开云体育

That's interesting. I didn't think of using return loss instead of VSWR. I was thinking my VSWR would be much improved at the end of a lossy cable and I couldn't think of a way to correct for it. But return loss is the way. Simple math actually.
But I wonder how calibrating at the end of the lossy coax will affect my sweep. Do I lose dynamic range?
Albert

Hi Albert
You can cal at the end of the coax, but you don't have to.
Lets say the coax has about 2 dB of loss.2 dB of the signal is lost going up, and 2 dBis lost coming back down.Thus the antenna is 4 dB worse than you though it was.
In dB a measured Return Loss of 20 dB RL was really 16 dB RL.
In SWR ....... the math takes a while! hihi.? ? ??
In either case, you are making sure the antenna is on frequency, the coax shouldn't change that.?(Yes, I know a few exceptions)? and if you adjust the antenna for best RL/SWR, then you have adjusted the antenna for best RL/SWR.? It's as good as you are going to get!
Good luck with your project.? ?Kent WA5VJB? ?Antenna Editor CQ Magazine

I had the same problem, which appears when making the calibration and SHORT probe does no electric contact is made with the CH0 connector.
Solving was by tightening the warm wire in the CH0 connector.

I want to sweep my 2 meter antenna from the end of a 15 to 18 foot coax. This is because I need to get the antenna out in the clear away from nearby objects to get a good indication of the actual VSWR. The antenna has an SMA connector. I have two pieces of coax I can use. One is 15' of LMR-100 with SMA connectors on both ends. The other is an 18' piece of RG-8X with PL-259 connectors at both ends. In either case I can move the plane of calibration to the end of the coax where the antenna connects.
So I have two questions for those who have a lot of VNA experience in this group:
1) Since LMR-100 is lossy at 2m frequencies, will I lose much precision in the sweep by calibrating at the end of the coax?
2) Am I better off using RG-8X knowing that I will need to adapt the PL-259 to SMA at both ends? I understand that PL-259 connectors are not so good at VHF and adding adapters will only make them worse.
Which is the better alternative?

Albert KK7XO

On 19/09/2021 12:56, DougVL wrote:

FYI, when you're looking at your own posting, underneath are 3 links that say
Reply Like More
If you click on the "More" link, you can choose to edit your post (and correct typing mistakes). I have had to do that a few times.
-- Doug, K8RFT

Although as an e-mail subscriber it can be frustrating to see a whole string of messages all but one of which say "Edited message" (or whatever it is). Of course, it's a very useful feature at times, but I would rather folk review their message /before/ posting rather than after!

(I'm as guilty as anyone at making mistakes, of course!)

73,
David GM8ARV
--
SatSignal Software - Quality software for you
Web:
Email: david-taylor@...
Twitter: @gm8arv

FYI, when you're looking at your own posting, underneath are 3 links that say

Reply Like More

If you click on the "More" link, you can choose to edit your post (and correct typing mistakes). I have had to do that a few times.
--
Doug, K8RFT

Buy, I meant *buy* an original Hugen NanoVNA!

Update. I did return the non Hugen "NanoVNA". Call them fakes, counterfeit, clones or what you will. The clone's branding and packaging is clearly intended to deceive the purchaser.
I have now received a genuine unit with the correct QR code and the specified mcu. I now realise how poor the copy was, the fit of the case, lower quality switches, screen, out of date firmware and some unmarked components of dubious origin.
Do the right thing and yourself a favour. But an original, real product.

I think these RF test boards help, trains you on results, if you have set the vnanano up and calibrated correctly, getting the prescribed results gives us ou some confidence in the nano.

Interestingly, inside KEIL, STM32F302CB and STM32F302CC use the same memory description file. So in fact STM32F302CB and STM32F302CC may use the same Die, just marked as different models when packaging. You can try to refresh the latest firmware, then save the calibration data to SAVE6, then Recall 6, if the data can be recalled normally, then STM32F302CB and STM32F302CC have the same flash.

A cheap LCR meter is also a good option for measuring variable capacitors.
The value of most types of capacitors does not vary wildly with measurement
frequency. The value of inductors does, which is why the LCR meter is not
always a good option for measuring them.

On Thu, Sep 16, 2021 at 1:43 PM Roger Need via groups.io <sailtamarack=
[email protected]> wrote:

I did some measurements on a variable capacitor in my junkbox. It was
from an old MW receiver.

I measured the closed position on a calibrated DE-5000 LCR meter and it
was 367 pF.
In the open position it measured 8.5 pF

I set my NanoVNA-H4 to a stimulus range of 1 to 10 MHz. I used a RG-58
cable with alligator clips on the end. This type of test jig is OK at
lower HF frequencies but stray capacitance/inductance affects the results
as you go much higher.

A SOL Calibration was done using the alligator clips and a 50 ohm carbon
resistor (non inductive). When connecting the alligator clips to the
variable capacitor make sure the ground alligator clip goes to the frame of
the variable capacitor.

The NanoVNA gets the most accurate reading when |Z| is 50 ohms. In the
fully closed position the capacitor measured 365 pF when the magnitude of Z
was 50 ohms (bottom of Smith Chart). Only 2 pF difference compared to the
DE-5000. Connecting the the NanoVNA to a PC and using the NanoVNA app the
measured capacitance from 1 to 10 MHz was plotted. The measured
capacitance ranged from 371 pF to 363 pF which is quite close.

Next the variable capacitor was set to the open position. It is not
possible to measure at |Z| = 50 ohms because this occurs around 375 MHz..
However the NanoVNA-H4 still gives a reasonable estimate ranging from 8 to
8.3 pF over the 1 to 10 MHz. range which is close to the 8.5 pF measured on
the LCR meter.

Experimental results are attached.

Roger

I did some measurements on a variable capacitor in my junkbox. It was from an old MW receiver.

I measured the closed position on a calibrated DE-5000 LCR meter and it was 367 pF.
In the open position it measured 8.5 pF

I set my NanoVNA-H4 to a stimulus range of 1 to 10 MHz. I used a RG-58 cable with alligator clips on the end. This type of test jig is OK at lower HF frequencies but stray capacitance/inductance affects the results as you go much higher.

A SOL Calibration was done using the alligator clips and a 50 ohm carbon resistor (non inductive). When connecting the alligator clips to the variable capacitor make sure the ground alligator clip goes to the frame of the variable capacitor.

The NanoVNA gets the most accurate reading when |Z| is 50 ohms. In the fully closed position the capacitor measured 365 pF when the magnitude of Z was 50 ohms (bottom of Smith Chart). Only 2 pF difference compared to the DE-5000. Connecting the the NanoVNA to a PC and using the NanoVNA app the measured capacitance from 1 to 10 MHz was plotted. The measured capacitance ranged from 371 pF to 363 pF which is quite close.

Next the variable capacitor was set to the open position. It is not possible to measure at |Z| = 50 ohms because this occurs around 375 MHz.. However the NanoVNA-H4 still gives a reasonable estimate ranging from 8 to 8.3 pF over the 1 to 10 MHz. range which is close to the 8.5 pF measured on the LCR meter.

Experimental results are attached.

Roger

I have gotten good measurements on a variable capacitor doing it exactly the same way I measure an inductor. I use only CH0 and measure reflection (S11). I connect the 2 conductors from CH0 to either end of the capacitor. I get a pretty stable result at any frequency from 100 kHz to several MHz. My capacitor (14 - 387 pF) hits self-resonance up around 21 MHz, and then shows inductance, when turned to the low-C side. I get the C value from the CH0 Smith chart.

If you have successfully measured numerous inductors, you are also set to
measure the variable capacitors. They are both a source port (S11)
measurement. Connect the same as you did for the inductors. Instead of
plotting on the top of the Smith Chart, it will plot on the lower half of
the Smith Chart. Use the cursors to measure at a specific frequency.

Dave - W?LEV

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

Hugen. Thanks for your thoughts. Your point about the demands of new firmware is important. I wish it were easier to get hold of your original product here in the UK.

Larry, thanks for your very detailed responses. I will try to upload two phots of the PCB of my unit (apologies for the phone camera quality). People can use them to make informed choices about their own purchases. Looks like the MCU part is STM32F302 CBT6 but very blurry print.

Hello
One walk around.
Get a known inductor.
Put unknown capacitor in parallell
See what frequency it resonate

C = 1/(L×fxf)

Regards


El jue., 16 de septiembre de 2021 7:38 a. m., Joe WB9SBD <nss@...>
别蝉肠谤颈产颈ó:

OK,
I wandered the group site for a long time and went to the HOW TO on the wiki site, Nothing helped me.

I have the steps for measuring an inductor down solid. And have probably done a hundred by now.

But I now want to measure a variable cap. to learn it's max and min values.

But not having any success on how to do that.

Anyone?

Joe WB9SBD

Since the STM32F303CCT6 is currently difficult to order, some clones may be replaced with some inexpensive processors.
Earlier versions of the NanoVNA-H4 firmware, which allows only 101 scan points and uses relatively little RAM in order to remain compatible with the NanoVNA-H, can also run on the STM32F303C8T6 and STM32F303CBT6 as well as the STM32F302C8T6 and STM32F302CCT6.
The new firmware takes full advantage of the STM32F303CCT6 RAM and ROM with more scan points and save options with the help of DiSlord, so it can only run on the STM32F303CCT6 and STM32F302CCT6 which have larger memory.

Stephen
I found the functional block diagrams of both F302 and F303 series processors, if you're interested.
Again, a lot of the chip's internal functionality is not used. The F303 was originally chosen due to its ability to drive a 4 inch display which the F072 in the original design could not. The STM32F303CCT was also a pin for pin drop-in replacement for the F072 and available on an STM32 experimenter board.
More info: /g/nanovna-users/topic/61079748#9206

Try installing DiSlord's latest H4 version of FW - 1.0.69 - on the unit, calibrate it and check out all the functions including screen brightness control and battery charge sense. I don't think you'll have any issues - but again, if new software functionality is added to the H4 in the future that depends on something that the F303 has and the F302 does not (like memory), then return it.

BTW - what's the complete processor part number, including suffix? A real H4 uses an STM32F303CCT6. If you have a GD32F303 processor, return it. With more info, I can update the forum's Wiki about this model and list any caveats associated with the device.

Your call.