PocketVNA Any idea what its like?

[Kalvin]

Everyone seems to be building 2 port VNA's.  My personal view is that this is just because there are now some really nice RF IC's available at low cost, and engineers are thinking "great IC, lets build a 2 port VNA).

Even Pico Technology have just built one.  They certainly have their uses, but most seem to be destined to become posh VSWR meters.

What would be really useful is a USB connected (powered via mains!) 4 port VNA that goes from 300KHz to 6GHz (to replace my aging 8753C).....

This doesn't go down to 300 kHz but nevertheless inexpensive and nice design with detailed explanation:

http://hforsten.com/cheap-homemade-30-mhz-6-ghz-vector-network-analyzer.html

With some improvements:

http://hforsten.com/improved-homemade-vna.html

The design files can be found in the Github.

[fcb]

That's a really impressive project - might be very interesting if he does a third version.

300KHz is overkill - but 30MHz still very usable as the bottom end.

[Wirehead]

It seems Martin quietly released a beta version of the new pocketVNA software. Much improved calibration screens and methods.. I'll give it a go tomorrow

[Wirehead]

Unfortunatelly, you can't check the port 1 impedance without another VNA, because port 2 doesn't have reflection measurement capability (it's not a bidirectional VNA, so only S11 and S21 measurements). But it's clear by looking at the PCB that the design arround port 1 (CN2 connector!) is far from ideal.

You can check the effect of port 1 matching by measuring a couple of meters long coaxial cable. First calibrate the VNA and measure the coax S21 as usual, save the trace. Then place a 10dB attenuator at port 1, calibrate the VNA again and remeasure the same coax. You will see that measurements doesn't match, and the first one will show a huge ripple. You can use the 10dB pad technique to improve this behaviour, but you will reduce the VNA dynamic range by 10dB.

Now imagine trying to adjust or optimize a high Q circuit, like a band pass cavity or duplexer: it will not be possible because it's behaviour will be very different when connected to the VNA. Resonances will change, loss, rejection... You could be able to get the desired response, but it will not work the same way when attached to a 50&50ohm system.

If your VNA behaves differently, please take a photo of the PCB to look for the differences.

I found a good site explaining that technique with the attenuator.. Might be interesting for others:
https://www.evaluationengineering.com/features/0200loss.htm
It also mentions "The technique is not appropriate for S-parameter VNAs because a full 12-term, two-port calibration inherently will produce accurate results." So the issue could be overcome by not using a pad; but doing a 12-term 2port calibration? Maybe that's why it's in the new beta? Hopefully eb4fbz can chime in 

[xaxaxa]

I found a good site explaining that technique with the attenuator.. Might be interesting for others:
https://www.evaluationengineering.com/features/0200loss.htm
It also mentions "The technique is not appropriate for S-parameter VNAs because a full 12-term, two-port calibration inherently will produce accurate results." So the issue could be overcome by not using a pad; but doing a 12-term 2port calibration? Maybe that's why it's in the new beta? Hopefully eb4fbz can chime in 

12-term calibration can't be done on a T/R vna because the second port does not have a signal source and can't measure reflection.

Both the miniVNA tiny and pocketVNA are of the T/R (transmission/reflection) type.

The full two port version of the xaVNA supports this kind of calibration.

[Wirehead]

Ok, so hardware limitation of not having a full blown VNA... the 10db pad is a good idea then.

Do drop the sales pitch please. No need for that when discussing technical issues.

[xaxaxa]

Ok, so hardware limitation of not having a full blown VNA... the 10db pad is a good idea then.

Do drop the sales pitch please. No need for that when discussing technical issues.

Not sure why you would think that was a sales pitch; I was trying to be helpful and clear up some confusion about T/R two port vs full two port, since some manufacturers are vague about it. 

[Jaimon]

I also faced the same problem with calibration , But with modified version of the software " BETA" it works fine...

[hendorog]

 

I found a good site explaining that technique with the attenuator.. Might be interesting for others:
https://www.evaluationengineering.com/features/0200loss.htm
It also mentions "The technique is not appropriate for S-parameter VNAs because a full 12-term, two-port calibration inherently will produce accurate results." So the issue could be overcome by not using a pad; but doing a 12-term 2port calibration? Maybe that's why it's in the new beta? Hopefully eb4fbz can chime in 

12-term calibration can't be done on a T/R vna because the second port does not have a signal source and can't measure reflection.

Both the miniVNA tiny and pocketVNA are of the T/R (transmission/reflection) type.

The full two port version of the xaVNA supports this kind of calibration.

There is a cal type called one path two port which can do this on t/r network analysers. You need to manually reverse the dut.

[0xdeadbeef]

I wonder if there is the verdict on the pocketVNA has changed in the meantime? Is it usable in the 2G range or does it have a design flaw?
I figure there is a SW 1.0 available since end of October 2018 but not much development since then.
Anyway, for the MiniVNA Tiny, things look even worse. I guess the developers have stopped supporting it after the Chinese copycats took over.

[OwO]

I wonder if there is the verdict on the pocketVNA has changed in the meantime? Is it usable in the 2G range or does it have a design flaw?

The design flaw is in using those all-in-one phase & log magnitude detector chips. No way you're gonna get good accuracy out of those because of DC, nonlinearity and drift errors. AFAIK all cheap USB VNA's have this problem. Edit: except those Si5351 based cheap 300MHz ones on aliexpress that actually do DSP on the micro.

[hendorog]

I wonder if there is the verdict on the pocketVNA has changed in the meantime? Is it usable in the 2G range or does it have a design flaw?

The design flaw is in using those all-in-one phase & log magnitude detector chips. No way you're gonna get good accuracy out of those because of DC, nonlinearity and drift errors. AFAIK all cheap USB VNA's have this problem. Edit: except those Si5351 based cheap 300MHz ones on aliexpress that actually do DSP on the micro.

Negative ghost rider, pocketVNA doesn't use that technique either. See the PCB layout and the block diagrams which have been posted above.

[ogden]

I wonder if there is the verdict on the pocketVNA has changed in the meantime? Is it usable in the 2G range or does it have a design flaw?

The design flaw is in using those all-in-one phase & log magnitude detector chips.

It is not about detector but front end design flaw.  PocketVNA have no log / magnitude detector but rffc5071 mixer with ad7192 ADC. It is clearly RF mixing and IF sampling architecture.

[AF6LJ]

Not to mention the first time you key a transmitter directly into one, or when a nearby transmitter fires up and blows the snot out of the detector you are not out big bucks or that HP you bought on E-Bay is waiting for you to buy another one to use as a Parts queen.

When you know their limitations you can get some use out of the cheap ones.

[KD0CAC John]

1st off Hi Sue
I have several antenna analyzer / VNAs , MFJ-269 , Sark 110 , and a MiniVNA Pro .
I keep all inputs covered with my calibration loads , either a 50 ohm load , short or open - probably in that order for effectiveness .
I started doing that after repairing several RF test gear and overloaded inputs MFJ 259s , TimeWave TZ-900 etc.
TimeWave is a friend of mine , club member and we talked about that , and ever since - capped all inputs .

[AF6LJ]

Long time no see.
Good to see you again.

Indeed this stuff is really fragile.