Thanks Dave,
that was exactly my point. We should take it as the wonderful and cheap
tool, that opens up world of VNAs that many of us would never have been
able to see otherwise. I don't think for most of us it would matter that
much if impedance would be 50.00 Ohms or 51.4 Ohms. In terms of return
loss that difference would not be a big deal and with real antennas it
would be well in the range of influence of environment to an antenna.
Even the factors like some stuff being located nearby the antenna,
position at the roof, the ground heigth, the fact if you are touching
the SMA port shield (and ground it more or less) and so on will have at
least that much influence on measuring the DUT's, so let's not get
carried away by hunting for superlatives but enjoy the usefulness.
vy 73
Wolfgang (DG7NEF)
Am 17.08.20 um 19:28 schrieb David Eckhardt:
I have N, SMA, and OSM series precision cal. kits from HP. I made a DC
measurement of the 50-ohm standard loads for these three connector series
from these kits using the following instruments (what I have).
1) Bench DMM: HP 3478A
2) Hand-Held DMM from Harbor Freight: P37772
3) Hand-Held DMM from Sperry: DM6400
*Results follow:*
*HP** Harbour Freight Sperry*
50.375 45.50 50.20 Type-N
(HP 909C)
50.179 45.50 50.10 SMA
(HP 909D)
50.330 47.00 50.10 OSM
Connector (the expensive ones from HP)
AVERAGE OF ALL READINGS: 48.801 ¦¸ Clearly the Harbour Freight unit
is dragging down the average.
AVERAGE OF ALL READINGS (excluding the Harbour Freight measurements):
50.214 ¦¸ (0.43 % error against 50.0000 ¦¸)
These are precision HP cal. standards. Sure, they are better than what
comes with the NANOVNA's. But consider the cost!!!!!! Is it worth arguing
and expecting something less than 2.26 % error when the load reads 51.13
ohms? Really, now?
This is a hobby, not a metrology lab!!!!!!
Dave - W?LEV
On Mon, Aug 17, 2020 at 4:20 PM Wolfgang Wilde via groups.io <wwilde69=
[email protected]> wrote:
Fully agree with Dave. Are you not awaiting a little bit to much from
the NanoVNA? What precision can be expected regarding impedance, levels,
noise figures and so on? The FR4 material may not be precise enough to
guarantee 100% precise 50 Ohms impedance for the input/ output lines of
the NanoVNA itself!
Can it really work that precise? Over what frequency range? And in what
impedance range will it show the right numbers?
Would you really expect similar figures and precision from it than you
would expect from HP/Agilent/Keysight/Tektronix/Rohde&Schwarz labor
grade measurement equipment that costs 100 times as much or more? Don't
get me wrong, it is really a brilliant DIY project that helps me a lot
and is very handy tool for me to get a rough idea of what is going on.
But I would never expect the NanoVNA being absolute precise as I think
it never was target on that and I doubt it could deliver that precise
results. It already does almost magic, so let's adjust expectations a
little bit for reality.
vy 73,
Wolfgang, (DG7NEF)
Am 17.08.20 um 17:33 schrieb Dr. David Kirkby, Kirkby Microwave Ltd:
On Fri, 14 Aug 2020 at 01:52, Glen Jenkins WB4KTF <wb4ktf@...>
wrote:
The 50 OHM load (SMA-Male) that came with my nanoVNA-H4 measures 51.13+
ohms. Not a good start for calibration. Where is a good source for
GOOD
loads that are accurate?
--
-----
Glen Jenkins, WB4KTF, Austin, TX
Do you have a specification in mind? DC resistance doesn¡¯t tell you much,
other than it puts an *upper* limit on return loss. As soon as you go up
in frequency, the return loss will most probably decrease.
If you measure up to 3 GHz, it gives you no idea what it will be like at
6
GHz. Sometimes loads worsen dramatically with frequency, but others, from
the same batch will get better.
The last I looked, there was nothing in the NanoVNA firmware that
properly
accounted for the open or short calibration standards. So there are
limits
to what you accuracy you are going to achieve if you had a really
expensive
load, as the phase errors will kill you.
Dave
--
