Author Topic: Why do I need a VNA rather than a SA with directional coupler to tune antenna?  (Read 9258 times)

0 Members and 1 Guest are viewing this topic.

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
I am wanting to do a bit of work tuning antenna. Be that PCB, helical or other. I have an SA with TG and a directional coupler on the way but someone told me I really need a VNA for this kind of work. Anyone explain why to me or point me in the direction of some useful article(s) that explain why?

Thanks

Trev
 

Offline xaxaxa

  • Regular Contributor
  • *
  • Posts: 248
  • Country: ca
There are two main reasons, the most common being that you only get magnitude and not phase information with a SA + directional coupler; with phase information you can make better guesses as to what changes to make to your antenna to tune it.
The less known reason is that directional couplers have limited directivity, and a VNA allows you to calibrate away the imperfections in your coupler (and you can easily measure down to -50dB S11) whereas a SNA could not.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
I am wanting to do a bit of work tuning antenna. Be that PCB, helical or other. I have an SA with TG and a directional coupler on the way but someone told me I really need a VNA for this kind of work. Anyone explain why to me or point me in the direction of some useful article(s) that explain why?

Thanks

Trev
Hope these help:
http://www.eham.net/ehamforum/smf/index.php?topic=91151.0
https://mediacdn.eu/m/media/downloads/Siglent-RBSSA3X20-Reflection-Bridge-Manual.pdf
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
There are two main reasons, the most common being that you only get magnitude and not phase information with a SA + directional coupler; with phase information you can make better guesses as to what changes to make to your antenna to tune it.
The less known reason is that directional couplers have limited directivity, and a VNA allows you to calibrate away the imperfections in your coupler (and you can easily measure down to -50dB S11) whereas a SNA could not.
Hi,  xaxaxa. Will your vna design enable antenna tweaking/tuning? How long do you feel it will be before they are available?

Sent from my SM-G950F using Tapatalk

 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
I am wanting to do a bit of work tuning antenna. Be that PCB, helical or other. I have an SA with TG and a directional coupler on the way but someone told me I really need a VNA for this kind of work. Anyone explain why to me or point me in the direction of some useful article(s) that explain why?

Thanks

Trev
Hope these help:
http://www.eham.net/ehamforum/smf/index.php?topic=91151.0
https://mediacdn.eu/m/media/downloads/Siglent-RBSSA3X20-Reflection-Bridge-Manual.pdf
Thanks 

Sent from my SM-G950F using Tapatalk

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
I am wanting to do a bit of work tuning antenna. Be that PCB, helical or other. I have an SA with TG and a directional coupler on the way but someone told me I really need a VNA for this kind of work. Anyone explain why to me or point me in the direction of some useful article(s) that explain why?

Thanks

Trev
Hope these help:
http://www.eham.net/ehamforum/smf/index.php?topic=91151.0
https://mediacdn.eu/m/media/downloads/Siglent-RBSSA3X20-Reflection-Bridge-Manual.pdf
Thanks 
Check the usage docs on the SSA3kx Siglent FAQ page on the websites.
I haven't got it open on this laptop but there should be some more guidance there.
There's nothing you want there it seems.  :(
http://www.siglentamerica.com/prodcut-pxzl.aspx?id=5113&tid=227&T=2
« Last Edit: September 30, 2017, 10:33:42 pm by tautech »
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline xaxaxa

  • Regular Contributor
  • *
  • Posts: 248
  • Country: ca
Hi,  xaxaxa. Will your vna design enable antenna tweaking/tuning? How long do you feel it will be before they are available?

Yes it is definitely sufficient for antenna design; I have used it to tune up several antennas already. If the kickstarter is successful I can start shipping by march of next year; I have already gotten quotes for pcba and components, and will put up the kickstarter soon.

If you need it sooner you can always build my design or hforsten's design; both are fully open sourced and have all design files online.


 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
Do you have any spare boards that you want to sell? Better to give you some money than build from scratch myself. I would like to reward your hard work if possible.

Thanks for making it open source, that really is awesome of you

Trev
 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
tautech, thanks for your suggestions. I will check the FAQ out. You have a lot of energy for answering in the forum. I know you are a distributor in NZ but still, you seem to work hard and it is very much appreciated. The SSA3021X appears to be a beast. I dont really know what I am doing at the moment but I really like the unit so far.

Trev
 

Offline xaxaxa

  • Regular Contributor
  • *
  • Posts: 248
  • Country: ca
Do you have any spare boards that you want to sell? Better to give you some money than build from scratch myself. I would like to reward your hard work if possible.

I have just the one prototype shown in the pictures, and I generally don't sell hand-assembled boards because of reliability concerns (solder joints are not perfect); 
If your directional coupler has >20dB directivity, it is good enough for antenna tuning and should do for now.

It mainly depends whether you're designing antennas or just fine tuning antennas. If you are building an existing antenna design that has already been characterized, a SA+coupler is more than sufficient since you only need to fine tune the resonance to the right frequency. A vna is only needed for designing new antennas when simulations are either unavailable or not accurate enough.

 
The following users thanked this post: trevwhite

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
tautech, thanks for your suggestions. I will check the FAQ out. You have a lot of energy for answering in the forum. I know you are a distributor in NZ but still, you seem to work hard and it is very much appreciated. The SSA3021X appears to be a beast. I dont really know what I am doing at the moment but I really like the unit so far.

Trev
Thanks Trev my energies are juggled between Dist, hobbies, other business commitments, family and EEVblog.
I too don't have much experience with analysers but I know there's a lot you can find online and in the forum here.

I'll do some checks with my SSA to see if Phase measurement is possible....member hendorog gave me a tiny SMA directional coupler and along with a WiFi SMA antenna it shouldn't be hard to cobble something together to get some SWL info and maybe phase info.

Why ? Because I'm learning too and the some/little knowledge gained helps me support customers better and you guys too.
As Arnie says; I'll be back, hopefully with something that's useful.
Give me a bit, it's lunchtime here and I have to hunt all the bits out.

Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 
The following users thanked this post: trevwhite

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
Thanks for the replies.

I have an rf module that comes in a few different frequencies. 433, 868 and 915Mhz. I have the module tracked to a board with a smallish ground plane of around 6cm x 6cm. I am not sure if a 17cm wire will be okay with this ground plane for the 433Mhz version. My thoughts were to solder a wire to a bare board. Then connect the output of the coupler to the ground plane and the antenna and trim the wire accordingly and see what I get. I was then thinking of doing something similar for the other frequencies.

After that I wanted to experiment with some helical designs to see if they work as well as they improve on space.

Am not sure how I will be able to compare designs for gain values at the moment. It is all a bit new and experimental at this stage.

 

Offline PA4TIM

  • Super Contributor
  • ***
  • Posts: 1161
  • Country: nl
  • instruments are like rabbits, they multiply fast
    • PA4TIMs shelter for orphan measurement stuff
http://www.pa4tim.nl/?p=1594 I made a series vna tutorials.
The difference is that a SA+TG only gives you the return loss but no phase info. So you do not know if there is a phase jump. Besides that you can not calibrate a SA the same way as a VNA. And that is the most important part of a measurement. The result can never be better as the calibration. Without that you only have an indication and the result can be so way off that it is kinda useless. The higher the frequency the more important calibration is for the end result.

For measurements on a pcb or other cases you can not calibrate and measure direct at the vna port you need to do more as a normal calibration and modern VNAs have functions for that, they can shift the reference level. A SNA is useless for those kind of measurements.
www.pa4tim.nl my collection measurement gear and experiments Also lots of info about network analyse
www.schneiderelectronicsrepair.nl  repair of test and calibration equipment
https://www.youtube.com/user/pa4tim my youtube channel
 
The following users thanked this post: Mechatrommer

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
http://www.pa4tim.nl/?p=1594 I made a series vna tutorials.
The difference is that a SA+TG only gives you the return loss but no phase info. So you do not know if there is a phase jump. Besides that you can not calibrate a SA the same way as a VNA. And that is the most important part of a measurement. The result can never be better as the calibration. Without that you only have an indication and the result can be so way off that it is kinda useless. The higher the frequency the more important calibration is for the end result.
Educate us please.  :-//
Is the calibration you mention not the same as Normalise ?
After Normalise I can set the displayed Reference level and numerous other parameters to obtain detailed info on a spectrum of interest.

I'll do some checks with my SSA to see if Phase measurement is possible....member hendorog (thanks Roger)gave me a tiny SMA directional coupler and along with a WiFi SMA antenna it shouldn't be hard to cobble something together to get some SWL info and maybe phase info.
Grrrr.....I need an adaptor that I looked at online recently and wondered what on earth would I use that for. Male SMA male to Male SMA female....no not the outer thread but the inner pin.  >:(
Two Male SMA females for the WiFi antenna won't get a connection, will they ?  ::)  :palm:

More RF bits needed. <sigh>  :scared:
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
http://www.pa4tim.nl/?p=1594 I made a series vna tutorials.
The difference is that a SA+TG only gives you the return loss but no phase info. So you do not know if there is a phase jump. Besides that you can not calibrate a SA the same way as a VNA. And that is the most important part of a measurement. The result can never be better as the calibration. Without that you only have an indication and the result can be so way off that it is kinda useless. The higher the frequency the more important calibration is for the end result.
Educate us please.  :-//
Is the calibration you mention not the same as Normalise ?
After Normalise I can set the displayed Reference level and numerous other parameters to obtain detailed info on a spectrum of interest.

I'll do some checks with my SSA to see if Phase measurement is possible....member hendorog (thanks Roger)gave me a tiny SMA directional coupler and along with a WiFi SMA antenna it shouldn't be hard to cobble something together to get some SWL info and maybe phase info.
Grrrr.....I need an adaptor that I looked at online recently and wondered what on earth would I use that for. Male SMA male to Male SMA female....no not the outer thread but the inner pin.  >:(
Two Male SMA females for the WiFi antenna won't get a connection, will they ?  ::)  :palm:

More RF bits needed. <sigh>  :scared:
I think the issue with the SA + coupler is that we will get a frequency response indicating the resonant frequency but what we do not get is the phase angle of the impedance? The VNA shows this phase angle and thus indicates if a circuit needs more L or C  to bring the antenna in phase with the transmitter.

Sent from my SM-G950F using Tapatalk

 

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
Looks like I need to build a VNA!

Sent from my SM-G950F using Tapatalk

 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28139
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
I think the issue with the SA + coupler is that we will get a frequency response indicating the resonant frequency but what we do not get is the phase angle of the impedance? The VNA shows this phase angle and thus indicates if a circuit needs more L or C  to bring the antenna in phase with the transmitter.
Yep I get that, it was PA4TIM's comments on calibration.
What expressly does that mean ?
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline hendorog

  • Super Contributor
  • ***
  • Posts: 1617
  • Country: nz
I think the issue with the SA + coupler is that we will get a frequency response indicating the resonant frequency but what we do not get is the phase angle of the impedance? The VNA shows this phase angle and thus indicates if a circuit needs more L or C  to bring the antenna in phase with the transmitter.
Yep I get that, it was PA4TIM's comments on calibration.
What expressly does that mean ?

I don't quite understand what PA4TIM meant either. A SNA can be normalised (calibrated) so that the measurements have meaning.
Maybe he meant that the VNA can compensate for the coupler directivity:
With an SNA measurements of return loss of an antenna do become less and less accurate as the return loss approaches the directivity of the coupler - i.e. you need a high quality coupler to measure a high quality antenna. Good summary with pictures here (Figure 5):
http://www.wetterlin.org/sam/Reflection/3BeadBalunBridge.pdf

A VNA can compensate for this effect. It can also compensate for other limitations in the equipment like the return loss of the test ports. The cost of this is more calibration time and the need for precisely characterised calibration standards.
I read somewhere about a retiring HP directional coupler guru saying that he wasn't needed anymore as powerful CPU's made calibration cheap so now any old coupler would do the job.

Edit: I realised that PA4TIM meant reference plane extension - where a VNA can user clever maths to remove the cabling and adapters, and make measurements at the PCB itself as if the cables and adapters were never there. An SNA can compensate for the loss in the cables and adapters, but not for other effects like return loss of the adapters.
« Last Edit: October 01, 2017, 02:02:16 am by hendorog »
 
The following users thanked this post: tautech

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
Hi,  xaxaxa. Will your vna design enable antenna tweaking/tuning? How long do you feel it will be before they are available?

Yes it is definitely sufficient for antenna design; I have used it to tune up several antennas already. If the kickstarter is successful I can start shipping by march of next year; I have already gotten quotes for pcba and components, and will put up the kickstarter soon.

If you need it sooner you can always build my design or hforsten's design; both are fully open sourced and have all design files online.

Hi xaxaxa, I could not find the gerber files for your design? I found the cad files I think. Do I need to install the cad software and generate the build files to make my own? Also I know you said you only had the one prototype but what about bare PCBs unpopulated? Any of those kicking around for sale?

Trev

Trev
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
I've just done a quick spreadsheet comparing the measurement error using a 30dB directivity coupler (which is as good as they get) and a calibrated VNA that assumed 47dB effective directivty, which is about what I would expect with a freshly calibrated unit. The errors are +/- and all figures are in dB

RL   error, coupler   error VNA
10   0.872   0.123
12   1.099   0.154
14   1.388   0.194
16   1.757   0.245
18   2.230   0.308
20   2.844   0.388
22   3.660   0.489
24   4.785   0.616
26   6.454   0.776

So for a 12dB return loss antenna then there is 1.1dB of measurement uncertainty using a good coupler, compared with 0.15dB with a VNA.
For a 20dB return loss antenna then it is 2.8dB compared with 0.4dB.

Offline trevwhiteTopic starter

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: gb
Wow, that's great. Thanks, Kev. Are those measurements what I could expect with one of the VNA you have for sale?

Trev
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
Providing it's been calibrated, for that you'd need a cal kit, otherwise it's only as good as the coupler approach.

There are usually reasonable cal kits on ebay, Those from Dr Kirby are probably the most reliable route to a good one at a reasonable price.

Online KE5FX

  • Super Contributor
  • ***
  • Posts: 1878
  • Country: us
    • KE5FX.COM
For a 20dB return loss antenna then it is 2.8dB compared with 0.4dB.

Keep in mind, though, that this is the difference between an SWR of 1.16:1 for 22.8 dB RL and 1.22:1 for 20 dB.  So if you are the least bit budget-constrained, it's better to cheap out on the test gear and put the savings into the antenna system itself.
 

Offline KJDS

  • Super Contributor
  • ***
  • Posts: 2442
  • Country: gb
    • my website holding page
For a 20dB return loss antenna then it is 2.8dB compared with 0.4dB.

Keep in mind, though, that this is the difference between an SWR of 1.16:1 for 22.8 dB RL and 1.22:1 for 20 dB.  So if you are the least bit budget-constrained, it's better to cheap out on the test gear and put the savings into the antenna system itself.

It's plus or minus 2.8, so between 17.2 and 22.8dB for a measured 20dB.

Online KE5FX

  • Super Contributor
  • ***
  • Posts: 1878
  • Country: us
    • KE5FX.COM

True enough, but the difference still amounts to only 0.06 dB of mismatch loss. 

That's about what you might save by putting another $25 into your coax fittings, especially after they've been exposed to weather for a while. :)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf