开云体育

On Thu, Sep 19, 2019 at 06:53 PM, Reginald Beardsley wrote:

Most of the problem is because the SMA hex is so hard to get a grip on.

Reg, for this reason I picked up some "SMA Nuts" on eBay after I discovered I had the same issue with my NanoVNA. Now each of my three SMA standards wears a collar.

Below are a couple of eBay links to sellers in the States.





If they interest you, maybe there's someone in the UK selling the same? Try searching for "drone SMA nuts"

So far, they are great! (I used the first vendor, above). As for durability, we will see...

- Jeff, k6jca

slide 11

Simple and cheap: Back to back diodes
? Protection not dependent upon configuration
? Diode type is not critical (except, don’t use PIN diodes)
? Limited to low input power levels => receive only applications
? +30 dBm = 1 watt max (when using ? watt diodes)
? If either diode fails open => receiver front end not protected
? Spurious signals in receiver can be a problem
? Some mfgs offer choices on spurious levels (DX Engineering RG-5000 series)

So the question is, what precautions can be taken to protect the NanoVNA ?

Prayer is not an option ;-)

73 de Andy

The CH0 input of the nanoVNA is basically an attenuator connecting to a SA612 mixer without any additional protection.

Ps. I included the ssh key part as I could not "git submodule update
--init --recursive" without a one.

Interesting! I had the same issues, and instead finally resolved it by

editing the .submodules file:

- replace the 3rd line with "url =
- add a new line to the end of the file

Finally type "git submodule sync", and then "git submodule update --init
--recursive" works correctly.

I prefer your fix!

Rgds,
Dave

Thank you very much for your post, John. Posts like yours certainly give me
motivation to continue working on the application, even when I find issues
that are difficult to solve in an elegant manner. :-)

--
Rune / 5Q5R

On Fri, 20 Sep 2019 at 07:48, John Nightingale via Groups.Io <if455kc=
[email protected]> wrote:

This old man is just swept away by what you have been producing, Rune!

This very ordinary user has kept quiet about this superb application

and left comments to the savants. Now, though, with this 0.0.10 release,
I just have to convey a massive "thank you" on behalf of all of us
lesser users, ham radio operators particularly. We hams now have
something we could only have dreamed of until a couple of months ago;
it amounts to a $10k instrument.

A Linux system is in use here. The application works perfectly in

every detail tested so far. A special "thank you", too, for troubling to
furnish something better for old eyes than that minute font, Rune!

Mention will be made of something that defeated this user initially

because other less alert users may also bump into it. The reason for the
following has not been investigated but it will be something dim witted
done at this end because of being compelled to use the blunt instrument
of a 74 year old brain! The "SWEEP" option was found, repeatedly, grayed
out when trying to perform calibration. What was going on proved to be
very simple. The initial port assignment had been "/dev/ttyACM1". The
application had found the correct port by itself, as designed. What was
going on proved to be, somehow, reassignment to "/dev/ttyACM0". Of
course no sweep could be done with the application looking into a vacant
port. Manually forcing back the port to "/dev/ttyACM1" fixes the issue.

Many many thanks, Rune, on behalf of all the hams who now have a

truly superb tool.

John

at radio station VE7AOV

+++++



On 2019-09-18 1:36 p.m., Rune Broberg wrote:

I just released 0.0.10:


It's not the most exciting release, but it offers some quality of life
improvements, such as the ability to choose the font size (particularly
useful for Linux users, whose default is a massive 11 pt font).

It also adds debug logging: -d to get log messages to the terminal, or -D
filename.txt to log to a file. Useful if you see crashes!

Additionally, it now supports importing magnitude/angle touchstone files,
and there's been a number of little bugfixes.

As ever, I look forward to hearing what bugs you find, and what new
features you want! :-)

--

Hi David.

I've updated the post with instructions, on the opposing side of the SMA connectors CH0 and CH1, there are two pins on the PCB labelled VDD and BOOT0.
They need to be shorted with the powered off state, then with those shorted power on the device.

Sorry for the confusion.

Ps. I included the ssh key part as I could not "git submodule update --init --recursive" without a one.

Kind Regards
Ohan Smit
ZS1SCI

On Fri, 20 Sep 2019 at 07:09, Rune Broberg <mihtjel@...> wrote:

Hi Reg,
Ohan made a guide.

This sentence

“Flash by setting the device to DFU boot, either in software or by short of
pins, then:”

could usefully be expanded. I don’t know how to do either

Dave
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales.
Company number 08914892.
Registered office:
Stokes Hall Lodge,
Burnham Rd,
Althorne,
Chelmsford,
Essex,
CM3 6DT,
United Kingdom

Hi Reg,
Ohan made a guide. I'm not certain the SSH keys are required, but it's all
there:



--
Rune / 5Q5R

On Fri, 20 Sep 2019, 00:35 Reginald Beardsley via Groups.Io, <pulaskite=
[email protected]> wrote:

If you are modifying the FW *please* maintain a file in the Firmware
folder with release notes and version information along with the location
of the source repository.

If you make comments such as "TDR is now in the FW" *please* state which
version. This has turned into an Abbott and Costello skit.

I had asked about a post about instructions for building the FW on
*Linux*, not Windows. The search feature doesn't produce anything useful.
The maddening part is I'm pretty sure I thanked the person who did that,
but I cannot even find a list of my own posts.

I've downloaded both the ST development tools and the Gnu tools onto a
Debian 9.3 system, but I'm now so confused about who has done what that I'm
reluctant to actually do *anything*. I don't want to reinvent the wheel.
and unsure about whether the version that came on my unit is buggy and
needs fixing to use the command line console.

I'd like to see if I can improve the dynamic range at higher frequencies
with some DSP. Depending upon the nature of the situation, I think it
might be possible to pick up 20-30 dB improvement.

But at the moment I'm completely befuddled by "Who's on First".

Also, if you're quoting and commenting inline *please* delimit the quotes
from your own comments. There have been a number of posts I gave up trying
to figure out. They looked as if there were 3 or more authors.

Reg

Greetings Dr. David,

Thank you for flagging my less than clear and possibly misleading post! I did not intend to infer that my DIY 50 Ohm load has the quality of a 53 dB return loss instrumentation standard! After reviewing the wording I understand how one could get that impression. In an earlier post (#2562) in this thread I had previously described measuring my DIY BNC 50 Ohm load using the SMA loads supplied with my nanoVNA for the calibration. I provided a screen shot and S1P file showing results. Amazingly the RL match to the SMA references measured in the 50s up to 450 MHz. At 890 MHz it dropped to about 30dB. I should have clearly qualified the relative reference and referred to that earlier post. I was trying to provide a foundation as to why I think that BNC connectors can be used for routine Z measurement at frequencies VHF and lower.

My goodness my DIY loads are not even shielded and could possibly be used as lossy antennas at SHF, (:/)). The reality is that unlike yourself I don't have any means to properly compare my loads to any laboratory level standards. And as you point out it would be extremely difficult to establish a 53 dB reference. Sounds like your well equipped for the high end of the microwave spectrum!

Amazingly at 450 MHz and below (according to my nanoVNA) my BNC load including the BNC/SMA adapter matches the SMA load calibration with a return loss of around 50dB up to 450 MHz. I decided to see if I could repeat the test and have attached the files. This time 147MHz showed a RL of 48dB, 450MHz 53dB and 890MHz 47db (usually more like 30dB at 890MHz). I am amazed that I can repeat this test this well considering how critical a 50dB match is. It wouldn't surprise me to see a big reduction in these numbers with just a minor change of temperature.


I have not experimented with the nanoVNA S21 capabilities yet. I certainly agree about the need for 100 dB dynamic range for cavity tuning. A spectrum analyzer with a tracking generator works nicely but I imagine there are expensive VNAs that will do that job even better.


Best Regards,
Tom
VA7TA

This old man is just swept away by what you have been producing, Rune!

This very ordinary user has kept quiet about this superb application

and left comments to the savants. Now, though, with this 0.0.10 release,
I just have to convey a massive "thank you" on behalf of all of us
lesser users, ham radio operators particularly. We hams now have
something we could only have dreamed of until a couple of months ago;?
it amounts to a $10k instrument.

A Linux system is in use here. The application works perfectly in

every detail tested so far. A special "thank you", too, for troubling to
furnish something better for old eyes than that minute font, Rune!

Mention will be made of something that defeated this user initially

because other less alert users may also bump into it. The reason for the
following has not been investigated but it will be something dim witted
done at this end because of being compelled to use the blunt instrument
of a 74 year old brain! The "SWEEP" option was found, repeatedly, grayed
out when trying to perform calibration. What was going on proved to be
very simple. The initial port assignment had been "/dev/ttyACM1". The
application had found the correct port by itself, as designed. What was
going on proved to be, somehow, reassignment to "/dev/ttyACM0". Of
course no sweep could be done with the application looking into a vacant
port. Manually forcing back the port to "/dev/ttyACM1" fixes the issue.

Many many thanks, Rune, on behalf of all? the hams who now have a

truly superb tool.

John

at radio station VE7AOV

+++++?



On 2019-09-18 1:36 p.m., Rune Broberg wrote:

I just released 0.0.10:


It's not the most exciting release, but it offers some quality of life
improvements, such as the ability to choose the font size (particularly
useful for Linux users, whose default is a massive 11 pt font).

It also adds debug logging: -d to get log messages to the terminal, or -D
filename.txt to log to a file. Useful if you see crashes!

Additionally, it now supports importing magnitude/angle touchstone files,
and there's been a number of little bugfixes.

As ever, I look forward to hearing what bugs you find, and what new
features you want! :-)

--

When a VNA is connected to an antenna (often they are outdoors) there are some potential issues with electrostatic discharge (ESD.) In dry areas of the world ESD can be an issue even when working with objects that are purely indoors. I have seen issues when working on tables indoors. Other VNA devices are known to have issues. Does anyone know how sensitive the RF inputs to the nanovna are with regard to ESD? Is there any ESD protection built into the RF connections on the nanovna?
--
Bryan, WA5VAH

Reginald and Dr. Dave

Since you both completely missed the point of my post I must assume the fault for that is mine in the presentation.

My post did not focus on what parameters were being measured or on the accuracy of the value measured for any of the chosen parameters. The point of the post was the consistency of measurement among three versions of Nanovna and two independent devices.

Warren Allgyer
WA8TOD

The wikipedia definition of "return loss" is rather absurd. If you have a *perfect* match and *no* power is reflected, the "return loss" is infinite. Of course it *is* wikipedia.

I'm not about to try to change the EE community lexicon, but it seems to me reflected/incident makes a great deal more sense numerically. It also leads to what I'd expect. "Loss" as a negative quantity when expressed in dB.

Have Fun!
Reg

FWIW a bit of TDR porn on "finger tight" vs torqued. I think it should be noted that in this example, the ringing is at over 5 GHz, so it really doesn't matter for the current version of the nanoVNA.



This is a little more extreme than most cases, but I often am watching the 11801 while I tighten the connections and see similar results as I switch from the initial finger tight to torqued. Most of the problem is because the SMA hex is so hard to get a grip on. I commonly see such effects disappear while tightening N connections.with my fingers.

I replaced both of the SMA connectors on my nanoVNA with the bulkhead type, so with the connector clamped to a case, the PCB connections won't be strained.

I think a knurled round wrench about the diameter of an N-M would be entirely adequate until you got well above 10 GHz. Perhaps some enterprising merchant will offer some on ebay. Simple to make. Knurl a piece of 2 cm bar, part off, and mill a slot to fit the flats. About $1200 US in tooling and with practice probably 5 minutes to make. I'd buy a well made pair for $10-12 delivered in a heartbeat just for the convenience of not making them myself. So a person in the US should be able to make over $30/hr making and selling them. Pretty good business for someone who is retired and is only looking for play money to buy more toys. I've got the machine tools, so I'll make a pair eventually if no one decides to make them for sale.

Have Fun!
Reg

On Tue, 3 Sep 2019 at 00:11, Warren Allgyer <allgyer@...> wrote:

Roger

I have seen the article. Return loss measurements of 40 dB or greater turn
to mush on most instruments that cost less than $10,000. No surprise the
same would apply to the NanoVNA. But a couple of points:

I measured an SMA load to have a return loss of *56 dB* to 1.3 GHz. I am
fairly confident that the return loss is at least *37 dB*.

The equipment I used was an HP 8753ES VNA with a cheap (only $13,464)
85052B 3.5 mm calibration kit.



The loads in the cal kit I believe have a return loss specification of at
least 48 dB to 2 GHz. About 10 dB better than the DUT is needed for an
uncertainty of about 1 dB.

48-10-1= 37 dB.

Some of the figures people believe that they are achieving are just wrong.

Dave.


--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales.
Company number 08914892.
Registered office:
Stokes Hall Lodge,
Burnham Rd,
Althorne,
Chelmsford,
Essex,
CM3 6DT,
United Kingdom

When I am testing amplifiers, using one of my network analyzers, I normally start at -30 dBm output from the signal source, into the device under test. You don't to come anywhere close to saturation, or the results will not be accurate.
A good indication that a linear amplifier is going into saturation can be determined by watching the bias current. As you raise the input level, the current will start to rise as the device saturates; however you need a resolution of at least 10 ma on the ammeter. The inexpensive Chinese bench power supplies, with digital meters are very good for amplifier testing. For the most part, an analog meter is useless.
Stuart K6YAZLos Angeles, USA

Tks for mentioning that Dave. I forgot to add that besides the amplifier gain and max input, the 1 dB compression point should be taken into account when determining the system budget.

On Thu, 19 Sep 2019 at 03:41, w0lwa <trberger@...> wrote:

Rod

Have a look at

and start reading on page 24 of

Whilst his methods of making open short and loads is not had, his
verification procedure looks seriously flawed to me. Unless I am mistaken,
he performs a calibration using his standards, then believes that they are
working well as when he remeasured them, they give the same results. He
could almost use any standards and that would happen - even a wire wound
resistor!

Perhaps I have misunderstood it, but it seems very flawed to me.

Dave


--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales.
Company number 08914892.
Registered office:
Stokes Hall Lodge,
Burnham Rd,
Althorne,
Chelmsford,
Essex,
CM3 6DT,
United Kingdom

On Fri, 20 Sep 2019 at 00:12, hwalker <herbwalker2476@...> wrote:

All my low noise RF preamps have a maximum input of 0 dbm and can handle
the maximum output power of the nanoVNA without an attenuator.

Whilst an LNA can probably take 0 dBm without damage, it would be driven
well beyond its linear region.

I just took as quick example, one of the Minicircuit amplifiers I have here



1 dB compression point +5 dBm
Gain > 24 dB

That means a signal of 5-24 = -19 dBm would drive that into compression. A
typical amateur radio LNA would almost certainly have a 1 dB compression
point of < 0 dBm.

I reckon something around a 10 dB attenuator on the output of port 0 would
be desirable before an LNA, and another 20 dB or so after the amplifier.

A typical amateur LNA would use a FET. The largest 1 dB compression point
would be achieved with a high drain current and lowest noise at a smaller
drain current.


Dave

Dave.
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...

Telephone 01621-680100./ +44 1621 680100

Registered in England & Wales.
Company number 08914892.
Registered office:
Stokes Hall Lodge,
Burnham Rd,
Althorne,
Chelmsford,
Essex,
CM3 6DT,
United Kingdom