开云体育

When measuring a 433MHz filter the rejection can not be measured below 60dB due to the limited S21 isolation of the NanoVNA.
Because of the measurement approach of the NanoVNA there is a trick you can use to gain 20dB. This trick ONLY works above 300MHz and ONLY when measuring attenuators or filters that remove everything below 300MHz
First connect a 20dB attenuator and 20dB amplifier between ports 1 and 2, so port1 -> attenuator -> amplifier -> port2 and measure the S21 as can be seen in second graph.
The combination gives a magnitude of -1dB. Seems the amplifier performs a bit below the promised 20dB
Then connect the filter instead of the attenuator, so port 1-> filter -> amplifier-> port 2 and measure the S21 as can be seen in the third graph
Be aware the top of the scale is now at +20dB. Everything that was above -60dB in the first measurement stay's the same but the uppper band rejection drops down nicely to -80dB where the lower band stay's at -50dB.
To double check you should measure the S21 with a disconnected amplifier input to see the noise floor as in the last graph.

The headroom of the port 2 SA612 clearly has no 20dB headroom (more like 5dB) so this ONLY works because the filter effectively removes the fundamentals and the harmonics are at a much lower magnitude.
The reference SA612 also sees the much lower harmonics so this drop gets compensated in the S21 calculation (S21/Reference).
When connecting the amplifier between port 1 and 2 without attenuator or filter you see there is indeed insufficient headroom as the port 2 SA612 saturates

Eric,
Thanks for your input. I haven't tested your procedure yet, but it is definitely on my to do list. Maybe at some point an "applications" section can be added to the list where user's can contribute application notes on using the nanoVNA (i.e. "Increasing the measurement range of the nanoVNA above 300 MHz with a pre-amp"). Both software development and user documentation have grown over the past month for the nanoVNA, and an applications note section would assist new users in exploring the capabilities of this really useful device.

The headroom should be more than 5dB on the SA612 when running in overtone mode.
If I remember correctly the the third overtone is about -15dBC on a Si5351 so I would expect at least this headroom + whatever was already there for normal mode, or do I miss something?

If I remember correctly the the third overtone is about -15dBC on a Si5351
so I would expect at least this headroom + whatever was already there for normal mode,
or do I miss something?

Hardware has to deal with full stimulus energy; DSP sorts overtones

You are correct, the headroom is about 20dB when running in overtone, that is why I chose the 20dB amplifier, but only when you remove (or reducing with at least 15 dB) the fundamental, and that is exactly what you're doing when using this measurement approach.
When you are not filtering out the fundamental and use a 20dB amplifier the SA612 will be driven into non-linear and the overtone response will deviate from what is should be.