VNA in high noise floor

[Randy222]

Just curious what others may be doing.

Lets say you get a handful of 2.4Ghz wifi omni antenna to test on your VNA, but your immediate environment already has a high noise floor in that band, how do you account for that noise being in the measurements? As example, my own local AP hits a nearby Rx at around -40dbm, that's pretty loud. I know it's best to turn off local wifi, but still getting quite a bit of noise from countless other wifi AP's in my urban area. Certainly this issue is not limited to just wifi bands, etc.

Omni is one type, what about gain antenna like folded corner or yagi, etc.

My anechoic chamber is not finished yet, probably wont ever be.

[Bud]

Reduce the VNA's detector bandwidth. Some may have specific settings, some would automatically adjust when sweep time is increased. Mickey-Mouse VNAs may not have that though.

[pdenisowski]

My anechoic chamber is not finished yet, probably wont ever be.

Doesn't have to be a whole chamber - a small shield box might be enough.  A "proper" commercial one (like the R&S Z10) is probably well out of your budget, but you could build a semi-proper one for a couple hundred USD - either a metal box or even a cardboard box with RF shielding cloth.

The picture below shows a roll of that material we used to isolate a non-removable / non-shut-off-able source of interference in a building by building a "curtain" around it.  Knocked the signal down by 30-40 dB, and we weren't even that careful in wrapping it

[radiolistener]

For a VNA, this is generally not a big issue. The main concern is that the external RF signal should not be strong enough to overload the VNA receiver.

A typical VNA can tolerate a fairly strong interfering signal on the input, and something around -10 to 0 dBm of interference is usually not catastrophic (though the exact limit depends on the particular VNA). So if your local AP is around -40 dBm at the VNA input, I wouldn't expect it to have much effect on the measurement.

For example, I've measured a filter while a 5 W VHF transmitter was operating nearby at 145 MHz on 20-30 cm distance (near field region). It had practically no effect on the measurement results, at most the trace would fluctuate slightly.

[pdenisowski]

For a VNA, this is generally not a big issue. The main concern is that the external RF signal should not be strong enough to overload the VNA receiver.

A typical VNA can tolerate a fairly strong interfering signal on the input, and something around -10 to 0 dBm of interference is usually not catastrophic (though the exact limit depends on the particular VNA). So if your local AP is around -40 dBm at the VNA input, I wouldn't expect it to have much effect on the measurement.

For example, I've measured a filter while a 5 W VHF transmitter was operating nearby at 145 MHz on 20-30 cm distance (near field region). It had practically no effect on the measurement results, at most the trace would fluctuate slightly.

The issue is mostly whether the external signal can get into the VNA's (measurement) receiver.  If you're using high quality coax with good shielding, a connectorized DUT, and solid connections with a torque wrench, then there's a much smaller chance of a signal entering the VNA measurement receiver.  Of course, it helps if the DUT has it's own shielding - the traces, etc. in a DUT can easily pick up strong RF and route back to the device input or output.

A good example of how cable quality is important is the semi-rigid loops on high end VNAs that can be removed to allow direct receiver access.  These loops are "exposed" to the ambient RF environment all day, every day, and often in RF "dense" places like labs without causing measurement issues. 

And I might be misunderstanding you, but a -40 dBm spur in the measured signal would absolutely wreck quite a few of the VNA measurements I've made.  There would also be the risk of that "second signal" creating intermod products within the measurement receiver.

Just my 0.02 

[Randy222]

So if your local AP is around -40 dBm at the VNA input, I wouldn't expect it to have much effect on the measurement.

-40dbm at about 30ft away from an AP on omni Rx antenna.
My VNA setup is located where that -40dbm is measured.
I then connect a wifi Yagi that has ~12dbi gain to my VNA.

Reflections from Tx, direction of Yagi (on axis with AP, off axis, etc). The VNA is gonna pick up some of that Tx signal, which will mess up VNA math?

[radiolistener]

And I might be misunderstanding you, but a -40 dBm spur in the measured signal would absolutely wreck quite a few of the VNA measurements I've made.  There would also be the risk of that "second signal" creating intermod products within the measurement receiver.

I was referring to the external signal actually getting into the measurement path. I deliberately tried this without shielding, placing the antenna of a 5 W transmitter very close to the DUT and its measurement connections.

Of course, once the interfering signal becomes strong enough to overload the VNA frontend, distortion and intermodulation products appear very quickly. But below that point, the effect of even a fairly strong RF pickup can be surprisingly small because the VNA is measuring its own coherent stimulus signal.

This is one of the advantages of a VNA. Unlike a scalar measurement system, which can be very sensitive to external RF signals, a VNA can largely reject signals that are not coherent with its own stimulus.

[radiolistener]

-40dbm at about 30ft away from an AP on omni Rx antenna.
My VNA setup is located where that -40dbm is measured.
I then connect a wifi Yagi that has ~12dbi gain to my VNA.

Reflections from Tx, direction of Yagi (on axis with AP, off axis, etc). The VNA is gonna pick up some of that Tx signal, which will mess up VNA math?

Theoretically it will be around -28 dBm, but I wouldn't rely on that estimate, since it's not clear what kind of omni antenna it is and how much attenuation it has.

I would measure the actual RF power at the Yagi output before connecting it to the VNA. Connecting an antenna to the VNA without knowing the possible signal level could be risky. -40 dBm at the VNA input itself should not be a problem, but if the directional antenna produces a much higher level (for example above ~0 dBm - depends on your VNA), it could overload or even damage the VNA input.