Timing for new batch available for order

[deandob]

Hi,

I sent an email to the email at the 'contact us' page on the xaxax web site but didn't hear back. Another batch for sale was apparently due in April, have these arrived or sold out? Do let me know how I can order an xaVNA T/R two port.

Thanks - and nice work with the development of this.

Regards,
Dean

[xaxaxa]

Hi,

Sorry, it looks like your email landed in my spam folder (I'll be sure to check it more often).
The boards have arrived but there was a delay with the enclosure. I think this should be sorted in a week and it will be available for order.

Thanks

[deandob]

Great, do post back here when its available for payment. Thanks.

[deandob]

I see these are now available from your website, I have placed an order to ship to Australia. Thanks.

[satfan52]

Hello everybody,

I want to publicly thank Cary for giving me the opportunity to assemble "the 2 port version" of his xaVna product. I can not publicly thank Cary on Kickstarter as I did not buy a device there so I decided to thank him and provide my feedback here instead.

Initially, I wanted to buy a two port version but since it was not yet available commercially Cary offered me to assemble a two port xaVna from one of his spare PCB (that was fortunately already partially populated with many smd 0402 components). After installing the additional components needed and after flashing the firmware, I ended up with a fully functioning device. My device was not as profesionaly assembled as the ones soldered in industrial environments (I am not a robot or a CNC machine!) but I am very happy with the final result. As a matter of fact, my home assembled two port xaVna device meets the formal specification of the two port xaVna and even exceeds them in terms of "frequency range" and "dynamic range". The performance of the industrially assembled two port xaVna, will of course be superior to the one of this manually handed assembled device as the industrially assembled boards will have less stray reactance (due to the use of better and truly profesional soldering techniques).

Cary helped me solved the many challenges that I encountered during the assembly. I therefore especially want to praise the quality of his technical support (as well as his patience and his availability). Please rest assured that if you buy some of Cary's products you will be offered top quality support (if you need it).

Here are a few pictures of the measurements that I have made with my home assembled two port xaVna (I have just finished assembling it today).

Picture 1 below: Low pass 60 Mhz filter - sweep between 35 and 250 Mhz - 200 points

As you can see in the picture below, the S21/S12 dynamic range clearly exceeds 70 dB in this frequency range, it is in the order of 80-90 dB so much more than the guaranteed 70dB !

I have only used 200 points, so roughly one measure every 1 Mhz , that is why the definition of the Smith trace appears a bit too low at some frequencies near the cut off the filter. Increasing the number of points would increase the definition of the trace course, but that is not really needed for this specific example aimed at demonstrating the excellent dynamic range of the device at lower frequencies.

Picture 2 below : 440 Mhz bandpass saw filter - sweep beween 400 and 500 Mhz

Note the good S21/S12 dynamic range as well as the excellent definition of the trace. As a matter of fact it is possible to get 70 dB dynamic range up to approximately 1.5 Ghz, so well above the 800 Mhz guaranteed for the xaVna.

Picture 3 below: Wifi Antenna - sweep between 2 and 3 Ghz - 500 points

Shows the traces of a Wifi Antenna resonating in the 2.4 Ghz band. Here I increased the number of points to 500 (instead of 200 in the first example) so I get a definition that is adequate to display a nice resonance loop on the Smith Chart. Note that the measurements extends to 3 Ghz without any problems; (so beyond the guaranteed 2.5 Ghz limit).

Picture 4 below: 2 Ghz low pass filter - full scan sweep between 35 Mhz and 4.4 Ghz - 200 points

This is my favourite measurement as it shows that the two port xaVna is fully functional up to 4.4 Ghz so well above the 2.5 Ghz guaranteed limit. It also shows that the S21/S12 dynamic range in the 2.75-4.4 Ghz range is in the order of 50-60 dB which is quite good considering that the performance guaranteed is 50 dB up to 2.5 Ghz.

Picture 5, 6, 7, 8 below: 3, 40, 50 and 59 dB attenuators - sweep between 2.75 and 4.4 Ghz

The xaVna is guaranteed up to 2.5 Ghz but works nicely up to 4.4 Ghz. This picture confirms that the exploitable S21/S12 dynamic range is in the order of 50-60 dB (depending on the frequency) between 2.75 and 4.4 Ghz. My two port xaVna device can also be used in this frequency range with a dynamic range adequate for making meaningful measurements.

Picture 5 below shows the S21/S12 curve of a 3 db attenuator, the linearity is very good. The atenuator is golden and guaranteed to work up to 10 Ghz +/- 0.5 dB

Picture 6 below shows the S21/S12 curve of a 40 dB atenuator, it is still quite good though towards 4-4.4 Ghz some deviations start to appear. The attenuator is golden and guaranteed to work up to 10 Ghz +/- 1.5 dB

Picture 7 below shows the S21/S12 curve of a 50 dB attenuator, it is possible to make indicative rf measurements with a 50 dB dynamic range up to 4.4 Ghz.

Picture 8 below shows the S21/S12 curve of a 59 dB attenuator, here it is clear that performance is only acceptable up to roughly 3.5 Ghz for port 1. Port 2 shows less deviation that port 1 above 3.5 Ghz but the result is not good either. Indicative rf measurements with a 60dB dynamic range can be done up to maximum 3.5 Ghz (roughly).

Picture 9, 10, 11, 12 below: Popular 4Ghz Aliexpress amplifier - full scan from 35 Mhz to 4.4 Ghz and close -up inspection between 3 and 4.4 Ghz

Picture 9 below shows the Gain (S21) and Isolation (S12) between 35 Mhz and 4.4 Ghz. An attenuator of -40dB is inserted at the Input of the amplifier (to avoid saturating it and port 2 of the VNA given the high gain below 1 Ghz). For this reason it is necessary to add 40dB to all measurements showed on the picture. Unfortunately the amplifier does not hold its promise in the 3.8 Ghz region as the S21 gain is negative..

Picture 10 below shows a close-up between 3 Ghz and 4.4 Ghz of the Gain (S21) , the input SWR (SWR(S11)) and the SMITH chart at the Input of the amplifier (VNA port 1). The output power of the VNA is reduced to -15dBm only to avoid using an attenuator at the input of the amplifier (so the figures on the picture do not need to be adjusted). There is no risk of saturating port 2 of VNA given the limited gain of the amplifier in this frequency range. The measure confirms that the amplifier does not hold its promises in the 3.8Ghz frequency range.

Picture 11 below shows a close-up between 3 Ghz and 4.4 Ghz of the Isolation (S12) , the output SWR (SWR(S22)) and the SMITH chart at the output of the amplifier (VNA port 2). The measure shows that the isolation considerably degrades in this frequency range. No attenuator is used so the figures on the picture are the real ones

Picture 12 below shows a close-up between 3 Ghz and 4.4 Ghz of the Gain (S21) and Isolation (S12), no atenuator is used so the figures on the picture are the real ones. The output power of the VNA is reduced to -15dBm to avoid having to use an attenuator

[deandob]

My VNA from the new batch arrived a week ago, well packed and well assembled, thank you.
I'm using it to tune a 915Mhz antenna but struggling a bit with setting up and measuring, more my lack of understanding how measure with a VNA although a user guide would have helped (have worked out how to calibrate so far). I'm also unclear how to use the 2nd port (I have the T/R version) for measurement. Stuff found on google is helping but if someone has a definitive quide to antenna tuning with a VNA please pass on a link.

[satfan52]

I am not familiar with the T/R version as I own the two port one. However assuming the T/R and two port work the same way, if you are using the VNA in reflexion mode to tune an antenna then you only need port 1 for making the measurement. With the two port version, I can use port 1 or port 2, it does not matter. Or I can connect two different antennas to each port and very easily compare their resonance.

If you decide to use an external rf bridge/coupler with a simple SOL calibration , then you will need to use the two ports of the T/R version obviously, but I don't see why you would want to do that with the xaVna

If you want to tune the antenna, the goal is to get it to an swr of 1 I assume. You can display the swr curve and you can look on the smith chart what is the impedance of the antenna. If you see it is too inductive you can short it, if it is too capacitive you can extend it (for example). If the impedance is ohmic but too low or too high compared to 50 ohm, then you need to match your antenna one way or another to 50 ohm using gamma match, ballun, choose a different feedpoint etc...This Smith chart will only help you see what is the effect of your matching attempts and it can help you calculate the reactive network that is required to do the matching. No software or vna manual can teach you about using the Smith Chart for antenna tuning, people have spent the best years of their lives on the subject of antenna matching and tuning using the Smith Chart as a tool, you will find plenty of info on google, YouTube has some excellent videos on the subject

I would use the settings in the picture below

[deandob]

Thank you satfan52, helpful.
Would be useful if someone could explain how to use the second T/R (transmit/receive?) port and why its different to a proper 2 port (eg. can we measure S21,S22 etc.)?

[satfan52]

You can use the second port of the TR/VNA to measure the Gain, loss and phase shift of a 2 port DUT like an amplifier or a filter for instance. If you build a TLT transformer for an LDMOS Power amp or a balun for your antenna you can measure it too (for instance to verify that the impedance transforming ratio is the right one as well as to check its bandwidth) Antennas are one port DUTs so you do not need the second port on the T\R model, only port 1. If you connect the 2 port DUT between port 1 and port 2, you will be able to see the input impedance on the Smith Chart on the left and on the right side you can for instance display the frequency response of the Gain of the amplifier or the loss of the filter (Mag S21 and Arg S21) . The key point about T/R is Is that you can measure transmission related info (S21) as well as reflexion related info (S11) at the same time, that is why it is called a T/R! I have another VNA that is not T/R and it is a pain in the neck as I need to different calibration procedures for reflection and transmission measurements. Seeing both gain and impedance at the same time is useful as you can see for instance how an swr improvement at the input of amp translates into a higher gain since less energy is reflected towards the generator With a T/R VNA, if you want to measure the output impedance of the 2 port DUT and the Gain or loss (ie, the isolation) in the opposite direction (so from output to input), then you need to physically reverse the DUT so its input is connected to port 2 instead of port 1 and its output to port 1 instead of port 2. With the two port VNA I don't need to do to this reversing because the two ports are active in the very same way, with one measurement I get everything related to the input and output of the 2 port DUT. That is especially useful for making measurements of non symmetrical devices like an amp, a mixer or an rf transformer for instance. I can for instance visualise how changes of the output impedance of an amp affects the input impedance due to the Miller effect. There are of course other advantages of using a 2 port VNA over a T/R one. Hope that helps Regards

…

On Tue, 18 Jun 2019, 14:32 deandob, ***@***.***> wrote: Thank you satfan52, helpful. Would be useful if someone could explain how to use the second T/R (transmit/receive?) port and why its different to a proper 2 port (eg. can we measure S21,S22 etc.)? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#25?email_source=notifications&email_token=ADR72WOKUULD7FOV3IOJJH3P3DIXJA5CNFSM4HK7TKUKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODX6HD2A#issuecomment-503083496>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ADR72WNO22LIUBKPR67IGY3P3DIXJANCNFSM4HK7TKUA> .

[deandob]

Useful thanks satfan52.
The VNA has already earnt its keep, I'm developing a LoRa solution and can see that the helix antenna we are using has quite narrow bandwidth that is just outside our frequency when its inside its box, and with some adjustments to the antenna layout and the matching circuit the gain has increased significantly!