Measuring balanced antenna feeds with the nano is easy if it is not connected to anything and well decoupled from "ground". It would be nice to remove the influence of connecting a USB cable to a PC so NanoVNA Saver can be used. I see two possible solutions:

- a wideband RF 1:1 transformer on the S11 port
- USB isolation of the RF path to ground

The nice thing about USB isolation is that it should be possible to make it generically broadband and there are lots of isolators out there cheap. As one might expect, none of the inexpensive products (not fiber optic) say anything about parasitic capacitance on the ground side. Does anyone on this list have good/bad experiences with specific brands and models that show low coupling in the 3 to 30 MHz region?

failing that, I'm thinking the RF transformer may be the best way to go.

All thoughts welcomed!

Brent, AB1LF

Use a laptop, use a BT to serial or WiFi to serial adaptor.


On Mon, 2 Dec 2024 at 20:26, Brent DeWitt via groups.io
<bdewitt@...> wrote:

Measuring balanced antenna feeds with the nano is easy if it is not connected to anything and well decoupled from "ground". It would be nice to remove the influence of connecting a USB cable to a PC so NanoVNA Saver can be used. I see two possible solutions:

- a wideband RF 1:1 transformer on the S11 port
- USB isolation of the RF path to ground

The nice thing about USB isolation is that it should be possible to make it generically broadband and there are lots of isolators out there cheap. As one might expect, none of the inexpensive products (not fiber optic) say anything about parasitic capacitance on the ground side. Does anyone on this list have good/bad experiences with specific brands and models that show low coupling in the 3 to 30 MHz region?

failing that, I'm thinking the RF transformer may be the best way to go.

All thoughts welcomed!

Brent, AB1LF

Exist USB isolators on ADUM3160 try search it
https://aliexpress.ru/item/1005002959825296.html?sku_id=12000022964317764&spm=a2g2w.productlist.search_results.0.48497f1aAGtFNW

having just been searching for a "copper-less" USB fiber optic connection (precisely for this sort of antenna testing application at work): they aren't easy to come by, and are kind of expensive ($900). You'll see lots of inexpensive AOC (Active Optical Cable) USB over fiber cables, but they have a couple copper conductors to provide power to the "far" end of the cable.



Optical serial port, or optical Ethernet are *much* easier to come by.

I've used a beaglebone green wireless to get remote USB access to a device, and also a Rpi. It's kind of clunky to be sure, but it does work, because you're doing two hops - your PC/Mac to the remote processor via IP network (WiFi) and then through the device.
And you've got to get power to the processor (battery works).

Since the NanoVNA has serial access (via a couple of pads on the board) that might be the easiest way.
I'd assume you'd use some sort of program (I use NanoVNA-Saver, but there's various other apps) to control the NanoVNA, so it's not like you need to see the screen.

You can use an old android smartphone connected to your nanovna via USB, running the TCPUART app to do remote access via WiFi.
Search my old posts for TCPUART for details.

USB 2.0 is hard to do over fiber, thanks to it beimg pseudo differential. So both ends of the fiber need smarts for it to work.

Onlybreally seen those Analog Devices made USB isolators on the market and they are for electrical safety/galvanic isolation and sometimes can imtroduce noise on their own.

to surpress ground current just turn the usb cable a few turns around a ferrite core (bigger donut shape)
that should work well enough
dg9bfc sigi

Am 03.12.2024 07:55 schrieb "Ismo Väänänen OH2FTG via groups.io" <ismo.vaananen@...>:

How about this: https://www.adafruit.com/product/2107

Might do the trick.

That adafruit.com isolator appears to only support USB 1.0 speeds.


--
--Bryon, NF6M

How fast do you need?

I have built several of these using ADI ADUM3166's and ADUM4166's. I've had zero problem with noise. But, you have to use good engineering practice in their design and execution.

The effective primary to secondary capacitance in an actual circuit is something less than 10 pF. Is that low enough? Depends, I guess, on what you're trying to do and what the circumstances are. 10 pF is around 500 Ohms at 30 MHz. I've built and measured some common mode chokes that had 20x that amount at 30 MHz.

Clarke

On Mon, Dec 2, 2024 at 11:26 AM, Brent DeWitt wrote:

failing that, I'm thinking the RF transformer may be the best way to go.

Brent, check out the Coilcraft SWB1010, SWB2010, and SWB3010 wideband RF transformers. They are cheap and readily available. Their response is very flat over HF. Depending on the accuracy you need, you may be able to get by without calibrating the VNA through the transformer, which would simplify use. Coilcraft does not specify the primary-to-secondary capacitance. If you get the model with a secondary center tap, when necessary you can try grounding it to create a voltage balun. This may work better than an isolated winding if the interwinding capacitive reactance is too low at higher frequencies.

Brian

Just a comment re the ADUM3160. I purchased one of these on Amazon just a
week ago and had to return it because it made my usb cable unusable to my
SDRuno. Basic usb cable worked. With the ADUM3160 in line windows issued a
failure.

I returned it. I'm not sure if it was a one time defective unit or not.

Just an FYI.

Fred
N4CLA

On Mon, Dec 2, 2024, 16:31 DiSlord via groups.io <dislordlive=
gmail.com@groups.io> wrote:

Exist USB isolators on ADUM3160 try search it

https://aliexpress.ru/item/1005002959825296.html?sku_id=12000022964317764&spm=a2g2w.productlist.search_results.0.48497f1aAGtFNW

All the Analog Devices isolator chips seem to specify an input to output capacitance in the region of 2pF, which is quite good, and comparable to the Coilcraft broadband RF transformers.

Using an isolator and adding some additional clip-on ferrites to the USB cable, would improve the HF common mode isolation still further.

--
Regards,

Martin - G8JNJ

yeah, but you still have a conductor in the near field of the antenna under test. It would solve the "measuring at the end of a balanced transmission line" thing. And, of course, galvanic isolation is handy in a lot of test scenarios.

I did a study on this. There are USB isolators, but they are not cheap, and most do not reach USB 3.0 speeds. The easiest way is to make a separation of the ground (case) connection with a grounded USB input and an ungrounded USB output. But easier is to use a separate USB HUB with its own power supply.

I have had very good success using wideband transformers, as Brian suggests.

If you use the configuration shown in Figure 2 here: https://www.minicircuits.com/appdoc/AN20-001.html you can get good results out to several hundred MHz. I use something like this in series with the CATV box to minimize the hum from the power leakage on the CATV feed. In that case, it's a 75 Ohm application and it was pretty flat to 1 GHz. Using a balun at the input and another at the output helped. The transformers shown in the attachment worked pretty well, with one "turned sideways" to use as a conventional flux coupled transformer.

Clarke

I would like to emphasize on the point Jim made: As long as you have an USB cable running into the antenna's near field, it will affect the measurement, even if an USB isolator or an RF isolation transformer is used. Also any isolator or transformer has capacitance, and even if it's just a few pF, this might be enough to cause an unacceptable measurement error in some cases.

So I think that the best thing to do is to connect the NanoVNA directly to the antenna, powered by its battery, and use any small battery-powered device to translate between USB and a fiber optic cable. Then run that fiber cable down outside the near field of the antenna, and interface it to the readout software.

As an alternative a radio link could be used. On Wifi or similar.

If suitable low-cost converters can't be found, it should be simple enough to write a program for an Arduino, to convert between USB and fiber, or use an ESP32 with its built-in WiFi. In fact, given how easy it is to use WiFi with the ESP32, it looks like the way to go!

With an ESP32 plus battery attached to the NanoVNA, there is just the capacitance of this small pack to the surroundings, and that should be really negligible on HF. But on VHF and higher even this is enough to significantly affect the measurements.


Manfred

There are plenty of serial to BT or WiFi modules available that work
with NanoVNA practically out of the box. Why do people insist on
reinventing the wheel?


On Sat, 7 Dec 2024 at 14:39, Manfred Mornhinweg via groups.io
<manfred@...> wrote:

I would like to emphasize on the point Jim made: As long as you have an USB cable running into the antenna's near field, it will affect the measurement, even if an USB isolator or an RF isolation transformer is used. Also any isolator or transformer has capacitance, and even if it's just a few pF, this might be enough to cause an unacceptable measurement error in some cases.

So I think that the best thing to do is to connect the NanoVNA directly to the antenna, powered by its battery, and use any small battery-powered device to translate between USB and a fiber optic cable. Then run that fiber cable down outside the near field of the antenna, and interface it to the readout software.

As an alternative a radio link could be used. On Wifi or similar.

If suitable low-cost converters can't be found, it should be simple enough to write a program for an Arduino, to convert between USB and fiber, or use an ESP32 with its built-in WiFi. In fact, given how easy it is to use WiFi with the ESP32, it looks like the way to go!

With an ESP32 plus battery attached to the NanoVNA, there is just the capacitance of this small pack to the surroundings, and that should be really negligible on HF. But on VHF and higher even this is enough to significantly affect the measurements.


Manfred

QUOTE (from Dragen): Why do people insist on reinventing the wheel?

WHY: Because they don't take the time to use the search engines. They're
lazy and just want someone else to put out the small required effort.

Dave - WØLEV

On Sat, Dec 7, 2024 at 6:16 PM Dragan Milivojevic via groups.io
<d.milivojevic@...> wrote:

There are plenty of serial to BT or WiFi modules available that work
with NanoVNA practically out of the box. Why do people insist on
reinventing the wheel?


On Sat, 7 Dec 2024 at 14:39, Manfred Mornhinweg via groups.io
<manfred@...> wrote:

I would like to emphasize on the point Jim made: As long as you have an

USB cable running into the antenna's near field, it will affect the
measurement, even if an USB isolator or an RF isolation transformer is
used. Also any isolator or transformer has capacitance, and even if it's
just a few pF, this might be enough to cause an unacceptable measurement
error in some cases.

So I think that the best thing to do is to connect the NanoVNA directly

to the antenna, powered by its battery, and use any small battery-powered
device to translate between USB and a fiber optic cable. Then run that
fiber cable down outside the near field of the antenna, and interface it to
the readout software.

As an alternative a radio link could be used. On Wifi or similar.

If suitable low-cost converters can't be found, it should be simple

enough to write a program for an Arduino, to convert between USB and fiber,
or use an ESP32 with its built-in WiFi. In fact, given how easy it is to
use WiFi with the ESP32, it looks like the way to go!

With an ESP32 plus battery attached to the NanoVNA, there is just the

capacitance of this small pack to the surroundings, and that should be
really negligible on HF. But on VHF and higher even this is enough to
significantly affect the measurements.

Manfred

--

*Dave - WØLEV*


--
Dave - WØLEV