Topics

reference plane

Ken Bozarth
 

Am I missing something big here? Does this nanovna calibrate the end of the cable as the reference plane? All VNAs should do this. Mine does not. Recalling cal 0, with the TX connector open, it looks like an open. If I do a complete calibration at the end of the cable, then leave it open at the end, it is not an open, at any frequency. What am I doing wrong?

John Galway
 

While not an expert on this I check my SWR at the antenna to take the cable out of the equation.  After adjusting the antenna I add the cable to the test to see what the addition of the cable does to the system.  Based on that I work to get the lowest SWR for the system.  That being said I would calibrate the VNA using none for at the most a very short cable.
John GalwayKH6JTE

On Thursday, October 10, 2019, 5:07:43 PM CDT, Ken Bozarth <kwbozarth@...> wrote:

Am I missing something big here? Does this nanovna calibrate the end of the cable as the reference plane? All VNAs should do this. Mine does not. Recalling cal 0, with the TX connector open, it looks like an open. If I do a complete calibration at the end of the cable, then leave it open at the end, it is not an open, at any frequency. What am I doing wrong?

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Thu, 10 Oct 2019 at 23:07, Ken Bozarth <kwbozarth@...> wrote:

Am I missing something big here? Does this nanovna calibrate the end of
the cable as the reference plane? All VNAs should do this. Mine does not.
Recalling cal 0, with the TX connector open, it looks like an open. If I do
a complete calibration at the end of the cable, then leave it open at the
end, it is not an open, at any frequency. What am I doing wrong?

You are doing nothing wrong - it is a limitation of the firmware in the
NanoVNA. The calibration standards supplied are male. When you add a
female-female thru on, you are adding a bit of transission line with
one-way delay of about 42 ps, so 84 ps on reflection. The firmware
currently has no way to correct for that.

So when you remove the calibration standard, the calibration is wrong.

Be aware that the male pin of an SMA protudes several mm above the
reference plane, which adds capacitance.

Hopefully someone will fix the firmware.

Dave

--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Thu, 10 Oct 2019 at 23:43, Dr. David Kirkby <
drkirkby@...> wrote:


Be aware that the male pin of an SMA protudes several mm above the
reference plane, which adds capacitance.
I should have been clearer here. There are two issues

1) The firmware prevents you getting a proper calibration at the reference
plane.

2) Even if the firmware is corrected, you can not expect a make SMA plug to
act like an ideal open.

Both a female SMA socket, as well as an APC7 connector are somewhat close
to an ideal open. A male SMA will never be, nor will any N connector.



Dave
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

John Gord
 

Ken,
As delivered, the NanoVNA default calibration, Location 0, is apparently set up with the S11 reference plane at the on-board connector. You can run another calibration at the end of your cable put that in another storage location if you like.
--John Gord

On Thu, Oct 10, 2019 at 03:07 PM, Ken Bozarth wrote:


Am I missing something big here? Does this nanovna calibrate the end of the
cable as the reference plane? All VNAs should do this. Mine does not.
Recalling cal 0, with the TX connector open, it looks like an open. If I do a
complete calibration at the end of the cable, then leave it open at the end,
it is not an open, at any frequency. What am I doing wrong?

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Thu, 10 Oct 2019 at 23:57, johncharlesgord via Groups.Io <johngord=
verizon.net@groups.io> wrote:

Ken,
As delivered, the NanoVNA default calibration, Location 0, is apparently
set up with the S11 reference plane at the on-board connector. You can run
another calibration at the end of your cable put that in another storage
location if you like.
--John Gord

But it is currently *totally impossible to get a valid calibration on a
cable with a male SMA on the end. *The firmware doesn’t permit one to
enter any characteristics of the female calibration standards. Female SMA
calibration standards will be nothing like text-book opens and shorts,
whereas male calibration standards will be fairly close to text-book opens
and shorts.

Without fixes to the firmware, this problem will not go away.

Dave.



On Thu, Oct 10, 2019 at 03:07 PM, Ken Bozarth wrote:


Am I missing something big here? Does this nanovna calibrate the end of
the
cable as the reference plane? All VNAs should do this. Mine does not.
Recalling cal 0, with the TX connector open, it looks like an open. If I
do a
complete calibration at the end of the cable, then leave it open at the
end,
it is not an open, at any frequency. What am I doing wrong?


--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Kurt Poulsen
 

Hi Ken
Shure you can calibrate at the end of cable.
You have noted when CH0 left open you see an open after a full calibration with Short nothing as open and laod. It has been compensated internally by 50fF which is the delay of the female SMA adaptor. When you calibrate at the end of a SMA male male adaptor you are forced to use the female female adaptor to which you fit your male calibration prt being short, nothing and the load. When removing the male calibration part and leave the female female adaptor in place the you see the same open as when calibrating directly to Ch0. You probably want to remove the fem female adaptor for measuring on a device and then you have to do the following trick.
1. Enable a Ch0 phase trace with 1 degree/divisionplus reference level 5, and then you place the marker mid screen or any other place and note the phase shift. Then remove the female female adaptor and enable Display/Scale/Electrical delay and enter about 65ps delay and then fine tune this delay until you have the same phase shift/phase trace as before thus removing the female adaptor. What you do here is pulling the calibration plane backwards by the amount the female female adaptor has as delay and not the measurement plane is the calibration plane of the SMA male adaptor.
Remember the NanoVNA as a native device assumes ideal short (0ps delay) Ideal open (however compensated to 50fF) and ideal 50 ohm load. Lucky you the supplied load is pretty goods and the male short is almost 0 ps. When using other calibration parts then use NanoVNA.saver and enter the data you find/get for thes calibration parts. I jut uploaded the data for the female kit based on the suppled parts and then you do not need to do the trick just described. But the you need to bring the PX with you to e.g. an antenna.
Have Fun
Kind regards
Kurt


-----Oprindelig meddelelse-----
Fra: nanovna-users@groups.io <nanovna-users@groups.io> På vegne af Ken Bozarth
Sendt: 11. oktober 2019 00:08
Til: nanovna-users@groups.io
Emne: [nanovna-users] reference plane

Am I missing something big here? Does this nanovna calibrate the end of the cable as the reference plane? All VNAs should do this. Mine does not. Recalling cal 0, with the TX connector open, it looks like an open. If I do a complete calibration at the end of the cable, then leave it open at the end, it is not an open, at any frequency. What am I doing wrong?

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Fri, 11 Oct 2019 at 00:52, Kurt Poulsen <@TheBoss> wrote:

Hi Ken
Shure you can calibrate at the end of cable.

Yes, Kurt is right, you can compensate this in software by moving the
reference plane. It is far from ideal - it would be better if the data on
the calibration standards could be properly entered.
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Kurt Poulsen
 

Hi David
I think you should read my response to this question.
You negative conclusion is not valid.
Pragmatic solution are still useable and in addition using the NanoVNA-saver you got it all to perfection. You even was the person who motivated Rune to developed his calibration function with L and C coefficients and his load are even able to cope with the inductive loads such as the BNC calibration kit from SDR Kits. IT IS PLAIN PERFECT David.
I know you want to flood the firmware with all sort of default professional calibration kit data, It is you who are dreaming of a tolls only very few would benefit from (who want to fool around in outdoor environment with the professional calibration kit anyway) and as a bench tool I definitely prefer the NanoVNA to be as is, as a bloody good and accurate "vector measurement engine" and solve rest of the challenges by software.
And for a spot measurement device in the field, you can do with the described tricks very well even with other homemade calibration kits having same plane for short open and load and a known delay to the real calibration plane. The only improvement I can think off is the possibility to enter negative delay in the display/Scale/Electrical delay so we also can move the calibration plane forward.
Kind regards
Kurt


-----Oprindelig meddelelse-----
Fra: nanovna-users@groups.io <nanovna-users@groups.io> På vegne af Dr. David Kirkby from Kirkby Microwave Ltd
Sendt: 11. oktober 2019 01:14
Til: nanovna-users@groups.io
Emne: Re: [nanovna-users] reference plane

On Thu, 10 Oct 2019 at 23:57, johncharlesgord via Groups.Io <johngord= verizon.net@groups.io> wrote:

Ken,
As delivered, the NanoVNA default calibration, Location 0, is
apparently set up with the S11 reference plane at the on-board
connector. You can run another calibration at the end of your cable
put that in another storage location if you like.
--John Gord

But it is currently *totally impossible to get a valid calibration on a cable with a male SMA on the end. *The firmware doesn’t permit one to enter any characteristics of the female calibration standards. Female SMA calibration standards will be nothing like text-book opens and shorts, whereas male calibration standards will be fairly close to text-book opens and shorts.

Without fixes to the firmware, this problem will not go away.

Dave.



On Thu, Oct 10, 2019 at 03:07 PM, Ken Bozarth wrote:


Am I missing something big here? Does this nanovna calibrate the end
of
the
cable as the reference plane? All VNAs should do this. Mine does not.
Recalling cal 0, with the TX connector open, it looks like an open.
If I
do a
complete calibration at the end of the cable, then leave it open at
the
end,
it is not an open, at any frequency. What am I doing wrong?


--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Fri, 11 Oct 2019 at 01:15, Kurt Poulsen <@TheBoss> wrote:

Hi David
I think you should read my response to this question.

You negative conclusion is not valid.
Pragmatic solution are still useable and in addition using the
NanoVNA-saver you got it all to perfection.

Yes, but NanoVNA-saver doesn’t give one a *stand-alone, self contained unit*
like an Agilent FieldFox. Personally I don’t want to have to use an
external computer with the NanoVNA.

I obtained a couple of those T-shaped calibration standards with an open,
short and load. I paid less than $50 each. A few user-defined calibration
kits solves the problem.

*You and I have very different intended uses. I think we will just have to
agree to differ *

1) You wish to use it as a data collection device and do the processing in
software on a PC - much like the VNWA

2) I want to be able to use it outside, without having an external PC.

The Agilent FieldFox gives the best of both worlds.

* Totally self contained, so maybe used without any PC.

* Software that allows one stream data directly to a PC if one wishes.


Dave.
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

QRP RX
 

you're needs to perform RESET before calibration. If you don't do that, you will get strange results.

RFy
 

On Fri, Oct 11, 2019 at 09:58 AM, Dr. David Kirkby from Kirkby Microwave Ltd wrote:
The Agilent FieldFox gives the best of both worlds.
I think you should keep to use your FieldFox and ...
Try one of this group's EXPERT solutions. below
I think you should either

1) Throw it in the bin
2) Seek a refund.
I found someone asking 'feature request' without encouraging/cheering, Just giving complaint/discourage to the contributors.

Erik Kaashoek
 

On Thu, Oct 10, 2019 at 05:15 PM, Kurt Poulsen wrote:


The only improvement I can think off is the possibility to enter negative
delay in the display/Scale/Electrical delay so we also can move the
calibration plane forward.
@Kurt
Can you check this firmware
https://groups.io/g/nanovna-users/files/Firmware/NanoVNA%20firmware%20build%20by%20Erik/Release%200.2.3%20width%20modified%20scan%20command%20and%20negative%20edelay.dfu
It implements the ability to set negative edelay in the UI but I do not know if the rest of the calculations is correct

Dr. David Kirkby from Kirkby Microwave Ltd <drkirkby@...>
 

On Fri, 11 Oct 2019 at 07:10, RFy <gpdxdveil@...> wrote:


I think you should keep to use your FieldFox and ...
Try one of this group's EXPERT solutions. below

I think I should keep clear of FieldFoxs. I have had two that failed

* An Agilent reconditioned N9923A VNA that developed a fault requiring a
replacement main board
* A used N9912A combination analyzer that arrived DOA.

There’s no doubt that the NanoVNA is an amazing device for the money. I
have enoughed others at my amateur radio club to buy one - so far without
success.

Dave
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Richard G4TGJ
 

Comparing the NanoVNA to the FieldFox is a bit pointless considering it is 1000th the price.

--
73
Richard
G4TGJ

RFy
 

Who knows one of Ph.D might publish $50USD FieldFox-GNU.
If that happens, I will ask to contributer like below.
-'FieldFox-GNU is great performance/price ratio. But poor than R&S ZNA110. Any schedule of improvements?'
-'Provided calibration standards worse than 85521A. Is that really better than SDR-kit's SOL kit?'
-'Retina display is better, Please add play video or game function'

Kurt Poulsen
 

Hi Erik
Thank you very much for the hint. I will do so when my second NanoVNA arrives. I am not at all familiar with this "magic updating methods and must maintain my first NanoVNA as is not to "fuck it up" sorry for my language.
I am buying no2 for experimenting only
I hope you understand
Kind regards

-----Oprindelig meddelelse-----
Fra: nanovna-users@groups.io <nanovna-users@groups.io> På vegne af erik@...
Sendt: 11. oktober 2019 09:22
Til: nanovna-users@groups.io
Emne: Re: [nanovna-users] reference plane

On Thu, Oct 10, 2019 at 05:15 PM, Kurt Poulsen wrote:


The only improvement I can think off is the possibility to enter
negative delay in the display/Scale/Electrical delay so we also can
move the calibration plane forward.
@Kurt
Can you check this firmware
https://groups.io/g/nanovna-users/files/Firmware/NanoVNA%20firmware%20build%20by%20Erik/Release%200.2.3%20width%20modified%20scan%20command%20and%20negative%20edelay.dfu
It implements the ability to set negative edelay in the UI but I do not know if the rest of the calculations is correct

Kurt Poulsen
 

Hi David
Well I will go that far if a menu point allowed entering delays for open and short in addition to C0 for open, then Resistance shunt C series L for the load then it would be flexible and cope with the many calibration standards to purchase form SDR-Kit where there is full documentation available. But that is the end of it, because you forgetting totally that there is no means for doing 6/12 term error correction and entering any commercial calibration kit definitions is an insult to these commercial calibration kits
Chew on that, please....and then dream on, but not providing much more noise on the subject
That is my humble invitation
Kind regards
Kurt

-----Oprindelig meddelelse-----
Fra: nanovna-users@groups.io <nanovna-users@groups.io> På vegne af Dr. David Kirkby from Kirkby Microwave Ltd
Sendt: 11. oktober 2019 02:58
Til: nanovna-users@groups.io
Emne: Re: [nanovna-users] reference plane

On Fri, 11 Oct 2019 at 01:15, Kurt Poulsen <@TheBoss> wrote:

Hi David
I think you should read my response to this question.

You negative conclusion is not valid.
Pragmatic solution are still useable and in addition using the
NanoVNA-saver you got it all to perfection.

Yes, but NanoVNA-saver doesn’t give one a *stand-alone, self contained unit* like an Agilent FieldFox. Personally I don’t want to have to use an external computer with the NanoVNA.

I obtained a couple of those T-shaped calibration standards with an open, short and load. I paid less than $50 each. A few user-defined calibration kits solves the problem.

*You and I have very different intended uses. I think we will just have to agree to differ *

1) You wish to use it as a data collection device and do the processing in software on a PC - much like the VNWA

2) I want to be able to use it outside, without having an external PC.

The Agilent FieldFox gives the best of both worlds.

* Totally self contained, so maybe used without any PC.

* Software that allows one stream data directly to a PC if one wishes.


Dave.
--
Dr. David Kirkby,
Kirkby Microwave Ltd,
drkirkby@...
https://www.kirkbymicrowave.co.uk/
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

Ken Bozarth
 

Thanks to all for the guidance. I found that I need to Reset before doing a new cal. The reference plane stays at the end of the cable, where it should. In fact, I now use a pair of 3-foot RG188 cables, and the reference plane is behaving itself. I use 100MHz and down, so the little things do not bother me. I wish the unit were heavier, as the RG188 cables easily drag the unit off the table!

reuterr@...
 

Hello Kurt,

Well I will go that far if a menu point allowed entering delays for open and short in addition to C0 for open,
then Resistance shunt C series L for the load then it would be flexible and cope with the many calibration standards
to purchase form SDR-Kit where there is full documentation available.
But that is the end of it, because you forgetting totally that there is no means for doing 6/12 term error correction
and entering any commercial calibration kit definitions is an insult to these commercial calibration kits
Chew on that, please....and then dream on, but not providing much more noise on the subject
Thank you very much for the clear words.

I think there are 4 major topics in the group:
1. Applications
2. Newbies
3. Repair
4. Tuning

For my taste the tuning topic is too much, and the Applications is too few.

73, Rudi DL5FA

Oristo
 

The only improvement I can think off is
the possibility to enter negative delay in the display/Scale/Electrical delay
so we also can move the calibration plane forward.
edy255 is adding to mainstream firmware:
https://circleci.com/gh/ttrftech/NanoVNA/115

Erik Kaashoek
 

Yes, I did send him my code .
Its tested now and works well

WB2UAQ
 

Ken,
You hit the nail on the head. At low frequencies, at VHF and down, I wouldn't worry about picosecond delays or femto-farad fringing C's. And I bet 99% of the Nano users will be looking at SWR only............no worries about the measurement plane.
I am mounting my Nano on a plate and adding N connectors with short cables connected to the plate to strain relief the Nano's connectors. This will give it a little more mass to deal with even the tugging of the USB cable:)
73