开云体育

On Sunday 12 July 2020 04:32:00 am Gyula Molnar wrote:

Hi,
You can also visit this website about ferrites:


The first ferrite measurements with nanoVNA were discussed by forum members here: /g/nanovna-users/message/6996 and in subsequent discussions. (/g/nanovna-users/topic/nanovna_saver_0_2_0/54247421?p=Created,,,20,1,20,0&jump=1)

73, Gyula HA3HZ

Wow. Lots of good info there to digest... :-)

--
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 Mon, Jul 13, 2020 at 01:23 PM, Brian Garber wrote:

I think my SD card is recognized and is writing successfully (see below), but
other than taking the SD card out, is there a way to get to the saved data?

At this moment no, USB work only as Serial USB Driver
USB as Mass Storage Device not implemented

I think my SD card is recognized and is writing successfully (see below), but other than taking the SD card out, is there a way to get to the saved data?

Thanks!
Brian

ch> S file
CMD0 Ok
CMD8 Ok
CMD8 0xAA010000
CMD55 + ACMD41 97
CMD58 OCR = 0xC0FF8000
CSD = 40 0E 00 32 5B 59 00 00 ED C8 7F 80 0A 40 40 C3
CardType 12
Mount = 0
Open VNA_200713_202057.s2p, = 0
disk_ioctl(0) = 0,
Close = 0
Read speed = 474074 Byte/s (count 10, time 108)
Write speed = 1545660 Byte/s (count 32, time 106)
Total time: 35ms (write 404289 byte/sec)

Some si5351 chips can non stable work at 300Mhz (on datasheet max 200MHz)

Tru reduce it maximum by set 290-301MHz:
'threshold 301000000' and wait result
if not help reduce
'threshold 299000000' and so on

I check it, but on H4 use longfilenames, and name + filecreation time some

PS on H filename = filecreation time in Hex

So 0x50ED8956
in binary =0b1010000111011011000100101010110
year 0b101000 + 1980 = 40+1980 = 2020
mounth 0b0111 = 7
day = 0b01101 = 13
hour = 0b10001 = 17
min 0b001010 = 10
sec 0b10110 * 2 = 22*2=44

2020/7/13 17:10:44

I check, possibly exist error on time convert

Anton
That is quite a wide freq range. What kind if common mode Z are you looking for? You must maintain the characteristic Z of the RF path so coax wrapped around a core is the way to go. Bifilar is tough to maintain a characteristic Z. A device like this will go thru self resonance like an inductor as frequency is increased. Above self resonance the common mode Z is capacitve and drops with increasing freq. Is the CMC for HF or really for 50k to 300 MHz?

Anton ask,
I guess an attempt to create a decent CM-choke for the 50kHz~300MHz span would be DIFFICULT, more so for a wider span.
Are there any engineering instructions on the topic?

I don't know a lot about CMCs, But I built anantenna matching pcb with a large CMC, and found it rolled off at 1.7MHz.
So, I ask on another group.

I put a CMC on a twisted pair antenna feed line.
It was between the antenna matching transformer
and the feed line. It started to severely attenuate
the RF signal at about 1.7MHz.
The CMC is a 25Mh.

------ What don't I know about CMCs? ------

The response I got,

That they have nonzero leakage inductance? ;0

25 millihenrys with k=0.999 is still 25 uH, which is 270 ohms at 1.7 MHz.
****

So.... you have to have low leakage and the inductance can't be huge, just because it seems like that is better.
Mikek

Has anyone else noticed if the files saved on the SD card are at the correct time?
I saved the data at 16:10 and the file is 17:10.
I checked on the PC and smartphone. Both times, I have the same schedule (17:10).
Pictures in the annex (after a few minutes).

Hi, a week back I got an H4, after initial playing with it I upgraded it to the v.0.8.4.7 fw. Nice!
I see an issue with it, however. With fully charged battery, not connected to PC, CH0 terminated with the stock coax and load, calibrated, range 50kHz-900MHz, it shows expected results after power-on, but after ~15minutes the information on the display gets weird.
The H4 still provides measurements, but the display shows something like this - see attached.
I do not know whether it comes from that fw version, or it is something with H4s hw.
I have not modified the H4s hw yet (ie. xtal + sdcard socket).

Hello,
I exported some data from my nanoVNA S-A-A V2 using the nanoVNA saver software, but when I viewed the .s2p file, S11 values were the exact same as my S21, even though that was not the case in the S11 and S21 plots. I tried to export the data multiple times with different frequency ranges, but I am still having this issue. I would appreciate some help, thanks!

Yes, we somehow must measure with the balun choke "in place".

Here is my practical approach (see my powerpoint):

We can compare that in-place result with another measurement with no balun.
From the ratio of both currents we can calculate the actual effect of the the balun choke.
(We may want to try different balun chokes and can compare, which one of them is best.)

I have done that as can be seen in the powerpoint attached, by measuring
common mode currents in both cases.

To do that I had built a ferrite coupler that I clamped over the coax near my actual Kenwood TS-930S rig.
Don't ask me for the transducer's exact specs. I only found it indicates it's - 17 db between 1 and 30 MHz
in comparison to direct (not clamp over, but in coax) measurement. But as we use the device twice,
it's actual specs don't really matter too much. What matters, is the ratio of both results (in dB). In that
case the same specs apply and cancel out by division.

For detection I first used an RF sniffer circuit together with attenuators, later a very modestly
priced Chinese RF uV meter (shown near the end).

I think, but have I not done that yet, one can also use CH1 of the NanoVNA as well for detection.

It may be possible to obtain a frequency sweep graph across a band of interest
using the NanoVNA with such clamp over transducer.

If breaking the ferrite ring is a problem, one may also use an unbroken ring and slip it over the PL-259 plug of the coax
instead. It just takes a few seconds longer, hi.

I gueth that the NanoNVA's common Ground of CH0 and CH1 should be no problem,
but I have not figured that out in practice.



-----Ursprüngliche Nachricht-----
Von: [email protected] <[email protected]> Im Auftrag von Dana Whitlow
Gesendet: Montag, 13. Juli 2020 14:47
An: [email protected]
Betreff: Re: [nanovna-users] A test fixture for common-mode current chokes AKA 1:1 baluns #measurement

I had a crazy idea for the common mode impedance measurement (of the antenna/feedline system, not of the balun itself), although it could only work well at fairly low frequencies (say, well below 100 MHz as a guesstimate).
This would take advantage of the nano-VNA's very compact size (especially the small ones with the 2.8" display), and amounts to "floating" the VNA.
It's definitely not something I'd ever thought of in the context of lab instruments!

The idea is this: at the physical location where you want to place a common mode choke, short the inner to the outer conductor of the coax line, and short the two connections of the antenna's balanced feedline to each other. Then connect the VNA's CH0 between the two shorting jumpers using the shortest possible connections and do an S11 measurement at that point. First I'd just watch the VNA's display while doing things like grabbing the VNA or letting go, making the measurement with CH0's GND connected to the coax shorting jumper and then repeating the measurement with CH0's GND connected to the balanced line's jumper. The idea is to see if the measurements made under the varying conditions show any reasonable consistency. If they do, I think you can get some useful information in this way.

Note that for this measurement to be meaningful, the shack end of the coax should be connected to the normal equipment in the normal configuration in which it is used, included AC power connections being plugged in. The common mode current along a feedline system can have standing waves too, and the position of the peaks and nulls varies along the feed system and how the ends are "terminated" common-mode- wise. If the VNA measurement described above yields a high impedance at the frequency of interest, this means the common mode choke will do very little, because the common mode current would have been small even without the choke. In stubborn cases it is worth trying adding about a quarter-wave of extra coax to the system and re-evaluating.

But remember that this scheme is not going to work at high frequencies where the size of the VNA is more than a small fraction of a quarter-wave at the frequency of interest.
And the VNA readings will be most accurate if it's "hands off the VNA" while you're taking the final reading. One might even consider moving off a few feet and reading the VNA display with binoculars!

I hope this provides some food for thought. Have fun!

Dana K8yum




--
Diese E-Mail wurde von Avast Antivirus-Software auf Viren geprüft.

On Fri, Jul 10, 2020 at 11:32 PM, Larry Rothman wrote:

Interesting article:

All my life I have been interested in electronics. HOWEVER, I am severely dyslectic; so much so that some (not all) of my grammar school teachers thought me to be retarded. I entered high school as an 'educably handicapped' student; I graduated as a Westinghouse Science Scholar. Well, I'm not retarded as I have advanced degrees in Physics and Mathematics.

Now to my point. For years there was a requirement to learn morse code to get an amateur license. I had 0 (ZERO) chance of ever passing a test for morse code. Thus, I never got a license. One of my professors (I'm retired and studying Electrical and Computer Engineering) wanted all of his students to get license. I explained that I can't do morse code and he told me "you do not need it any longer". I do not know when the requirement was dropped, but......

So, now I have a license and have no idea what to do with it! At 75, I use the internet to keep in contact with people. I study electronics and computer work for fun. While I could probably get a reasonably priced transceiver, WHAT FOR!!! Back when it would have been helpful for my electronics interest, they would not let me get a license. I've read/heard 'hams' lament that people no longer need morse code. I keep my mouth shut, but I want to spit when ever I hear it.

Enough of my problems, larry

If you don't mind a bit of home brewing; Mini Circuits has lots of surface mount pre-amps dirt cheap (<$3), such as:

The only thing to watch is their overload performance. This one, in particular, is DC to 1 GHz, so you may want to add some frequency selectivity for that reason.

Stay well,
Brent

Kevin Schmidt, W9CF, has written about measuring an antenna system's differential and common-mode impedances -- he describes it in the appendix of this article:



I've tried using this method, but with so-so results. The issue, for me, with measuring an antenna system's common-mode impedance is that I want to know it at the point where I plan to insert the choke. For example, at a dipole's feedpoint (or my loop's feedpoint), where the two wire arms leave the coax, rather than back at the transmitter.

I haven't figured out a way to do this apart from simulations with EZNEC. So for the moment I simply shoot for a choke that has a large amount of common-mode resistance, hoping that it will do the job.

- Jeff, k6jca

P.S. Regarding the measurement of a choke's common-mode impedance -- those who have a two-port VNA that can make all four S-parameter measurements (S11, S12, S21, S22) might want to look at this technique that I just put up on my blog:

It offers some improvement over the S21 technique described by G3TXQ. But the S21 technique is perfectly fine, too, in many cases (for those who have, for example, a nanoVNA).

I had a crazy idea for the common mode impedance measurement (of the
antenna/feedline system, not of the balun itself), although it could only work
well at fairly low frequencies (say, well below 100 MHz as a guesstimate).
This would take advantage of the nano-VNA's very compact size (especially
the small ones with the 2.8" display), and amounts to "floating" the VNA.
It's definitely not something I'd ever thought of in the context of lab instruments!

The idea is this: at the physical location where you want to place a common mode
choke, short the inner to the outer conductor of the coax line, and short the two
connections of the antenna's balanced feedline to each other. Then connect the
VNA's CH0 between the two shorting jumpers using the shortest possible connections
and do an S11 measurement at that point. First I'd just watch the VNA's display
while doing things like grabbing the VNA or letting go, making the measurement
with CH0's GND connected to the coax shorting jumper and then repeating the
measurement with CH0's GND connected to the balanced line's jumper. The idea
is to see if the measurements made under the varying conditions show any reasonable
consistency. If they do, I think you can get some useful information in this way.

Note that for this measurement to be meaningful, the shack end of the coax should
be connected to the normal equipment in the normal configuration in which it is used,
included AC power connections being plugged in. The common mode current along
a feedline system can have standing waves too, and the position of the peaks and
nulls varies along the feed system and how the ends are "terminated" common-mode-
wise. If the VNA measurement described above yields a high impedance at the
frequency of interest, this means the common mode choke will do very little, because
the common mode current would have been small even without the choke. In stubborn
cases it is worth trying adding about a quarter-wave of extra coax to the system and
re-evaluating.

But remember that this scheme is not going to work at high frequencies where the size
of the VNA is more than a small fraction of a quarter-wave at the frequency of interest.
And the VNA readings will be most accurate if it's "hands off the VNA" while you're taking
the final reading. One might even consider moving off a few feet and reading the VNA
display with binoculars!

I hope this provides some food for thought. Have fun!

Dana K8yum

You need them solve one problem:
NanoVNA use interpolation then change frequency range/points count, made
measure from PC command. And need original calibration data from flash for
calculations.

I don''t mean use SD card instead of flash, I mean just give the user the option to backup all the current settings to SD card, and restore back to flash if need be.

Hello there!
I absolutely LOVE the v0.8.4.5 (on NanoVNA-H) for its' speed and general feel.
I wish that firmware would be used as THE golden standard for any further development.

Thanks, DiSlord, for you effort!

Warm regards,
Anton

Dear Cliffird,
I guess an attempt to create a decent CM-choke for the 50kHz~300MHz span would be DIFFICULT, more so for a wider span.
Are there any engineering instructions on the topic?

Warm regards,
Anton

On Sun, Jul 12, 2020 at 07:09 PM, Nels Nelsen wrote:

I just installed Linux Mint 20 (which is totally different from Ubuntu)
and revisited the Python VM instructions from Nick G3VNC.
I found I could make those instructions work if I replaced every mention of
"python3.7" with the words "python3.8"

/g/nanovna-users/files/NanoVNA%20PC%20Software/NanoVNA-Saver/nvna-s-pve-rev-c.pdf

Thanks for the LM20 test Nels.

Just tested these instructions again with nanovna-saver-0.3.4 and (apart from the version numbers) it seems to work with my nanoVNA Mk 1.

73
Nick

On Mon, Jul 13, 2020 at 02:47 AM, OneOfEleven wrote:

I was going to add the SD card menu to allow the user to save the config and
all 5 calibrations to the SD card, and allow them to restore them back from
the SD card, but currently too short on available flash area, I'll make it a
compile option with a #define

You need them solve one problem:
NanoVNA use interpolation then change frequency range/points count, made measure from PC command. And need original calibration data from flash for calculations.

Also DiSlord, if the number of sweep points was reduced from 101 to I think 95 points it would allow each calibration save to fit into 4096 bytes of flash

Yes i know (you can see my comments in code about this)

} properties_t;
//on POINTS_COUNT = 101, sizeof(properties_t) == 4152 (need reduce size on 56 bytes to 4096 for more compact save slot size)

And this /g/nanovna-users/topic/73017230
Better reduce ISOLN calibration points to 51, it contain ~noise