开云体育

Ken Boswoth reported error in calibrating on a cable with a male SMA on it.
Here’s a way to improve on that, despite the fact that the firmware doesn’t
allow one to do it properly.

Here are few facts

1) Typically the one-way delay of a female-female adapter is about 42 ps.

2) Typically the one-way delay of a male-female adapter, (often used as
connector savers) is about 55 ps.

3) You can easily get both types on eBay.

4) If you make up a female calibration standard by combining the male short
and female-female thru the calibration will have a significant error to
delay delay of the adapter.

*Now a possible trick*

* Use a female-female adapter as your female open. Do not just leave the
connector open

* Use a female-female adapter with a male short to make your female short.


After performing the calibration of the male plug, you will have the
calibration plane and reference plane differing by about 42 ps.

If you now add a male-female adapter onto that male plug, you will reduce
the error to the *difference* of the delays in the adapters. The error will
Reduce from 42 ps to 55-42= 13 ps.

As the director of a company making calibration kits, I am well aware that
there are some *gross* simplifications above. I am well aware of the fact
that the calibration standards can not just be considered delays.

However, the fact remains that adding a male-female adapter into the male
cable after calibration will reduce the error by about 70%.

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, 11 Oct 2019 at 01:00, Dr. David <drkirkby@...>
wrote:

Ken Boswoth reported error in calibrating on a cable with a male SMA on
it. Here’s a way to improve on that, despite the fact that the firmware
doesn’t allow one to do it properly.

Kurt is right. Moving the reference plane in software is going to be better
than doing the trick I suggested.

I don’t know why I didn’t think of this - I regularly use port extensions
to move the reference plane.

However, proper support of calibration standards would give more accurate
results.
--
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, Oct 11, 2019 at 03:14 AM, Dr. David Kirkby from Kirkby Microwave Ltd wrote:

However, proper support of calibration standards would give more accurate
results.

The problem here is that we don't know parameters of cal-kit which is supplied with nanoVNA...

So, just adding ability to setup cal-kit parameters won't help. Because we're needs to know what parameters needs to be entered.

Does someone tried to measure parameters for cal-kit supplied with nanoVNA?

Hi qrp.dds
YES !!! I did publish these yesreday
Kind regards
Kurt
-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af qrp.ddc@...
Sendt: 11. oktober 2019 03:42
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the calibration of a male SMA connector

Does someone tried to measure parameters for cal-kit supplied with nanoVNA?

Hi Kurt,

yes, I just found your message.

Could you please clarify how I can check if cal-kit setting will help or
not? What I should check after C0 change?

Hi QRP RX aka qrp.dds ?

The best way to check if the calibration done by stand alone NanoVNA using ideal calibration condition and by using a calibration by NanoVNA-saver is by sweeping a some 25cm long semirigid cable which has low loss and well defined Z0=50 ohm. For a male calibration the will be not much difference as the male calibration part are fretty good and the SMA female adaptor for Ch0 is already compensated in open condition by internal 50fF. It it still unclear why 50fF and for what. Maybe the designer also included the supplied male open adaptor. However I found that using nothing for open was looking like the right choice.

What you should see it a Smith chart trace from 50KHz to 900MHz following closely the circumference of the Smithchart and not running outside the circumference. I attach an image where I sweep a 20cm HP airline and with a semirigid cable I got the same fine result. Although it is done with the NanoVNWA-saver and using my HP85033C 3.5mm kit for calibration I get same result using the suppled male kit. This image proof the high quality of the NanoVNA and the best condition used for it investigation. The small bump extreme left is due to the 300MHz spikes when NanoVNA switches over to harmonic operation. OZ2M Bo has explained some time ago why the NanoVNA should not have gone beyond some 275MHz in fundamental mode, but "who in hell" measures around 300MHz ?.

The benefit of using my data is when you calibrate at the end of a test cable and want a calibration at the SMA male adaptors calibration plane, then you must use NanoVNA-user and enter the female data. When the sweeping a semirigid cable - now with a female adaptor at the end of the semirigid cable, (I have male and female in the two ends) you will see the trace is running outside the Smithchart if you have a female female adaptor with loss, because thee loss is embedded in the calibration (see attached image). If not then the trace will follow the circumference just like the male calibration did.
If you do not have a semirigid cable use a good quality short SMA test cable of some 220 to 30cm length and see it spins inward due to the loss in the cable. Select a S11 dB trace and observe the trace and note marker values for the two different calibration. For male calibration at Ch0 directly will show no difference I am sure but time will show what you experience for a female calibration.

Hope this is of help
You may show us your findings

Kind regards

Kurt

-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af QRP RX
Sendt: 11. oktober 2019 12:15
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the calibration of a male SMA connector



Hi Kurt,



yes, I just found your message.



Could you please clarify how I can check if cal-kit setting will help or not? What I should check after C0 change?

Am I correct in assuming that SMA connector/adapter delays are a neligible problem at HF?

Hi W5DXP
Correct
Kind regards
Kurt de OZ7OU

-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af W5DXP
Sendt: 11. oktober 2019 13:52
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the calibration of a male SMA connector

Am I correct in assuming that SMA connector/adapter delays are a neligible problem at HF?

Fri, 11 Oct 2019 at 12:49, Kurt Poulsen <kurt@...> wrote:

i QRP RX aka qrp.dds ?

Although it is done with the NanoVNWA-saver and using my HP85033C 3.5mm kit

for calibration I get same result using the suppled male kit. This image
proof the high quality of the NanoVNA and the best condition used for it
investigation. The small bump extreme left is due to the 300MHz spikes when
NanoVNA switches over to harmonic operation. OZ2M Bo has explained some
time ago why the NanoVNA should not have gone beyond some 275MHz in
fundamental mode, but "who in hell" measures around 300MHz ?.

The benefit of using my data is when you calibrate at the end of a test
cable and want a calibration at the SMA male adaptors calibration plane,
then you must use NanoVNA-user and enter the female data. When the sweeping
a semirigid cable - now with a female adaptor at the end of the semirigid
cable, (I have male and female in the two ends) you will see the trace is
running outside the Smithchart if you have a female female adaptor with
loss, because thee loss is embedded in the calibration (see attached
image). If not then the trace will follow the circumference just like the
male calibration did.
If you do not have a semirigid cable use a good quality short SMA test
cable of some 220 to 30cm length and see it spins inward due to the loss in
the cable. Select a S11 dB trace and observe the trace and note marker
values for the two different calibration. For male calibration at Ch0
directly will show no difference I am sure but time will show what you
experience for a female calibration.

Hope this is of help
You may show us your findings

Kind regards

Kurt

-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af QRP RX
Sendt: 11. oktober 2019 12:15
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the calibration
of a male SMA connector



Hi Kurt,



yes, I just found your message.



Could you please clarify how I can check if cal-kit setting will help or
not? What I should check after C0 change?



On Fri, Oct 11, 2019 at 12:57 PM Kurt Poulsen < <mailto:kurt@...>
kurt@...> wrote:

Hi qrp.dds

YES !!! I did publish these yesreday

Kind regards

Kurt

-----Oprindelig meddelelse-----

Fra: <mailto:[email protected]> [email protected] <

<mailto:[email protected]> [email protected]> P? vegne af

<mailto:qrp.ddc@...> qrp.ddc@...

Sendt: 11. oktober 2019 03:42

Til: <mailto:[email protected]> [email protected]

Emne: Re: [nanovna-users] A way to improve the accuracy of the

calibration of a male SMA connector

Does someone tried to measure parameters for cal-kit supplied with

nanoVNA?

--
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892

Tel 01621-680100 / +44 1621-680100

--
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

Please ignore my empty message - it was sent in error.

On Fri, 11 Oct 2019 at 14:46, Dr. David Kirkby <
drkirkby@...> wrote:

Fri, 11 Oct 2019 at 12:49, Kurt Poulsen <kurt@...> wrote:

i QRP RX aka qrp.dds ?

Although it is done with the NanoVNWA-saver and using my HP85033C 3.5mm

kit for calibration I get same result using the suppled male kit. This
image proof the high quality of the NanoVNA and the best condition used for
it investigation. The small bump extreme left is due to the 300MHz spikes
when NanoVNA switches over to harmonic operation. OZ2M Bo has explained
some time ago why the NanoVNA should not have gone beyond some 275MHz in
fundamental mode, but "who in hell" measures around 300MHz ?.

The benefit of using my data is when you calibrate at the end of a test
cable and want a calibration at the SMA male adaptors calibration plane,
then you must use NanoVNA-user and enter the female data. When the sweeping
a semirigid cable - now with a female adaptor at the end of the semirigid
cable, (I have male and female in the two ends) you will see the trace is
running outside the Smithchart if you have a female female adaptor with
loss, because thee loss is embedded in the calibration (see attached
image). If not then the trace will follow the circumference just like the
male calibration did.
If you do not have a semirigid cable use a good quality short SMA test
cable of some 220 to 30cm length and see it spins inward due to the loss in
the cable. Select a S11 dB trace and observe the trace and note marker
values for the two different calibration. For male calibration at Ch0
directly will show no difference I am sure but time will show what you
experience for a female calibration.

Hope this is of help
You may show us your findings

Kind regards

Kurt

-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af QRP RX
Sendt: 11. oktober 2019 12:15
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the
calibration of a male SMA connector



Hi Kurt,



yes, I just found your message.



Could you please clarify how I can check if cal-kit setting will help or
not? What I should check after C0 change?



On Fri, Oct 11, 2019 at 12:57 PM Kurt Poulsen < <mailto:kurt@...>
kurt@...> wrote:

Hi qrp.dds

YES !!! I did publish these yesreday

Kind regards

Kurt

-----Oprindelig meddelelse-----

Fra: <mailto:[email protected]> [email protected] <

<mailto:[email protected]> [email protected]> P? vegne af

<mailto:qrp.ddc@...> qrp.ddc@...

Sendt: 11. oktober 2019 03:42

Til: <mailto:[email protected]> [email protected]

Emne: Re: [nanovna-users] A way to improve the accuracy of the

calibration of a male SMA connector

Does someone tried to measure parameters for cal-kit supplied with

nanoVNA?

--
Dr David Kirkby Ph.D C.Eng MIET
Kirkby Microwave Ltd
Registered office: Stokes Hall Lodge, Burnham Rd, Althorne, CHELMSFORD,
Essex, CM3 6DT, United Kingdom.
Registered in England and Wales as company number 08914892

Tel 01621-680100 / +44 1621-680100

--
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

--
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, 11 Oct 2019 at 12:49, Kurt Poulsen <kurt@...> wrote:

Hi QRP RX aka qrp.dds ?
attach an image where I sweep a 20cm HP airline and with a semirigid
cable I got the same fine result.

Kurt,
I have kept meaning to ask you this for ages.

Can you attach a photograph and/or description of your 200 mm airline, and
a part number.

I have two airlines here, but are reluctant to use them for anything other
than their intended purpose, which is one of the 4 parts of the HP 85051B
(APC7) and 85053B (3.5 mm) verification kits, because the airlines are both
very delicate.

Neither have complete RF connectors on the end. Both have only the outer
conductor, with the inner conductor stored in a vial. One is expected to
add a coaxial short one end and support the centre conductor on that. At
the other end one is expected to support the centre conductor on the cable
one is connecting to. One is supposed to wear gloves when handling the
inner conductor. ????

See photographs of the APC7 outer on the far left, the APC7 in the centre,
and the 3.5 mm outer on the right.

The APC7 airline is 105 mm long, and the 3.5 mm airline 80 mm long.

I would like to find something a bit longer, and less delicate. I have seen
some with APC7 connectors on the end, but of course one needs an adapter to
SMA. The ones I have seen are either too expensive or too short to make
them attractive to buy. Someone in the UK had a 300 mm APC7 airline, but he
waned ?350 (GBP) which I was not willing to pay.

PS, I have seen some APC7 airlines on eBay, which are sold as “tested”, but
which are missing the centre conductor. Perhaps they were tested to see if
they could extend a hosepipe???

Kurt

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

I forgot to attach the picture of the two airlines. I hope you can see
they are not convenient to use
--
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 David

Mine are pure APC7 in both ends
1 pc HP 11567A 20.35cm and 2 pc. 11566A 10.21cm.
To fit to 3.5mm and SMA plus N I have male and female 3.5mm and N to APC7 adaptors

Just goggle the Airlines

I bought a full 3.5mm verification kit on e.bay but the seller "forgot" to deliver the center conductors ?

Kind regards

Kurt



-----Oprindelig meddelelse-----
Fra: [email protected] <[email protected]> P? vegne af Dr. David Kirkby from Kirkby Microwave Ltd
Sendt: 11. oktober 2019 16:30
Til: [email protected]
Emne: Re: [nanovna-users] A way to improve the accuracy of the calibration of a male SMA connector



I forgot to attach the picture of the two airlines. I hope you can see they are not convenient to use

--

Dr. David Kirkby,

Kirkby Microwave Ltd,

<mailto:drkirkby@...> 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, 11 Oct 2019 at 16:03, Kurt Poulsen <kurt@...> wrote:

Hi David

Mine are pure APC7 in both ends

That's *much* more convenient. I can see the unsupported ones are
technically better, but they are impractical for every testing.

1 pc HP 11567A 20.35cm and 2 pc. 11566A 10.21cm.

I just had an offer of $50 accepted for a 11567A. Unfortunately, the
carriage was more than the cost of the item. ????

I have been looking for a while. I was searching on “APC7 airline”, which
was not showing up much. I will look up part numbers.

To fit to 3.5mm and SMA plus N I have male and female 3.5mm and N to APC7

adaptors

Just goggle the Airlines

Cheers.

I bought a full 3.5mm verification kit on e.bay but the seller "forgot" to

deliver the center conductors ?

I see a lot of incomplete items on eBay that are sold as “tested”.

Kind regards

Kurt

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