开云体育

Recently, I measured a 30 m loading coil for a short vertical antenna with my NanoVNA. I powered the VNA, calibrated for a span from 9 MHz zu 11 MHz with Nano VNA Saver 0.2.0 and started measuring (average 9/4) S11 (R+jX and Q). Repeating the scan, I found R slightly higher, but X beeing the same as before. So I measured again, with the same effect: R drifts higher, X stays the same.

Continually running scans, a pattern evolved. Checking the attached figure, it is clearly observable that R increased for about half an hour before the measured value stabilizes. X however is stable, right from the beginning.

I stopped the scans after 37 minutes, when I had the feeling that it leveled out. After re-calibrating I measured again. It came as no surprise, that the NanoVNA continued to measure the same number for X. However, R was down (even a touch lower) to the value which I measured at the very beginning. But now it stayed at that number.

My conclusion is, that there is a thermal effect during the warming up of the NanoVNA. Once the device has operating temperature this drift disappears. For whatever reason this seems to effect low |Z| (in my case around 1 Ohm for the real part of the impedance) to a much higher degree than higher values. I assume, this effect is lowest for |Z| around 50 Ohm.

If you are aiming for high precision, especially at low |Z|, my advise is to let the NanoVNA warm up for about half an hour, and only than calibrate and start to measure.

Has anybody encountered anything similar?

Vy 73,
Chris, OE1CGS

did you performed measurement with USB cable connected to NanoVNA? Try to disconnect USB cable to disable accumulator charging. It will eliminate such effect.

Yes, I measured with NanoVNA Saver 0.2.0, so having the USB connected is a must. The battery actually was already fully charged at the beginning, so shouldn't be too much of a problem unless the charging process is not optimal.
I don't see any possibility to suppress charging when using NanoVNA Saver anyway, besides removing the LiPo if this is the problem. Is there any?

Thx,
Chris, OE1CGS

Any expensive proffesional instrument needs 30 warmup minutes. It is really ussual. Calibration has to be done after warmup too.

In the miniVNA (VNA / J) software there is a thermometer.
The green light only occurs above 45 °. I missed this information on nanoVNA.

The difference is, that after warmup one gets repeatable results. When I calibrated the cold VNA and started to measure, maybe the very first results showed true numbers, but very soon results drifted (see the figure to my initial posting). I feel much better when repeated measurement show the same number. Otherwise there is always a bad taste...

Chris

Yeah, the original Nano (and mini/pocket VNA, plus most old HP gear) are like that. It's possible to remove the drift (and I mean not having to re-calibrate after warmup) with an internal correction switch (basically a switch that shorts port 1 to ground) which I've added in my design; hopefully others will follow.

On Wed, Dec 18, 2019 at 03:24 PM, Christoph Schw?rzler wrote:

I don't see any possibility to suppress charging when using NanoVNA Saver
anyway, besides removing the LiPo if this is the problem. Is there any?

Then connect usb and wait until battery will be charged to 100 % and charge led stops blinking. After that you can start measurement. The effect will be significantly reduced.

My guess warming battery.
But in my room temps vary from 6C to 15C and that requires recalibration.

Thanks QRP and jlxsolutions,

it seems as if I intuitively did the right thing. But I gives a much better feeling, reading about your observations and solutions.

Happy New Year,
Chris, OE1CGS