开云体育

This is what I ordered from Amazon. They are high quality and each measured 50 ohms. USA made by Amphenol that makes the best RF connectors.

Amphenol Connex 132360 (Pack of 4) RF Terminators SMA Male Resistor Cap 50 OHM

Gary
AA5I

You can set time by use time command

Filaname VNA_19800101_02850 is 1980 year 01 mounth 01 day 2850 second

See this post /g/nanovna-users/message/13957

You use this? from post /g/nanovna-users/message/13941
NanoVNA H4 v0.9.3.2 beta - SD Card

v0.9.3.3 is only H version at this moment (for H4 i not compile it) in this version i speed up write from 450kb/s up to 850kb/s on good SD cards (remove wait SD card ready write answer after write, wait only on receive write command)

I'm just starting to learn how to use and what everything means, and have never used a VNA.
I'm using a NanoVNA H4 Purchased from R&L. As you may know it comes with OSL and a barrel to calibrate the end of the cable.
So I did that and tuned on correction. (not sure why or what it does).
After Calibration, the resistor measures 49.9Ω and 200Ph inductance, that bounces a little so Plus or Minus 100pH

Then I installed a female SMA edge mount connector (curiosity of my friend Bob) with short leads to plug into the Test Board.
See the attachment.
Then I put the VNA supplied, 50Ω on the SMA connector, the measurement with the same cal as above but through the test board,
is 50.0Ω with 5.9nH But much more stable Plus of minus 0.4nH.

Next I installed just the 50Ω load that I solder together from the SDR-Kits kit, and it
measured 49.0Ω with 900nH plus or minus 250nH.

I'm not sure I did exactly what you ask, but, did you get the info you wanted?
If not explain a little more and I'll try again.

Mikek

On Tue, Jun 9, 2020 at 09:07 AM, Mikek wrote:

Does -1.5db loss for 10 turns wound over 7 turns on a #43 binocular core sound
about correct, or would you expect lower?

Your loss is too high and you can get a fraction of a dB with the right ferrite mix selection, number of turns and winding arrangement.

From your design description it seems that you want to go as low as 500 kHz. Your primary was 7 turns with a system impedance of 50 ohms. The rule of thumb is that the reactance should be 4x50 = 200 ohms when nothing is connected to the secondary. Using the calculator at toroids.info we get 68 ohms at 500 kHz. using an FT50-43 toroid. You need 12 turns to get the desired 200 ohms on the primary which means 17 turns on the secondary for a 50:100 transformer. If you wind the primary on top of the secondary you will get a lot of capacitive coupling between the windings. There will also be a lot of capacitance between turns. You can try winding on opposite sides of the core and on top and compare your results.

It is generally best to have fewer turns and this means a core with higher permeability. I suggest using a FT50-75, a BN-43-202 or a BN-73-202. With the FT50-75 you only need 4 turns on the primary and 6 turns on the secondary. I like binocular cores because they give a high coupling factor between the primary and the secondary with only a few turns and less coupling capacitance. A BN-43-202 is a good choice and only requires 6 turns on the primary and 8 to 9 turns on the secondary. Also a BN73-202 with 3 turns on the primary and 4 turns on the secondary might work for you.

You can see my calculations in the graphics below. There is also a photo showing how to wind binocular cores for those not familiar with them.

Roger

Thanks to all of you who tried to answer my question [rather than complaining about me asking it :-) ]

Thanks, particularly, to Bruce for some support there!

The idea of using two 50 ohm loads on a tee had not occurred to me. I guess that up at, say, 5GHz the length of the tee would be an issue - but not at these lower frequencies. Thanks for that idea.

It also occurred to me that the S11 of any attenuator with an open or short on its output should measure 2x the attenuator's value - so there is another simple test that can tell me, to some extent, how well the device is working.

The time domain measurements are interesting too. I note that my device seems to be reasonably accurate in determining the length of a stub (adjusted for velocity factor.

So... a few sanity checks like this tell me that it works pretty well.

As a double check I would calibrate at the end of the cable using the supplied S-O-L Then hook up to the SDR board with load and see how it compares to Cal Load. A perfect 50 ohm SM resistor is hard to find.

Very interesting writeup and beyond where I am at the moment. I'm having to learn all over again about how to setup the NanoVNA.

Sorry, its not the KARN, its the ANNE-50X+. Someone else mentioned another
model from Minicircuits. Pick whatever has the best guaranteed performance
over the range you want. Ignore the typical performance - that is based on
a sample of 1 from Mincircuits, which is a pretty crap way of getting a
typical set of data to me, but I was told by a Minicircuits employee that
"typical" means one random one taken from the production line.

Dr David Kirkby Ph.D C.Eng MIET
Email: drkirkby@... Web:

Kirkby Microwave Ltd (Tel 01621-680100 / +44 1621-680100)
Stokes Hall Lodge, Burnham Rd, Chelmsford, Essex, CM3 6DT.





On Tue, 9 Jun 2020 at 18:23, Dr. David Kirkby <
drkirkby@...> wrote:

On Tue, 9 Jun 2020 at 18:13, Wes Plouff <ac8jf@...> wrote:

I received my NanoVNA-H4 from AliExpress in just 10 days, and am happy
with my purchase. However, the 50 ohm SMA calibration load goes
intermittently open. The cap rotates even when the load is screwed firmly
to the VNA port connector, so I think the part is defective. The
calibration load from my hugen NanoVNA works properly on the H4.

What are good sources for a replacement part that have good performance
to 1.5 GHz, are reasonably priced, and ship from the USA? Am I stuck buying
an entire calibration kit? Thanks for any advice.

A Minicircuits KARN-50X+ is a pretty decent male load. Mechanically they
are good, and do not rotate as you screw them in. However, unless you
either pay lots of money, or pay someone to measure one for you, there's
never any guarantee of what it will be like. But Minicircuits have a
specification, and 99.999% of their devices meet that specification. You
may find another Minicircuits load with a better guaranteed performance to
a few GHz.

On Tue, 9 Jun 2020 at 18:13, Wes Plouff <ac8jf@...> wrote:

I received my NanoVNA-H4 from AliExpress in just 10 days, and am happy
with my purchase. However, the 50 ohm SMA calibration load goes
intermittently open. The cap rotates even when the load is screwed firmly
to the VNA port connector, so I think the part is defective. The
calibration load from my hugen NanoVNA works properly on the H4.

What are good sources for a replacement part that have good performance to
1.5 GHz, are reasonably priced, and ship from the USA? Am I stuck buying an
entire calibration kit? Thanks for any advice.

A Minicircuits KARN-50X+ is a pretty decent male load. Mechanically they
are good, and do not rotate as you screw them in. However, unless you
either pay lots of money, or pay someone to measure one for you, there's
never any guarantee of what it will be like. But Minicircuits have a
specification, and 99.999% of their devices meet that specification. You
may find another Minicircuits load with a better guaranteed performance to
a few GHz.

Thanks DiSlord, I installed miniSD reader and 32.768 kHz xtal on PC 14 and PC15 on my H4 as per your instruction.
The screenshots, s1p and s2p can be saved now on the SD card. I used the NanoVNA H4 v0.93.3 beta_SD Card
LSE_Clock.dfu firmware.

This model from Mini-Circuits is a good quality unit:

About $7.

I received my NanoVNA-H4 from AliExpress in just 10 days, and am happy with my purchase. However, the 50 ohm SMA calibration load goes intermittently open. The cap rotates even when the load is screwed firmly to the VNA port connector, so I think the part is defective. The calibration load from my hugen NanoVNA works properly on the H4.

What are good sources for a replacement part that have good performance to 1.5 GHz, are reasonably priced, and ship from the USA? Am I stuck buying an entire calibration kit? Thanks for any advice.

Thanks Erik, I reread the page I got that from and ya, I got it backwards.
" If S11=0 dB, then all the power is reflected from the antenna and nothing is radiated"
I see it now.
Thanks, Mikek

On Tue, Jun 9, 2020 at 09:07 AM, Mikek wrote:

I read that Ch0 10db/ -20.89db is the return loss, so, looking into the first
transformer,
it Is high (not 0db) because the 1.41uH is 8.9Ω at 1MHz and that causes most
of the -20.89 return loss?

Return loss is determined by how much the the thing connected to CH0 is deviating from pure 50ohm.
The more perfect the thing, the higher the return loss (e.g. no echo from bad impedance matching)
So -20dB is a very good match and indeed is measures 50.7Ω 1.41uH


--
NanoVNA Wiki: /g/nanovna-users/wiki/home
NanoVNA Files: /g/nanovna-users/files
Erik, PD0EK

I have the Test board from SDR-Kits as shown in the Attachment.
I've had inconsistent results measuring two back to back 50 ohm to 100 ohm transformers.
At one point I measured -0.4db loss for two and now consistently -2.78db At 500kHz to -3.16db at 4MHz,for two transformers.
Does -1.5db loss for 10 turns wound over 7 turns on a #43 binocular core sound about correct, or would you expect lower?
Here's what the display reads,
Ch0 10db/ -20.89db Ch1 LogMag 10db/ -2.77db
CH0 Smith 1.0FS 50.7Ω 1.41uH CH1 Phase 90*/ -3.051*
1MHz

What I think I know about those numbers, is, if I divide the -2.77 in half, that is a single transformer's loss,
Looking into the first transformer it does look like 50.7Ω with just a bit of inductance at 1MHz..
Looks like a slight phase shift, maybe caused by that bit of inductance.
I read that Ch0 10db/ -20.89db is the return loss, so, looking into the first transformer,
it Is high (not 0db) because the 1.41uH is 8.9Ω at 1MHz and that causes most of the -20.89 return loss?
Is that correct?

I posted the Test board to verify my cal procedure.
Cal-Calibration
Insert open 7E to 7F, push- Open
Insert short 7E to 7F, push- Short
Insert 50 ohm 7E to 7F, push- Load
Insert 50 ohm 1E to 1F, push- Isolate
Install wire from D1 to D7, push- Thru
Push-Done
Push Back
Push Save
Pick a slot-Save1

When I finish the cal, with the thru wire still in, I have the following readings.

CH0 10db/ -48.46 CH1 LogMag 10db/ 0.00db
CH0 Smith 1.0FS 50.3Ω 5.71nH Ch1 Phase 30*/ -0.002

Anything I need to change?

What does CORRECTION do?
If it's in the wrong position I get 68db loss on my transformers, but I have no clue what it's doing.

I think I learned a lot while formulating my question! It forced me to look things up! :-)

Thank you for your help, Mikek

Whose firmware are you using?
Edy555 and Hugen's firmware is more conservative and does not push the limits of the device.
DiSlord's firmware is cutting edge and pushes the device's capabilities.
You need to chose what you want - or just install the F/W from each and experiment to see which one your device works best with.

I've got a NanoVNA from Aliexpress so I'm not entirely sure who actually manufactured it but when I use the latest firmware to get the extended frequency range I see loads of noise in most of the UHF. I cal 2m and 70cm ranges (either as a single sweep or two different collections). Using either the build-in interface or NanoVNA saver I get a pretty flat 50ohm response on the 2m range (and expected responses with open and short) after calibration but the 70cm range is an unusable mess. The analog front ends don't have cans on them and I'm not sure if that's something installed on "official" hardware builds or is missing from my particular one. That's the obvious place I see for noise introduction. Is this a known problem? Ideas where the noise might be coming from? Should I build cans for the front ends and see if that helps?

For a nonaVNA-V2/SAA-2 there are in my relative unexperienced view two easy to do measurements that tell a lot about the performance (of course after calibration)

1: Connect the ports with a 60dB attenuator. If the S21 is nice and flat up to above 3.5GHz you can be happy. If you remove the connection and you see the S21 dropping (a lot) you can even be happier

2: Connect two 50 ohm loads to a Tee and connect the tee to port 1 (CH0). If the S21 is a constant 25ohm (+/- 1ohm) with a only a phase rotation with increasing frequency that you can remove by adding a port 1 delay you can be happy. If this pure rotation continues above 3.5GHz you can be very happy.






--
NanoVNA Wiki: /g/nanovna-users/wiki/home
NanoVNA Files: /g/nanovna-users/files
Erik, PD0EK

Hi Dale,

Nice to know that:) I think it was Mark Twain who said "reports of my death were greatly exaggerated" :) When my father was around about 30 years old anther man with the same name and close in age was killed in a car wreck. People who knew my father were astonished when he walked into the coffee shop.

73,

Bill KU8H

bark less - wag more

The compiler toolchain puts things like date stamps into the binary files. Every time you build you will get a different file, and this is not a problem.

There are options to disable it, or you can convert the binary to a format that doesn't have the variable data, I recall (but it's been years since I wanted to do that).

On 6/8/2020 11:16 AM, Bill Cromwell wrote:

Hi,

There was a trade school franchise known as "RETS" for Radio Electronics Television School. I attended some of their classes then after my time in the Navy. I doubt they are still in business. Another franchise I knew about "CIE" is long out of business. Good luck with your search.

73,

Bill? KU8H

bark less - wag more

CIE is still in business.

I took there courses while I was in the Army back in the late 80's early 90's.



--
All the Best & 7 3s
Dale Miller, KC2CBD
Cookeville, Tennessee
Putnam County TN ARES
Emergency Coordinator
Vice President Cookeville Repeater Association
Ham Operator since 1997 (Extra)

stpatrick2@...
Registered Linux User: #317401
Linux since June 2003
Registered Ubuntu User #26423

--



--
All the Best & 7 3's
Dale Miller, KC2CBD
Cookeville, Tennessee
Putnam County TN ARES Emergency Coordinator
Vice President Cookeville Repeater Association

Ham Operator since 1997 (Extra)
kc2cbd@...

Registered Linux User: #317401
Linux since June 2003
Registered Ubuntu User #26423