getting back to my ham radio roots.
It took a few weeks of going over material on the web to get comfortable
with Smith Charts.

I don't feel I have touched bottom unless I can perform the computations.
Python is an excellent choice, complex numbers are implemented in the core
language.
For example, in Jeff's tutorial

his first example using numbers can be summarized as follows:
Gamma = (Zload - Zo) / (Zload + Zo); If Zload = 50 + j200, then
Gamma = 0.8 + j0.4

Here's a complete python session to duplicate that computation:

def Gamma(Zload, Zo):

... return((Zload - Zo)/(Zload + Zo))
...

Gamma(50+200j, 50)

(0.8+0.4j)

The above example will run the same under Python 3.x as it does under 2.x
Both will give an answer of (0.8 + 0.4j)
But I recommend you stick with Python 3.x, as 2.x is now ancient history.

I'm thinking of someday creating a Python exercise to do the math
(including the plots)
of Jeff's tutorial. Start it out with an introduction to exactly what
complex numbers are.

I agree, a good explanation written for somebody starting from scratch
on such a "complex" subject matter as Smith Charts is not easy to cook up.
But writing it is a good exercise, you soon realize there are nooks and
crannies
of the subject you haven't quite figured out as well as you thought.

Jerry, KE7ER


On Mon, Mar 30, 2020 at 09:42 AM, Nbridgema wrote:

Jerry, thanks; it was enough to point out that I really hadn't missed

some

piece of information. I understand what you pointed out although I hadn't
followed it through as thoroughly as you. You were hearing my

frustration of

not being able to follow the statement verbatim.

Now retired, I've been re-engaging with RF electronics, which was a

childhood

fantasy. So I've been looking for tutorials on any number of subjects,

like

the Smith chart (as a refresher), and Python (about which I know less

than

zero, and lack any programming knowledge).

I observe a common denominator: the one skilled in a subject forgets how

much

they already know that will be required of the listener to understand

their

explanation. And I can't throw stones!! I now know I must have done the

same

thing explaining some aspect of propagation in optical fiber, and saw the
listener's eyes glaze over!

-Paul
I_B_Nbridgema

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

On Mon, Mar 30, 2020 at 11:49 AM, DiSlord wrote:

I try 12kHz offset (in my last commits i add dynamic update dsp sin/cos table
on offset change, work only on 1kHz step) i don`t see diffrences, but possible
i mistake

Also i try speedup audio codec from 48kHz to 60kHz, it allow speedup sweep and
not lost bandwidth

Its option Bandwidth, you can use generator and send AM modulated signal to CH1 port and see signal spectre. Use Bandwidth and see better and better bandwidth filter of signal

See

Paul,

The groups.io website mangled my python session rather badly.
Here it is in a somewhat more readable form:

#######################
def Gamma(Zload, Zo):
return((Zload - Zo)/(Zload + Zo))

Gamma(50+200j, 50)
#######################

In response to that final line, the python interpreter will print this:
(0.8+0.4j)

Jerry

I try 12kHz offset (in my last commits i add dynamic update dsp sin/cos table on offset change, work only on 1kHz step) i don`t see diffrences, but possible i mistake

Also i try speedup audio codec from 48kHz to 60kHz, it allow speedup sweep and not lost bandwidth

Fix screensot (shold be)
Set minimum freq as 10kHz
Remove DFU menu

Paul,

I got it that you had some familiarity with Smith Charts, and were just stumped on
why the tutorial was incomplete in identifying the frequencies involved.
I wrote the explanation such that others in the forum could also figure it out.

I'm in about the same boat as you.
Slept through EE classes on using the Smith Chart some 45 years ago,
made a career in digital design. Now getting back to my ham radio roots.
It took a few weeks of going over material on the web to get comfortable with Smith Charts.

I don't feel I have touched bottom unless I can perform the computations.
Python is an excellent choice, complex numbers are implemented in the core language.
For example, in Jeff's tutorial
his first example using numbers can be summarized as follows:
Gamma = (Zload - Zo) / (Zload + Zo); If Zload = 50 + j200, then Gamma = 0.8 + j0.4

Here's a complete python session to duplicate that computation:

def Gamma(Zload, Zo):

... return((Zload - Zo)/(Zload + Zo))
...

Gamma(50+200j, 50)

(0.8+0.4j)

The above example will run the same under Python 3.x as it does under 2.x
Both will give an answer of (0.8 + 0.4j)
But I recommend you stick with Python 3.x, as 2.x is now ancient history.

I'm thinking of someday creating a Python exercise to do the math (including the plots)
of Jeff's tutorial. Start it out with an introduction to exactly what complex numbers are.

I agree, a good explanation written for somebody starting from scratch
on such a "complex" subject matter as Smith Charts is not easy to cook up.
But writing it is a good exercise, you soon realize there are nooks and crannies
of the subject you haven't quite figured out as well as you thought.

Jerry, KE7ER


On Mon, Mar 30, 2020 at 09:42 AM, Nbridgema wrote:

Jerry, thanks; it was enough to point out that I really hadn't missed some
piece of information. I understand what you pointed out although I hadn't
followed it through as thoroughly as you. You were hearing my frustration of
not being able to follow the statement verbatim.

Now retired, I've been re-engaging with RF electronics, which was a childhood
fantasy. So I've been looking for tutorials on any number of subjects, like
the Smith chart (as a refresher), and Python (about which I know less than
zero, and lack any programming knowledge).

I observe a common denominator: the one skilled in a subject forgets how much
they already know that will be required of the listener to understand their
explanation. And I can't throw stones!! I now know I must have done the same
thing explaining some aspect of propagation in optical fiber, and saw the
listener's eyes glaze over!

-Paul
I_B_Nbridgema

Oh it's clickable, it takes me to a Google doc. This is a screenshot of the
doc!

I'm hoping to have access to a more expensive VNA later this week for comparisons but for now, the nanoVNA-H looks very promising in terms of accuracy near the top of its frequency range:



73,
--
John AE5X

Not sure why it's not click-able...probably needs the prefix.

--
John AE5X

Edy555 also has been experimenting with an IF of 12kHz as this frequency enables a reduction of DSP processing power required. Not sure if the experiment came to a conclusion.


--
NanoVNA Wiki: /g/nanovna-users/wiki/home
NanoVNA Files: /g/nanovna-users/files
Erik, PD0EK

On Mon, Mar 30, 2020 at 08:41 AM, Nbridgema wrote:
After installing ttrftech/NanoVNA version 0.7.1, F_min is again 50 KHz and F_ max is 2.7 GHz, although noisy > 1.5 GHz.
Confirm?
==========================================================
Nbridgema,
I can confirm. edy555 opened the max frequency up to 2.7 GHz in the 0.7.0 release to to allow NanoVNA-H 3.4 version devices to experiment past 1.5 GHz. The hardware changes to the 3.4 version boards made operation to 1.8 GHz pretty good, and from 1.8-2.0 fair to poor. Past 2.0 GHz measurements are not reliable.

I expect similar performance on the NanoVNA-H4 as its hardware improvements are at least as good as the version 3.4 boards.

edy555 also changed the lower operating frequency from 10 kHz back to 50 kHz in the 0.7.0 release. If enough people lobbied for a change back to 10 kHz, I don't see a technical reason for not doing so. I do know edy555 doesn't want the lower operating frequency to get too close to the IF frequency.

- Herb

There are several dozen videos on Larry Benko’s website showing Smith chart basics AND exploring various RF topics.

A good start might be



When I was first introduced to the smith chart it took me a year to get comfortable. Nowadays with the computer doing all the heavy lifting, it’s easy to explore and learn how the Smith chart can be used to gain insight to many RF circuit design prolblems.

Ward
Ae6ty

On Mon, Mar 30, 2020 at 07:39 AM, Jos Stevens wrote:

...works normally under NanoVNASaver. ?
... The extra bandwith settings seem to work when standalone
... I see that the smaller the bandwith, the slower the display updates.
... The DFU option does not work on the H4.
===============================================
DiSlord,
I'm adding the following comments to Jos' reply:

1. The beta firmware is also working fine with all the current software applications I am using. The only thing that isn't working, that was working before, is a script for capturing the NanoVNA-H4 display using the "capture" console command. The following worked previously:

image_data = ser.read(320 * 480 * 2)
rgb_data = struct.unpack(">153600H", image_data)

After updating to the beta firmware, a "struct.error" is always thrown at the second line. All other console commands seem to work.

2. The DISPLAY-BANDWIDTH menu option is a nice addition to the NanoVNA-H4. My expectation was that decreasing the bandwidth would decrease the noise floor, but what actually happens is that the trace is smoothed. It would probably be more accurate to call the menu AVERAGE or VIDEO AVG. As would be expected, the lower the bandwidth setting (higher number of averages applied) the slower the sweep.

3. As Jos noted, the NanoVNA-H4 cannot enter DFU mode from a firmware menu option and you can delete CONFIG-DFU from your code.

4. The bigger font sizes really improve readability of the display. This is especially true of the menu fonts. The menu fonts are now more legible on the 4" NanoVNA-H4 display than they are on my 4.3" NanoVNA-F display, which has an 800x400 resolution.

5. The improved UI features from the 0.7.1 firmware release make navigating the display better. The display markers are easier to grab, and tapping a marker to change trace selection is easier.

6. I noticed vbat_offset, color and bandwidth console commands have been added in this beta release.

Really a nice piece of coding. Its hard to believe you didn't even have a NanoVNA-H4 to work with while writing the beta firmware. I hope hugen decides to merge most or all of the changes into his next official release. Please let me know if you have an idea of what has changed with the "capture" command.

- Herb

At whis version added all my additional Bandwidth improvements, speed

optimizations + bigger font from my old fixes to AA version

Jerry, thanks; it was enough to point out that I really hadn't missed some piece of information. I understand what you pointed out although I hadn't followed it through as thoroughly as you. You were hearing my frustration of not being able to follow the statement verbatim.

Now retired, I've been re-engaging with RF electronics, which was a childhood fantasy. So I've been looking for tutorials on any number of subjects, like the Smith chart (as a refresher), and Python (about which I know less than zero, and lack any programming knowledge).

I observe a common denominator: the one skilled in a subject forgets how much they already know that will be required of the listener to understand their explanation. And I can't throw stones!! I now know I must have done the same thing explaining some aspect of propagation in optical fiber, and saw the listener's eyes glaze over!

-Paul
I_B_Nbridgema

Well John, I'm sure that you love me but what the heck is up with that link?

After installing ttrftech/NanoVNA version 0.7.1, F_min is again 50 KHz and F_ max is 2.7 GHz, although noisy > 1.5 GHz.
Confirm?
--
I_B_Nbridgema

I ordered mine from this Amazon vendor:
shorturl.at/ahsS7

Order placed on Saturday, Amazon said it would arrive Monday (today) but it was delivered at noon yesterday, 20 hours after placing the order. Compared to the first of these nanoVNA's I ordered 6 months ago, this one seems much more refined in terms of packaging, having a hard case included, etc.

--
John AE5X

On Sun, Mar 29, 2020 at 08:39 AM, jac hil wrote:

The problem with P O Box is the battery in the VNA.
PO rules prohibit a lot of things that burn and record how the parcel was
shook in transit.

That is absolutely not true. The only restriction for Lithium batteries (or any other hazardous materials) is air freight. Nothing to do with a PO Box being the final recipient. The problem is not at the destination, but rather in the process to get it there. I have received many times various Lithium batteries for radios, cell phones etc, even the backup cell I needed to refit in my HP3468A DMM via my PO Box absolutely without any issues. Some of the bigger packages even had the Lithium battery inside sticker warning as required.

The ONLY valid reason to exclude a PO Box is if the shipment is being sent by UPS, Fedex, or DHL, which require a signature upon delivery, so they do not deliver to a PO Box address.

There are US based sellers on ebay (a lot of them chinese) that advertise shipping with USPS in capital big letters and still exclude PO Box on various smallish items, nothing hazardous. The reason is usually that they are drop shipping, and are not really the entity that will handle the shipping so they have no control over it, and are not even able to provide you with a tracking number, or just send you a fake one. Sometimes you end up with an Amazon box even tough you made your purchase on ebay. Been there seen it happen, now I steer clear of any seller that is US based that excludes PO Box. Something funny going on there for sure. Life is short why waste time with drop-shippers.

And as I said before, the PO Box address is my ONLY practical option for small packages. I am not in the US. I do have a street address as well, but only for freight cargo so it has a base charge and does not make sense for small items. Been doing mail forwarding for many years with 99.99% success rate.