开云体育

I agree that it leads to a different measurement. My claim is that it leads
to a more accurate measurement than using the supplied "open", since the
software does not compensate for the characteristics supplied open, but
rather assumes a near-perfect open.

The fully characterized cal kit I recently received also uses an open
connector (with suitable compensation values) as open, not an SMA cap like
supplied with the NanoVNA.

--
Rune / 5Q5R

I'm not doubting that having a 50R load on Ch1 is better, because I have never tried it.
I've never had to do it in the past on equipment in my old professional past either.
But just for fun, I've ordered some SMA 50R terminations to test the difference, I'm willing to be educated ;-)
Does anyone have data on measurements taken with or without ?

However, since MOST users will only have the basic calibration kit supplied, I write with that in mind,
and I get sensible results so far.

As for the debate regarding the OPEN standard, well the original poster is messing about with a 630m LPF.

I doubt 0.5pf across the terminals is gonna matter at 475Khz.

73 de Andy

Rune Broberg, the problem with "nothing connected to CH0" case is that it has different geometry configuration than usual SMA connector. It leads to different capacitance and inductance than usual SMA connector and also leads to different delay line, which leads to incorrect measurement. But it depends on what you needs. If you're work in 1-30 MHz range, you may keep "nothing connected to CH0" it will not affect result much. But if you want to use frequencies above 30 MHz, you're needs to use proper open terminator.

Given the assumptions of the NanoVNA, using the supplied open is *not*
necessarily better. The NanoVNA assumes that the open presents 50fF
capacitance and no frequency dependent capacitance - and you may well be
closer to that with a "true" open than with the added capacitance of the
"open terminator" from the kit.

--
Rune / 5Q5R

Andy G0FTD, there are two mistakes in your calibration procedure:

1) Press OPEN with nothing connected Ch0

this is incorrect. You're needs to place open terminator from calibration kit.

2) Keep the LOAD connected on CH0 and press ISOLN

this is incorrect. You're needs to place 50 Ω terminator on CH1 for ISOLN.

Probably you have incorrect calibration for ISOLN and THRU. Did you performed RESET before calibration? If no, try to perform RESET before calibration. Without it, you may get some issues with calibration.

Also try to connect CH0 with CH1 with cables which you're using for connect the filter. It should show 0 dB. If it shows something different, then you have wrong calibration.

Joe,

This looks exactly like the issue I had because I did not follow calibration procedure correctly.

If you do not follow calibration procedure correctly then you always end up with about -10db as a baseline.

Please try this..

Turn on NanoVNA.
Select desired freq range or default of 50Khz - 900MHz
From the menu press CAL
Press RESET
Then press CALIBRATE
Press OPEN with nothing connected Ch0
Menu now moves to where it says SHORT (a light shade of green)
Connect the SHORT on CH0 and press SHORT
Menu moves to LOAD and turns a light shade of green
Connect your 50R load to Ch0 and press LOAD
Menu moves to ISOLN with alight shade of green.
Keep the LOAD connected on CH0 and press ISOLN
Menu moves to THRU in a light shade of green.
Connect a short lead (provide from Ch0 to Ch1
Press THRU
Menu moves to DONE in a light shade of green.
Press DONE
Then press SAVE 0



73 de Andy

Mark K0NIA,

No. For ISOLN calibration, you're needs to put it on CH1.

For OPEN / SHORT / LOAD calibration you're needs to put proper terminator on CH0.

Amplifiers have higher noise with open input. This is why you're needs to place 50 Ω terminator on CH1 input for ISOLN calibration.

For OPEN / SHORT / LOAD calibration you're needs to put proper load on CH0.

For ISOLN calibration you're needs to put 50 Ω load on CH1.

But if you want better accuracy, it worth to use two 50 Ω loads on both CH1 and CH0 for ISOLN calibration.

Maybe stupid question, did you check if without filter you get an attenuation of 0dB?

Good morning guys,

I'm new to the group, and I got my nano last week.
The measurements on antennas are plausible and I was able to follow them with my "miniVNA", the situation is different with filter measurements.
Here for example on a LPF for 70cm. The filter was measured on a Wiltron Network Analyser and had an attenuation of 0.15 dB in the passband. In the blocking range at 600 MHz, the attenuation is -62 dB. With the "nano" I measured an attenuation of almost -11 dB in the passband.
A measurement on a 630m LPF yielded the same result.

Does anyone have similar measurement results for filter measurements, is my "nano" defective, or is the source of the error in front of the "nano"? :o)

73 Joe
DF2JP

Forgot to add screenshot, here it is

Here is TDR VSWR for your BNC connector. It seems that this BNC has about 422 picoseconds delay in comparison with supplied SMA reference.

Try to measure this BNC with calibration on supplied calibration kit and ELECTRONIC DELAY setting 422 picosecond. You can find ELECTRONIC DELAY setting in the SCALE menu.

Hello Dave,

Thanks for the reply. I've come to the same conclusion. I've ordered some
tiny push switches to replace the jog/select switch (single pole three
throw momentary).
I can mount these side-by-side along the top edge of the board with the
buttons facing upwards. I bought ten for ?1.39 delivered and I'll have them
by the weekend.

Best regards,

John

On Wed, 18 Sep 2019 at 07:33, Dr. David Kirkby from Kirkby Microwave Ltd <
drkirkby@...> wrote:

On Tue, 17 Sep 2019 at 22:07, <jimcking@...> wrote:

I broke my switch and hugen sent me a link to purchase a new on. I
purchased 10 because the shipping cost more than the 10 switches. A

search

for hugen posts should find the link.

Jim K.

What type of switch is it? Singe pole double throw (SPDT), DPDT etc?

To me at least, the unit is so poorly made that the only sensible choice is
to mount it inside a box with *only the screen accessible.* Everything else
needs to be on the outside of the box. Personally I would not bother
replacing the switch with another of the same quality. Instead I would run
wires out to a well made panel mounted switch.

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 Tue, 17 Sep 2019 at 22:07, <jimcking@...> wrote:

I broke my switch and hugen sent me a link to purchase a new on. I
purchased 10 because the shipping cost more than the 10 switches. A search
for hugen posts should find the link.

Jim K.

What type of switch is it? Singe pole double throw (SPDT), DPDT etc?

To me at least, the unit is so poorly made that the only sensible choice is
to mount it inside a box with *only the screen accessible.* Everything else
needs to be on the outside of the box. Personally I would not bother
replacing the switch with another of the same quality. Instead I would run
wires out to a well made panel mounted switch.

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 Tue, Sep 17, 2019 at 03:44 PM, Dana Shtun wrote:

Erik where is the 1500 Mhz code?

Greetings,

In order to get a feel for the quality of my SMA-BNC adapter I decided to calibrate with the supplied SMA references then measure my DIY BNC 50 Ohm load through the adapter. Attached are the results which show a return loss of 53dB, VSWR 1.005:1 for both 148 and 450 MHz. The RL at 890 is 29 dB/1.07:1. The results are attached. My cheap adapter is better than I expected.

--
Enjoy!
Tom
VA7TA

Hello everyone, I'm a member of NanoVNA-F develop team, BH5HNU(flyoob)'s aliexpress store is on business now. The NanoVNA-F and RF DEMO KIT is on shelf.


Welcome to our shop, my friends!
If you interested in our product, you can join /g/nanovna-f ~

Hi Eric;

Nice job with the extended range, plus the other added features. It mostly works OK, and as stated earlier; it interfaces nicely with NanoVNA Saver. Kudos to Rune also on his "beyond the call" value added effort.

There appears to be a bug in the calibration... at least in my case... While attempting to perform a 2-port 50 kHz to 30 MHz local cal (not via one of the Windows GUIs), The corrected S21 data plots consistently in excess of -6 dB after several cal attempts, and I'm unable to toggle the correction on and off. I've checked my cables, and this wasn't an issue with my original firmware install, so I'm pretty certain it's something in the firmware. I'm going to reflash the firmware tomorrow, but I don't suspect a bad install would do something wrong consistently and not eventually crash. :-)

Has anybody else been putting this extended range firmware through its paces and observed something similar?


--
72/73

Gary, N3GO

This thread discusses the thumbwheel switch.
/g/nanovna-users/topic/32796736
In post 869 of that thread, Hugen tells us where he buys the switch that the nanoVNA ships with
In post 1305 of that thread, Hugen recommends a better switch.

This thread discusses other options:
/g/nanovna-users/topic/34151142

And another:
/g/nanovna-users/topic/34166411

The thumbswitch is mechanically unreliable, mine often gets stuck.
One solution for this would be to drill out the front panel to take 3 button switches,
or fab a replacement using one of the many low cost board fab shops.
Drill a hole in the main nanoVNA board, run 3 wires plus ground from the thumbswitch
up to the three new button switches. No need to remove the old switch, the three pins
we are interested in are exposed.

I have a PCB with three button switches from some other gear,
may just glue the strip of three down to the top of the nanoVNA top panel.

But beyond the unreliable thumbwheel switch, we also have a software issue.
Seems the switches are polled, and if the ARM processor
is busy doing other stuff it may not see your switch action.
I'd think it should be interrupt driven, and simply bail out of processing the current crop
of data when the switch is pressed.

Jerry Gaffke, KE7ER