Re: Trouble calibrating NanoVna Classic after upgrade to 1.2.20
Flash old firmware and see if you get the same problem.
toggle quoted message
Show quoted text
On Fri, 25 Aug 2023 at 19:58, kumuzu1 <bulchandani@...> wrote:
I'm hoping someone can help me with my inability to calibrate my NanoVna
classic after upgrading
sucessfully to 1.2.20h.dfu.
While attempting to debug this problem, I did the following:
1) Ran a 'clearconfig 1234' via putty
2) Did a Touch Cal
3) Reduced the si535I frequency to 290 Mhz
4) Reset the calibration before starting the calibration process
But the nanoVna just wouldn't confirm a successful calibration when I
tested the Smith Chart with a 50 ohm or short load. I've attached an image
of what i see.
The odd part is that every so often the nanoVna does calibrate successfully
but when
I try and repeat the process it fails to calibrate.
Any clues? I love this little device and it is now a matter of personal
pride to get it to work!
Sanjay
AI6PM
|
Trouble calibrating NanoVna Classic after upgrade to 1.2.20
I'm hoping someone can help me with my inability to calibrate my NanoVna
classic after upgrading
sucessfully to 1.2.20h.dfu.
While attempting to debug this problem, I did the following:
1) Ran a 'clearconfig 1234' via putty
2) Did a Touch Cal
3) Reduced the si535I frequency to 290 Mhz
4) Reset the calibration before starting the calibration process
But the nanoVna just wouldn't confirm a successful calibration when I
tested the Smith Chart with a 50 ohm or short load. I've attached an image
of what i see.
The odd part is that every so often the nanoVna does calibrate successfully
but when
I try and repeat the process it fails to calibrate.
Any clues? I love this little device and it is now a matter of personal
pride to get it to work!
Sanjay
AI6PM
|
Re: NanoVNA-App SWR graph does not match H4 graph
QUOTE: Consider this a benefit of the tiny battery powered Nano VNA. A measurement can be made with minimal impact on the device under test. Indeed. You hit the nail on squarely on the head. Consider the HP 8753C with the S-Parameter test set. It weight in around 100+ lbs. I can not use it with anything that requires a counterpoise or isolation from the instrument as its too large, sits on a metal shelf, and is connected to the power grid: a MAJOR increase in any counterpoise. The NANOVNAs allow me to make measurments that the 8753C really can not make. However, when one needs to measure at or above our 70-cm band, the size of the NANOs become significant. Dave - W?LEV On Fri, Aug 25, 2023 at 3:50?PM Greg Strickland <greg@...> wrote: Last night I was measuring a high Q resonant circuit on the bench with
Nano VNA on battery. The battery was getting low and without physically
moving anything I connected a USB cable from Nano VNA to laptop a few feet
away. The measurement moved just a bit. Not a cause for concern because
it makes sense. Consider this a benefit of the tiny battery powered Nano
VNA. A measurement can be made with minimal impact on the device under
test.
-- *Dave - W?LEV* -- Dave - W?LEV
|
Re: NanoVNA-App SWR graph does not match H4 graph
Dave, I understand your point about the counterpoise extension. However, I would think that the SWR readings would shift the same for both displays. What am I missing.
Bob, KN4HH
toggle quoted message
Show quoted text
On Aug 25, 2023, at 11:18 AM, W0LEV <davearea51a@...> wrote:
?When connected to the PC or laptop, the effective counterpoise is much
longer than when not connected. The counterpoise consists of anything
connected to the measuring unit, including the laptop/PC through the USB
cable. There is nothing wrong with either measurement. You must
electrically remove the "extension" of the counterpoise when connected to
the PC/laptop.
Dave - W?LEV
On Fri, Aug 25, 2023 at 3:05?PM Bob Watson via groups.io <kn4hhptc=
[email protected]> wrote:
I am using a NanoVNA H4 and NanoVNA-App for SWR display. I connected my
end fed half wave antenna to the CNA and compared the two graphical
displays. Both the NanoVNA and H4 displays were set for identical SWR and
frequency ranges (1:1 to 3:1 and 1-30Mhz).
The difference in displays is dramatic (ie 3Mhz. 2.1:1 vs over 3:1 SWR).
Has anyone observed a difference between the H4 and external display
software.
I would appreciate any comments.
Bob, KN4HH
--
*Dave - W?LEV*
--
Dave - W?LEV
|
Re: NanoVNA-App SWR graph does not match H4 graph
On Fri, Aug 25, 2023 at 08:50 AM, Greg Strickland wrote:
Last night I was measuring a high Q resonant circuit on the bench with Nano
VNA on battery. The battery was getting low and without physically moving
anything I connected a USB cable from Nano VNA to laptop a few feet away. The
measurement moved just a bit. Not a cause for concern because it makes sense.
Consider this a benefit of the tiny battery powered Nano VNA. A measurement
can be made with minimal impact on the device under test.
Performance/measurements of the NanoVNA might change slightly with battery level. Try making the same standalone measurement with a full battery. Roger
|
Re: NanoVNA-App SWR graph does not match H4 graph
Last night I was measuring a high Q resonant circuit on the bench with Nano VNA on battery. The battery was getting low and without physically moving anything I connected a USB cable from Nano VNA to laptop a few feet away. The measurement moved just a bit. Not a cause for concern because it makes sense. Consider this a benefit of the tiny battery powered Nano VNA. A measurement can be made with minimal impact on the device under test.
|
Re: VNA and antenna matching : calculating reactance needed to match
If you calibrate with the measurement plane at the antenna end of the choke
balun, you can measure the antenna impedance correctly. You need to
calibrate out the choke balun.
Dave - W?LEV
toggle quoted message
Show quoted text
On Fri, Aug 25, 2023 at 3:32?PM Fran?ois <18471@...> wrote: I have traps doublets (3 or 4 bands) connected by about 10 m of coax to
the station. I have always measured the impedance of my antennas at the
level of the center by calibrating (open short load) at the end of the
caoxial removed from the antenna, from my PC and nanovna-saver. I then
calculated and made my 'L' adapters to place them in the center of the
doublet. It worked without a problem, without a chock balun. The ROS
becomes correct and it is correct at the station, at the calculation
frequency only, obviously
When I go to the nanoVNA of the chock balun, I doubt that we can measure
anything correct beyond the chock balun. But hey, hope gives life, as well
as those who sell chock-balun.
73
--
F1AMM
Fran?ois
-- *Dave - W?LEV*
--
Dave - W?LEV
|
Re: VNA and antenna matching : calculating reactance needed to match
I have traps doublets (3 or 4 bands) connected by about 10 m of coax to the station. I have always measured the impedance of my antennas at the level of the center by calibrating (open short load) at the end of the caoxial removed from the antenna, from my PC and nanovna-saver. I then calculated and made my 'L' adapters to place them in the center of the doublet. It worked without a problem, without a chock balun. The ROS becomes correct and it is correct at the station, at the calculation frequency only, obviously
When I go to the nanoVNA of the chock balun, I doubt that we can measure anything correct beyond the chock balun. But hey, hope gives life, as well as those who sell chock-balun.
73
--
F1AMM
Fran?ois
|
Re: NanoVNA-App SWR graph does not match H4 graph
When connected to the PC or laptop, the effective counterpoise is much longer than when not connected. The counterpoise consists of anything connected to the measuring unit, including the laptop/PC through the USB cable. There is nothing wrong with either measurement. You must electrically remove the "extension" of the counterpoise when connected to the PC/laptop. Dave - W?LEV On Fri, Aug 25, 2023 at 3:05?PM Bob Watson via groups.io <kn4hhptc= [email protected]> wrote: I am using a NanoVNA H4 and NanoVNA-App for SWR display. I connected my
end fed half wave antenna to the CNA and compared the two graphical
displays. Both the NanoVNA and H4 displays were set for identical SWR and
frequency ranges (1:1 to 3:1 and 1-30Mhz).
The difference in displays is dramatic (ie 3Mhz. 2.1:1 vs over 3:1 SWR).
Has anyone observed a difference between the H4 and external display
software.
I would appreciate any comments.
Bob, KN4HH
-- *Dave - W?LEV* -- Dave - W?LEV
|
NanoVNA-App SWR graph does not match H4 graph
I am using a NanoVNA H4 and NanoVNA-App for SWR display. I connected my end fed half wave antenna to the CNA and compared the two graphical displays. Both the NanoVNA and H4 displays were set for identical SWR and frequency ranges (1:1 to 3:1 and 1-30Mhz).
The difference in displays is dramatic (ie 3Mhz. 2.1:1 vs over 3:1 SWR).
Has anyone observed a difference between the H4 and external display software.
I would appreciate any comments.
Bob, KN4HH
|
Re: VNA and antenna matching : calculating reactance needed to match
Based on my limited knowledge I think with regard to L networks there are choices in topology, depending on antenna impedance and desired phase shift in the L network. Also practical concerns such as desire to isolate DC from the transmitter, or to put an otherwise floating antenna at DC ground potential without having to use (and analyze) a separate "static choke".
|
Re: VNA and antenna matching : calculating reactance needed to match
B. Whitfield Griffith, Jr., "Radio-Electronic Transmission Fundamentals," McGraw-Hill, 1962 discusses the design of L-networks and provides graphical methods for such designs. ARRL used to sell this book, but I don't see it at the ARRL website now. Some used book dealers on the Web have it for under $10, although some of them want fantastic prices for it.
And there are many other text and reference books that provide design information for various networks.
"Octave for L-Networks," QEX, Mar/Apr, 2011 provides GNU Octave code for implementing Griffith's methods and points out that networks can often be oriented either way if the impedances to be matched are complex.
"More Octave for L-Networks," which I referenced earlier, provides Octave code that determines whether a network is reversible and, if so, provides element values for both orientations.
73,
Maynard
W6PAP
toggle quoted message
Show quoted text
On 8/25/23 00:14, Andrew (G1RVD G0Z) wrote: Thank you for all the superb replies.
I'm looking for a source that either has the relevant formulas or, where I can figure out the formulas. Ultimately to implement them in some form of auto tuner. (An intelligent version of an autotuner matching algorithm - if that makes any sense).
I'm investigating: GNU Octave code and TLW (i have a version of the ARRL Antenna Book).
I've used SimSmith, it's a great programme. (Effectively, I'm looking for the underlying math to automate the match.)
Kind regards,
Andrew
|
Thank you everyone for responding.
Last time I measured an AM broadcast antenna I used a Delta Electronics OIB-3 operating impedance bridge. Using the Delta OIB the measurement would have been completed in less than an hour with no complications. This time I am trying a VNA because I thought it would be more interesting. I will continue the parallel LC "trap" in-series approach. It is possible the trap could be designed into the ATU matching network and remain in the circuit.
If all goes well, eventually I will be using Nano VNA to measure the load "seen" by the transmitter output devices.
Goal is conveying the RF spectrum of the transmitter into radiated EM wave without significantly changing the RF spectrum in the intervening networks.
I am enjoying the VNA, but would eventually like to get accurate results in the field. Obviously, there is a difference between the lab or backyard antenna in a low ambient RF environment, and a broadcast antenna that has voltage from strong nearby RF sources.
|
On Thu, Aug 24, 2023 at 11:28 PM, Greg Strickland wrote:
I made a 40 dB Tee attenuator and did the open-short-load-save calibration at
the end of the attenuator
There is an 80dB round-trip loss through that attenuator; the analyzer does not have enough dynamic range to make sensible measurements through such a high loss. Best regards, Don
|
The other advantage is that some VNA's like the Anritsu Sitemaster and Cellmaster can overcome some interference. My Anritsu can send a coded signal that it can use to ignore the interference. If the interference is too strong the Anritsu will start beeping.? That's the difference between $40 and $$$$$$$$.
But the inexpensive VNA's can be a real handy hobby tool. They do have limited use on a crowded commercial site.
Joe, K1ike
toggle quoted message
Show quoted text
On 8/25/2023 1:54 AM, David McQuate wrote: These nanoVNAs have a smaller dynamic range than the (much more
expensive) professional VNAs.
|
Re: VNA and antenna matching : calculating reactance needed to match
Thank you for all the superb replies.
I'm looking for a source that either has the relevant formulas or, where I can figure out the formulas. Ultimately to implement them in some form of auto tuner. (An intelligent version of an autotuner matching algorithm - if that makes any sense).
I'm investigating: GNU Octave code and TLW (i have a version of the ARRL Antenna Book).
I've used SimSmith, it's a great programme. (Effectively, I'm looking for the underlying math to automate the match.)
Kind regards,
Andrew
|
Re: NanoVna H4 Showing blank Screen after updating to latest 1.2.20 firmware
On Fri, Aug 25, 2023 at 12:18 AM, Dragan Milivojevic wrote:
You probably flashed the wrong firmware variant, you used H version
instead of H4 version.
Yes I think that's the problem because I did exactly that with my H4 and got the same result. I flashed it again with the correct variant and it recovered. It's easy to miss the "4" when selecting the file to download. -- Mike
|
These nanoVNAs have a smaller dynamic range than the (much more
expensive) professional VNAs. With a 40 dB attenuator, the signal from
the VNA is attenuated by 40 dB on its way from the analyzer to the DUT
(or calibration standards), and any reflected signal is again attenuated
by 40 dB on its way back to the analyzer. When measuring a short or
open (~100% reflection) the analyzer sees a signal 80 dB weaker than the
outgoing signal. When the Load cal standard is measured, the received
signal will be even lower. The nanoVNA H4 has a dynamic range of 70 dB
(at lower frequencies). Thus, during calibration, it is measuring its
noise floor. There's nothing there. It doesn't matter what you connect
as a calibration standard. All you get is noise. The calibration is
quite useless. Even a 30 dB attenuator is going to give you a very poor
calibration. 20 dB -- maybe useful. 10 dB -- probably decent.
So, you're right, the VNA is not able to "see" beyond the 40 dB
attenuator.
David
toggle quoted message
Show quoted text
On 2023-08-24 20:28, Greg Strickland wrote: Thank you David. I made a 40 dB Tee attenuator and did the open-short-load-save calibration at the end of the attenuator not
connected to CH0. Maybe I did it incorrectly, because Nano VNA appeared to be measuring the shunt resistance of the Tee attenuator, and was not able to "see" beyond that. It appeared device under test side impedance was swamped out by the 3.2 ohm shunt resistor in the Tee attenuator.
Will try again tomorrow.
|
Thank you David. I made a 40 dB Tee attenuator and did the open-short-load-save calibration at the end of the attenuator not
connected to CH0. Maybe I did it incorrectly, because Nano VNA appeared to be measuring the shunt resistance of the Tee attenuator, and was not able to "see" beyond that. It appeared device under test side impedance was swamped out by the 3.2 ohm shunt resistor in the Tee attenuator.
Will try again tomorrow.
|
Port CH0 ideally presents a 50 ohm resistive impedance to anything
connected to it. If an attenuator is needed, it should have a Pi or Tee
form, so that a 50 ohm resistive impedance is still presented to
something connected to it. If a resistor is connected in series with
CH0, the impedance seen looking into it and CH0 would be 50 + R, clearly
a mismatch to the cal standards. If a Pi or Tee attenuator is used, do
the open-short-load-save calibration at the end of the attenuator not
connected to CH0. This way the calibbration will "take into account"
the presence of the attenuator.
David
toggle quoted message
Show quoted text
On 2023-08-24 18:35, Greg Strickland wrote: If I may ask a follow up questionsabout Nano VNA-H.
1. Is port CH0 input a 50 ohm resistor load?
2. If #1 is correct, would proper attenuator implementation for the port be a series resistor, to form a L pad with the internal resistor?
3. If #2 is correct, would the open-short-load-done-save calibration be done with the measurement plane on the source end of the L pad?
4. If #3 is correct, would scale or another adjustment need to be done to account for a pad between Nano VNA and antenna under test.
btw- the parallel LC "trap" worked at the antenna tower and it reduced the nearby RF source down to less than 100 millivolts. Moving the measurement plane to "cancel out" the trap did not work for me. I was able to obtain presumably credible results by removing the measured impedance of the trap by itself from the measured impedance of the trap and antenna. What concerns me is DC resistance of the trap inductor in series with an antenna that might be 27 ohms at 55 ohms of capacitive or inductive reactance at the operating frequency.
Good news is I have not "blown up" the Nano VNA yet.
Thoughts or suggestions are appreciated. Thank you
|
Bob Watson
Roger Need
Donald S Brant Jr